EPSRC Studentships
We offer fully funded Engineering and Physical Sciences Research Council (EPSRC) studentships in the following areas:
This call is open to UK Applicants only.
Applicants should be of outstanding quality and exceptionally motivated.
The studentships are funded for 3 years subject to satisfactory annual performance and progression review, and will provide for tuition fees and a tax-free stipend paid monthly.
Please note that there are more projects than funded studentships available and therefore this is a competitive application process which will include an interview. Shortlisted candidates will be contacted for an interview in person or via Teams. After interview the most outstanding applicants will be offered a studentship.
Application Details
Queries about the application process are welcomed. These along with completed forms, including all relevant documents should be submitted via-email to pgrscholarships@hud.ac.uk by the closing date which is 12 noon on Friday 12th April 2024.
Informal enquiries about individual projects should be directed to the lead supervisor listed for each project.
Summary of Award
Type of Award: |
Doctor of Philosophy (PhD) |
Eligibility: |
UK applicants only First Class or Upper Second Class Honours degree or equivalent in a relevant subject area, please refer to the entry requirements on the specific projects being advertised |
Location: |
Huddersfield |
Funding: |
3 years full time research covering tuition fees and a tax free bursary (stipend) starting at £19,237 for 2024/25 and increasing in line with the EPSRC guidelines for the subsequent years. Funded via the Engineering and Physical Sciences Research Council Doctoral Training Programme |
Duration: |
3 years Full-time plus 12 months writing up (please note that no funding is available for the writing up period) |
Revolutionising Glioblastoma research by developing a biomimetic organoid model, integrating AI, gene-editing, and tissue engineering
Recycle and Synthesis polyester and cellulose fibres from textile waste
Design and evaluation of novel cancer drugs targeting hypoxic tumour cells
The influence of route of manufacturing on the performance of amorphous solid dispersions (ASD)s
Chemically modified polymers will be created and analysed for their potential uses within healthcare
Computational drug design to identify and characterize the activity of enzyme mimetic agents.
Developing a new class of biomaterials using net carbon zero sustainable bio-circular economy protocols for drug delivery and wound healing applications
Targeting novel strategies for the challenging development of sustainable photosensitisers based on Earth-abundant metal-ion complexes
Exploring Fluoxetine's anticancer mechanism by identifying its molecular targets using innovative biorthogonal chemistry
Identification of material properties required to better withstand the high dynamic loads seen at railways switches and crossings
Addressing key computational challenges in breast cancer detection to improve the survival rate of patients
Mathematical and statistical modelling of the sample preparation process for cryoelectron microscopy
Radiation damage study of innovative accident-tolerant fuel (ATF) cladding materials of SiC-based composites
Development of a next generation real-time metrology surface instrumenstation top upgrade R2R manufacturing processes to responsive mode
This project investigates human-computer interaction in user-participatory musical performances within an extended reality (XR) framework
In-situ ion irradiation of nuclear materials exposed to corrosive liquid metals (Pb, Pb-Bi eutectic)
XCT Enhancement with Iterative Reconstruction
Sensing and context awareness for safe human-robot interaction (HRI)
Develop automated image processing algorithms to identify squats and grinding marks in photographs,
Advancing Battery Sustainability with Inspection
Design an AI system that intelligently monitors and predicts mental wellbeing of university students
Integrating AI-Driven Computer Vision with Ultra-Precision Machining for Enhanced Manufacturing Accuracy and Efficiency
Develop models and algorithms to estimate wind turbine parameters without relying on physical sensors
Using mathematical modelling of pipe flow to improve techniques to measure flow types
Enhancing Manufacturing Efficiency and Material Performance through Hybrid Machining (combination of additive and subtractive) Technologies
Integrating Additive Manufacturing Simulation with Metaverse: A Digital Twin-based Approach
Develop millimetre-wave optically transparent antennas and metasurfaces
Developing optimised osseo-integration surfaces and understand how this relates with infectious biofilms in implants for immune-compromised patients
Developing and validating CFD model for desing optimisation of gas turbine running with renewable fuels
Experimental and numerical investigation on combustion and emissions of carbon-neutral fuels in combustion engines
Manufacturing of fibre-reinforced bio composites and utilising novel fibrous networks for functional apparels
Developing cyber-physical system for emissions-free valves in Hydrogen-energy and O&G.
Developing a sensing system along cyber-physical system as well as multiscale model for developing digital monitoring system for mechanical polishing process
Designing intercranial drug delivery systems to prevent tumour recurrence following surgery to remove glioblastoma
Creating non-natural CYP enzymes with alternative metals to expanding their catalytic scope
Investigate explainable AI approaches to enhance the resilience of UK and global supply chains
Implementation of force sensors onto industrial robot arm (KUKA robots) to provide sensory feedback to perform maintenance tasks on delicate and fragile surfaces
Gain insights into visco-plastic behaviour of 3D-printed polymers and biomechanics of teeth movement
Novel CPSs insider threat detection using AI privacy-preserving methods, integrating socio-technical approaches for enhanced security
Argument mining in the context of legal decision making in the context of legal decision making
Automate event logs analysis to detect security breaches and propose mitigative and preventative solutions
Cyber-physical systems for acoustic-emission/vibration/ultrasonic measurement-based smart monitoring of pipelines/storage-tanks/vessels/valves
Explore deep learning approaches on multimodal data to enable robots to manipulate complicated objects
Leveraging advanced time-series analysis and AI techniques for defining earthquake regions to inform Early-Warnings
Development and analysis of novel metal allows for extrusion additive manufacturing
A novel approach to determining and characterising chemical damage in fabrics for criminal investigation
Developing & validating additive manufacturing parameters for printing medical grade magnesium alloy
Project Code |
EPSRC_2024_01 |
Supervisory Team |
Main Supervisor: Dr Anke Bruning-Richardson Co-Supervisor: Dr Yun Wah Lam |
This project aims to develop a biomimetic organoid model that recapitulates the biochemical, morphometric, and mechanical characteristics of Glioblastoma (GBM). This model comprises (1) patient-derived glioma cells, cocultured, (2) in hypoxic conditions, with (3) a panel of Crispr-activated fibroblast cell lines overexpressing extracellular matrix (ECM) proteins that reflects the stoichiometric composition of the GBM and normal matrisome, in (4) 3D-printed scaffolds that mimic GBM stiffness and morphometry, and (5) imaged using a custom-designed AI-driven algorithm. This system allows the unprecedented modelling and manipulation of tumour ECM milieu and will provide a more accurate model for GBM biology and drug discovery.
This project aims at transforming our understanding of glioblastoma (GBM), an aggressive and lethal brain cancer. Current therapies for GBM are largely ineffective, resulting in a bleak prognosis for patients. The existing models in GBM research, primarily based on 2D or 3D cultured cell lines, oversimplify the complex tumour microenvironment, resulting in limited clinical relevance. While some "tumour-on-a-chip" models have been explored, they often fall short of replicating the full biochemical and physiological complexities of GBM.
A growing body of evidence highlights the pivotal role of the extracellular matrix (ECM) in GBM growth, metastasis, and drug responses. Recent studies have revealed that ECM genes are closely linked to GBM prognosis, suggesting ECM remodelling is linked to GBM fatality. However, most cell culture models neglect to incorporate biologically relevant ECM components. Frequently used ECM materials such as rat tendon collagen and Matrigel bear little resemblance to natural tumour ECM.
To address these limitations, this project proposes a novel approach to engineer the ECM microenvironment for cell culture. We will employ the CRISPR-activation system in human fibroblasts to selectively induce the expression of specific ECM genes known to be differentially expressed in GBM biopsies. This precise genetic modification method, which operates at endogenous genomic loci, ensures that the overexpressed ECM proteins undergo correct post-translational modifications and secretions. The project will generate approximately ten CRISPR-activated fibroblast lines, each overexpressing an ECM protein highly expressed in GBM and another set overexpressing proteins in normal brain tissue. Patient-derived glioma cells will be cocultured with these engineered fibroblasts to create heterotypic tumour spheroids. These spheroids will then be placed into 3D printed scaffolds, carefully adjusted to match the morphometric and mechanical properties of GBM.
To enable non-invasive visualization of GBM cell growth and migration within this model, the researchers will stably transfect GBM cells with fluorescent-protein (FP) reporters, including markers for mitosis, focal adhesions, and cell cycle stages. Timelapse live-cell imaging data will be meticulously analysed using Cloudbuster, a custom-made AI-driven algorithm.
In summary, this project aims to establish a cutting-edge in vitro model for studying GBM, potentially reshaping our fundamental understanding of the tumour microenvironment. The incorporation of live-cell imaging, FP-reporter cell lines and AI will make this model adaptable for high-throughput screening in the future. The innovative technique of engineering tailored matrisomes using CRISPR-activated fibroblasts has broader applications in ECM biochemistry and can be adapted to other areas of cancer biology and tissue engineering.
Project Code |
EPSRC_2024_02 |
Supervisory Team |
Main Supervisor: Professor Chenyu Du Co-Supervisor: Professor Parikshit Goswami |
Synthetic polymers in textile waste are a major contributor to global plastic waste crisis. In the UK, over 1 M tonnes of textile waste are generated annually. This project aims to develop an integrated waste refining strategy for extracting plastic and cotton fibre from textile waste to be used as raw materials in the textile industry. Cellulose fibre in textile waste will be partially breakdown by enzymatic hydrolysis allowing PET fibre to be separated (e.g. extraction using very low cost ionic liquid alkylammonium $0.78/kg). Then recycled cellulose/PET will be characterised and then re-spun into new fibre for making textile products.
Textile industry is a major plastic consumer and plastic waste generator. In 2017, 438 M tonnes of plastic were produced worldwide; 62 M tonnes were used in the textile industry and 158 M tonnes in plastic packaging. While plastic packaging has attracted attention and concern for some time, textile waste has only become prominent recently, in part because recycling textile waste is a highly challenging task. This project aims to develop a novel integrated waste refining process targets recycling 80% of plastics (specifically, polyethylene terephthalate, PET) and 80% of cotton fibre.
Textile wastes with different ratios of cotton and PET compositions will be used and classified. Pre-treatment techniques, including dilute acid treatment and abrasion pre-treatment, will be investigated to determine the effects of exposing the cellulose to enzymatic treatment. The in-house produced cellulase for controlled partial degradation of cellulose in the textile waste will be investigated. The reduction of the Degree of Polymerisation (DP) of cellulose at various conditions will be determined, then fed into an Artificial Intelligence for learning and modelling the relationship between operation conditions and degradation degrees. Separation of PET after cellulose degradation will be explored. The recycled PET will be precipitated as clean polymers. The fibre-forming properties of the recovered PET will be evaluated via appropriate fibre-forming technologies. In the meantime, the degraded cellulose will be converted to recycled cellulosic fibre. The application of the novel recycled cellulosic fibre will be explored.
Qualification:
1st in BEng/MEng Chemical Engineering or relevant Engineering degree; or relevant Master of Research degree.
Key Skills:
Bioprocessing; Material science
Project Code |
EPSRC_2024_03 |
Supervisory Team |
Main Supervisor: Dr Duncan Gill Co-Supervisor: Professor Roger Phillips, Dr Simon Allison, Dr Marco Molinari |
Hypoxia (low cellular oxygen concentration) in tumour cells is associated with increased metastasis and low survival rates. Existing approaches to selective targeting of hypoxic cell, based on prodrug versions of existing anti-tumour agents activated by metabolic reduction, have met with limited success. In contrast, distinct metabolic pathways evolved by tumours to process carbohydrates in the absence of oxygen present alternative unexplored targets. In particular, the unique role of mannose phosphate isomerase (MPI) within the hypoxic tumour cell makes it an appealing therapeutic target, but further study is limited by the quality of available chemical matter, which this project will address.
One of the characteristic features of solid tumour biology is a poorly organised and dysfunctional blood supply, leading to the establishment of a ‘microenvironment’ characterised by regions of low oxygen tension or hypoxia. Cells that reside in this microenvironment grow and metastasise aggressively and are poorly responsive to chemotherapy. Despite over 5 decades of research into the development of hypoxia activated prodrugs, no hypoxia targeted therapies have been approved for clinical use.
A different approach is therefore required. In addition to low oxygen tension, cells in this microenvironment also exist under nutrient starvation conditions and low levels of glucose in particular. This raises the question of how hypoxic cancer cells generate energy when both glucose and oxygen are restricted. Recent studies in our laboratories have shown that mannose can replace glucose as a source of nutrients for glycolysis and the enzyme mannose phosphate isomerase (MPI) plays a key role in this ‘metabolic switch’. The purpose of this proposal is to develop potential inhibitors of MPI and to test them to see if they inhibit MPI and preferentially kill hypoxic/glucose starved cells.
The chemistry phase of this project (DMG) will identify potent MPI inhibitors to facilitate biological studies (RMP & SJA) and optimise these structures into potential drug candidates. Our preliminary modelling studies have identified a series of core structures that are readily synthesised by well-established and robust preparative chemistry, and potentially adaptable to a prodrug form designed to be activated under a reductive metabolic manifold, as is typical of hypoxic cells. Once the best chemotype has been identified, the design of more potent inhibitors will be guided by computational modelling of drug–target interactions (MM).
In summary, the student will join a team of multidisciplinary scientists with an interest in drug discovery. The integration of chemical synthesis with biological/pharmacological evaluation will provide the student with excellent training in the drug discovery process that will generate intellectual property, high quality publications and potential routes to commercialisation.
A MChem or BSc (Hons) qualified student with a strong interest and practical experience in synthetic organic chemistry is essential. Interest in medicinal chemistry and modelling of ligand – protein interactions is also desirable. The candidate should also have a keen interest in being trained in new experimental methodologies, such as the development of biological assays to evaluate the compounds prepared.
Project Code |
EPSRC_2023_04 |
Supervisory Team |
Main Supervisor: Dr Kofi Asare-Addo Co-Supervisor: Dr Martina Whitehead, Prof Liam Blunt, Dr Karl Walton |
Many new pharmaceutical compounds exhibit poor solubility and as such many strategies are used to mitigate against this. One of the formulation approaches that is being used is amorphous solid dispersions (ASD). This project also involves the use of a dissolution imaging technology in the early decision making of the ASDs manufactured using various routes of manufacturing. These manufacturing routes will include but not limited to hot-melt extrusion, vacuum-melt, spray drying, freeze drying and additive manufacturing (3D printing). Appropriate dosage forms for the produced ASDs of active the active pharmaceutical ingredients (APIs) will be investigated to determine their performance and suitability.
It is important to identify risks earlier in the drug development process especially for new-to-market drugs that must be formulated into a dosage form. Solid-state transformation which sometimes occurs in the formulation can affect efficacy and manufacturability.
The project therefore aims to manufacture ASDs via various manufacturing techniques such as hot-melt extrusion, vacuum-melt, spray drying, freeze drying and additive manufacturing (3D printing). Various API’s that present a spectrum of challenges will be identified as candidates (models) for the manufacturing of the ASDs. Commercial and novel polymers will also be sought as the appropriate matrices for the API’s. Manufacturing of ASDs with the APIs will be conducted using various manufacturing techniques and storage conditions (stability testing) will be investigated.
These formulations will be characterized further using solid-state characterization techniques such as DSC, FTIR, XRPD, TGA and NMR. The understanding and characterization of the appropriate dosage form will shed light as to where drug molecules can potentially change form during manufacturing, storage, and when dosed to the patient.
The novel UV dissolution imaging technique will be used with the traditional or compendial dissolution testing to aid an understanding of the complex mechanisms associated with potential form changes and its impact on the dissolution processes.
Candidates are expected to hold (or be about to obtain) a minimum upper second-class honours degree (or equivalent) in a related area/subject. Areas may include (but are not limited to): Biology, Pharmacy, Chemistry, Pharmaceutical Science, Pharmaceutical Chemistry. Candidates with previous laboratory experience are particularly encouraged to apply.
Training will also be provided in methods of data collection, statistical analysis and graphical presentation. Oral presentation skills will be developed through presentation of the results at internal laboratory meetings and both national and international conferences.
Project Code |
EPSRC_2024_05 |
Supervisory Team |
Main Supervisor: Professor Laura Waters Co-Supervisor: Dr Leigh Fleming |
There is no validated alternative to animal testing for determining penetration of compounds through skin. Researchers have proposed several methods to predict transdermal permeation yet none have achieved the simplicity and predictive ability required to realise their incorporation into analytical studies and remove the need for animal testing. The work proposed here will resolve this issue by creating optimised chemically modified skin-like materials for permeation prediction.
The aim of this project is to develop multifunctional polymer-based materials suitable as skin replacements. These functional materials will incorporate polymer-based membranes to replicate the rate and extent of permeation through human skin for use by researchers when analysing new compounds for permeation analysis, and to potentially incorporate antimicrobial activity for skin replacement material for wound dressing.
Preliminary permeation data using a synthesised material confirms the concept with successful manufacture of a modified material and desired percentage reductions in permeation compared with standard polymer membrane: ethyl paraben - 82 %, ibuprofen - 81 %, ketoprofen - 49 %, lidocaine - 73 % and tetracaine - 93 %. These preliminary (unpublished) results imply it is possible to more closely match clinical data, thus justifying its use as a skin mimic. The process will require the development of a number of more complex, modified polymer formulations and surfaces, and the integration of inorganic materials.
A variety of analytical techniques will be used to characterise these composite polymers and help understand how their performance relates to their structure and composition. The main analysis will be surface metrology, by using both optical interferometry and atomic force microscopy (AFM) to characterise modified surfaces and determine function from a topographical perspective. The AFM will also be used to develop a methodology for the characterisation of mechanical properties at the nanoscale. Functional testing will evaluate the materials use as a mimic in the development of standards to evaluate skin/device surface interactions where skin damage may occur. This aligns with a current KTP project to investigate evaluation of mattress performance for pressure ulcer prevention and work in the university SMART house to develop a digital twin for the purposes of predictive health maintenance.
This work is therefore an exciting collaborative project across research areas incorporating chemical modification, surface metrology and pharmaceutical analysis.
Following collation of data after the first year of the project, it is envisaged that two significant grant applications will be submitted using the results as ‘proof-of-concept’ data. One of these applications will be to EPSRC in the manufacturing of functional materials for healthcare technologies, and the second to a charity or group such as DEBRA (debra.org.uk) that allows more bespoke medically-oriented proposals, in our case on skin mimics.
A suitable candidate should have (or be expected to soon obtain) a degree in a STEM subject including Chemical/Pharmaceutical Sciences, Biochemistry, Biological Sciences, Physical Sciences, Materials Science, Chemical Engineering and any other engineering degrees or related subject areas that are willing to engage with Materials Science.
Project Code |
EPSRC_2024_06 |
Supervisory Team |
Main Supervisor: Dr Marco Molinari Co-Supervisor: Professor Roger Phillips, Dr Stavros Christopoulos |
Nanozymes are nanomaterials that mimic enzyme activity. These nanomaterials have been shown to act as potent anticancer agents. However, controlling their reactivity in the biological environment is hindered by the complexity of their surface morphology and speciation. Here, we apply computational design, including ab initio and machine learning approaches to generate complex nanozymes with differing therapeutic activity. Tuning nanozymatic activity is key to target simultaneously multiple reaction pathways of cancer development. Hence, this project paves the way for the rational design of highly active nanozymes that show great promise in the treatment of cancer.
Nanozymes are enzyme mimetic agents based on inorganic materials. Many metal oxides can be classified as nanozymes if they can replicate an enzyme activity. Hence, nanozymes have the benefits of both engineered nanomaterials and naturally occurring enzymes. Although nanozymes have been used in the biomedicine field, it is still challenging to obtain fundamental insights into their catalytic performance. Aside challenges of bioconjugation and biocompatibility, enzyme-like reaction mechanisms are still fundamentally unexplored due to the complexity of the nanomaterials’ surface. To complicate the matter, local biological environments may drive structural changes, and so nanozymatic activity may change. One of the most promising activities of nanozymes is related to the scavenging and the production of reactive oxygen and nitrogen species. Inducing the formation of reactive species is particularly valuable when it comes to kill cancer cells.
In this project, we will explore the mechanisms of reaction involving reactive oxygen and nitrogen species using ab initio methodologies, which can provide accurate insights at the atom level into the activation energies and the intermediate states of such reactions. We will aim to explore the chemical landscape using computational screening, which will ultimately inform targeted experimental synthesis of active nanomaterials. Tailoring the nanozymes surface morphology and speciation, and the strength of the binding of the reactive species to the surface, will allow us to map the chemical landscape of nanozymatic performance.
The chemical landscape of nanozymatic activity is however extremely complex as any structural and chemical feature of the nanozyme may perform simultaneously a different activity. Tuning nanozymatic activity is therefore a challenge and the full control of the nanozyme can only be achieved if the relationship between the structure and the activity can be fully addressed. To tackle this complexity, we will employ machine learning approaches to speed up the screening of nanozymatic reactivity via the development of tailored reactive force fields.
Ultimately, this project envisages to attain control over the nanozymatic activity by harnessing novel computational materials synthetic strategies and apply them to efficient screening for therapeutic cancer therapy.
Previous knowledge on simulation methods or programming is desirable but not necessary. Applicants should have, or expect to achieve, at least a 2.1 honours degree, in Chemistry, Physics, Pharmaceutical Sciences, Biochemistry, Biological Sciences, Physical Sciences, Materials Science, Chemical Engineering and other relevant STEM degrees. Fundings are available for home students only (UK residents, and EU students with (pre)settled status, pending residency requirements).
Project Code |
EPSRC_2024_07 |
Supervisory Team |
Main Supervisor: Dr Muhammad Usman Ghori Co-Supervisor: Professor Barbara Conway, Dr Shan Lou, Professor Jane Jian |
Developing new materials based on environment-friendly bio-circular economy and sustainability protocols is challenging. However, the invention and implementation of artificial intelligence and machine learning methods promise to accelerate and improve the traditional complex pathways. This project will develop new green sustainable materials with net-zero carbon footprint using machine-learning approaches while underpinning the novel bio-circular economy protocols for drug delivery and wound healing applications.
New sustainable materials are strategic drivers for societal advantage as they reinforce the use of methods and technologies that positively impact the global challenges associated with climate change, resource and economic sustainability. This avenue has opened some new scientific fronts that require a methodical approach for understanding, development and translation. To develop new materials using net carbon zero sustainable bio-circular economy approaches necessitate a negotiation with a complex space of instrumental engineering and chemical sciences while underpinning the different experimental conditions require for preparing these materials. Hence, developing, characterising and scaling up new material production is challenging, but artificial intelligence and machine learning promise to accelerate and improve the traditional complex pathways. Therefore, this project is aimed at developing a class of green sustainable materials with net-zero carbon footprint using machine-learning approaches while underpinning the novel bio-circular economy protocols for healthcare applications.
The key deliverables of this project are divided into the following stages.
(1) Artificial intelligence-based machine-learning method development using MATLAB to develop new materials.
(2) Development of an experimental protocol for scaling up the material production.
(3) The materials' functional, structural and chemical properties will be evaluated using a range of characterisation techniques, for example, XRD, FTIR, DSC, SEM, TEM, AFM, SAXS, and hot stage microscopy, white light interferometry and different microbial, chromatographic and wet chemistry assays.
(4) Investigation of the developed optimised materials for potential applications in drug delivery and wound healing. This involves designing novel smart drug delivery systems and wound dressings. Techniques such as drug encapsulation, in-vitro drug release, wound healing assays, texture analysis, microtomography, optical coherence tomography will be employed to assess the quality of the developed formulations.
Project Code |
EPSRC_2024_08 |
Supervisory Team |
Main Supervisor: Professor Paul Elliott Co-Supervisor: Dr Paul Scattergood |
Photoactive metal complexes for applications in solar energy conversion, photocatalysis, phototherapeutics and medical diagnostics has attracted enormous interest. These materials typically rely on the use of expensive metal elements that are amongst the rarest in the Earth’s crust. It is therefore imperative to develop new materials for these applications based on cheaper and far more abundant elements such as iron. This, however, presents a considerable scientific challenge and has become a top-topic in current chemical science. Building upon recent insights, this project targets highly novel 1st row transition metal sensitiser complexes, the investigation of their fundamental optoelectronic properties and probes their potential in luminescence and photocatalytic applications.
The project will involve organic ligand synthesis, the preparation of their metal complexes and their characterisation by a wide range of techniques (NMR and UV-visible absorption & emission spectroscopy, cyclic voltammetry with excited state investigations in collaborative transient-absorption spectroscopy).
The successful candidate will hold or be about to obtain a 1st class honours degree in Chemistry or the equivalent.
Project Code |
EPSRC_2024_09 |
Supervisory Team |
Main Supervisor: Dr Yun Wah Lam Co-Supervisor: Dr Farideh Javid, Professor Liam Blunt |
This interdisciplinary project aims to unravel the anticancer mechanism of Fluoxetine, a common psychiatric drug, through innovative biorthogonal chemistry techniques. We propose to synthesise clickable, PEGylated Fluoxetine derivatives and examine their intracellular and biochemical interactions in colon cancer cells. Using three biorthogonal methods - direct pulldown, in-cell pulldown, and proximity labelling - we aim to identify and analyse Fluoxetine-interacting proteins. This approach will potentially identify new molecular targets and pathways influenced by Fluoxetine, offering valuable insights for its potential use in cancer therapy. The methodologies established here could be adapted for exploring the mechanisms of other drug candidates.
The understanding of the mechanism of action (MoA) of small-molecule drugs, which entails the interactions between the drug and its target molecules, is a cornerstone of pharmacological research. To date, drug candidates are often empirically identified based on high-throughput phenotypic screens. For these compounds, even with promising efficacies, MoA is still essential for the rational design of analogues with optimised therapeutic activities and for the understanding of potential side effects and pharmacodynamic properties.
This project focuses on understanding drug MoA through biorthogonal chemistry. Using Fluoxetine as a proof of concept, we aim is to create clickable, PEGylated derivatives of the drug and study their interactions within cancer cells by advanced imaging and proteomic techniques.
Aim 1: Fluoxetine will be conjugated to a clickable alkyne group via a Polyethylene Glycol (PEG) linker. We will explore various linker lengths to find the most effective derivative. The products will be purified by HPLC and validated by mass spectrometry.
Aim 2: We will evaluate the efficacy of the PEGylated Fluoxetine in inhibiting the growth of a panel of cancer cell lines compared to the original drug. The most bioactive derivative will be selected for further studies.
Aim 3: Using a clickable fluorescent dye, we will investigate the intracellular distribution of the PEGylated Fluoxetine in colon cancer cells. In parallel, UV dissolution imaging will be used to visualise the uptake process and intracellular distribution of Fluoxetine. Data from the two techniques will be compared.
Aim 4: We will employ three biorthogonal methods to study Fluoxetine targets: direct pulldown, in-cell pulldown, and proximity labelling. Direct pulldown involves immobilizing PEGylated Fluoxetine on beads for interaction with cell lysates. In-cell pulldown studies interactions within living cells before lysis. Proximity labelling uses an APEX2 probe for the in situ biotinylation of proteins near the drug, facilitating their isolation and proteomic identification. These methods allow the mutual validation of data, establishing the core Fluoxetine interactome in cancer cells.
Aim 5: Bioinformatic analysis of the interactome will identify affected cellular components and molecular pathways, forming hypotheses about Fluoxetine’s role in cancer. These hypotheses will be tested through gene knockdown and overexpression experiments.
In summary, this project aims to identify the molecular targets and pathways influenced by Fluoxetine in cancer cells, which could inform its potential use in cancer therapy. The methods developed could also be applied to study the action mechanisms of other drug candidates, aiding in drug development and optimisation.
Qualifications:
Project Code |
EPSRC_2024_10 |
Supervisory Team |
Main Supervisor: Professor Adam Bevan Co-Supervisor: Dr Sam Hawksbee, Professor Yann Bezin, Professor Jay Jaiswal |
Railway switches and crossings (S&C) experience very demanding operating conditions as reflected in their disproportionate share of maintenance (24%) and renewal (23%) budgets for assets accounting for just 5% of the network. Low hardness of cast crossings causes plastic deformation that destroys the desired profile instigating accelerated degradation, frequent maintenance interventions, and reduced track availability. The research will identify the material properties required to better withstand the higher dynamic loads and supplements current research on profile optimisation to minimise loads. The study will also identify appropriate materials for Additive Manufacturing of crossings to reduce first installed and life cycle costs.
Although cast Austenitic Manganese Steel (AMS) crossings have served the industry well over many years and account for the vast majority crossing on UK network, their low installed hardness leads to appreciable plastic deformation and even shelling in the early stages both of which require frequent and costly maintenance interventions. Although the use of explosive hardened crossings reduces the rate of such degradation, weld restoration of deformed and damaged crossings continues to be practiced to enhance crossing life. The current manual repair processes have little process control resulting in inconsistencies in repair integrity and long repair durations. Increasing volume and density of traffic are expected to further exacerbate this position. Hence the two key areas requiring further research and development are (1) to minimise the dynamic loads experienced (being addressed by existing research within the Institute of Railway Research) and (2) the identification of alternative materials for crossings which better withstand the applied loading conditions. This second requirement forms the objective of proposed research.
The research project will be focussed on the following areas:
Define and undertake laboratory-based tests to simulate loading conditions experienced on the crossing nose and consequent effect on plastic deformation. The test will be designed to investigate the correlation of deformation of crossing nose shape to measured tensile and other material properties.
Investigate influence of impact loading and contact stresses on sub-surface initiation of cracks and their correlation with material properties.
Undertake comparative wear and RCF tests, using the large diameter twin disc rig already available within the Institute of Railway Research, on materials in current use and selected alternative materials that would be suitable for additively manufactured crossings.
Investigate the influence of cladded deposits on the reduction in roughness growth on wheel-rail contact surface and consequent benefits of reducing squeal noise.
Investigate the use of cladding material as an in-situ permanent modification of coefficient of friction to eliminate the need for regular grease lubrication. This aspect of the work is designed to provide a cost-effective solution to the costly current practice of frequent lubrication of rails at throats of busy railway stations with very tight curves and slow traffic.
The candidate should hold a good degree in a relevant engineering or science discipline. Good mathematical skills and an aptitude to learn the required experimental techniques are essential. Knowledge of the railway engineering and materials science would be strongly beneficial.
Project Code |
EPSRC_2024_11 |
Supervisory Team |
Main Supervisor: Professor Andrew Crampton Co-Supervisor: Professor Mauro Vallati |
Significant advances in image-based deep learning, together with rapid advances in medical imaging technology has provided new opportunities for novel research in detecting cancerous tumours from breast cancer images (mammograms). The impact of breast density on early tumour detection, imbalanced datasets, reducing false positives/negatives and increasing interpretability/explainability will all be examined. This work will examine how clinical data, such as breast tissue density, lesion size, location and patient outcomes, etc, provided with the OPTIMAM image sets, can be incorporated to provide a novel multi-modal prediction model. This approach aims to overcome known difficulties associated with a lack of interpretability.
OPTIMAM is a large-scale digital mammographic image database containing approximately 20,000 full-resolution biopsy-confirmed segmented cancer images, 2400 biopsy-confirmed segmented benign images, and 6,000 histologically confirmed masses that have not yet been segmented. Plus, approximately 40,000 normal breast images. The dataset contains 32 clinical features, including numerical, nominal, and binary attributes. A key feature of the OPTIMAM dataset is the inclusion of this clinical data, which allows researchers to explore the relationship between imaging features and clinical variables, potentially enhancing the performance of predictive models.
Though much has been done to advance the use of ML techniques for breast cancer detection, many challenges still remain. Examples are:
An ideal candidate would have a good honours degree in computer science, mathematics or a related scientific discipline. Have some knowledge of machine learning and/or data modelling/analysis. Be competent in programming (preferably Python). Have good self-management and organisational skills. Have good communication skills (written and verbal).
Project Code |
EPSRC_2024_12 |
Supervisory Team |
Main Supervisor: Dr Ann Smith Co-Supervisor: Professor William Lee |
Cryoelectron microscopy (CryoEM) is a technique for visualising biological samples by immobilising them in thin layers of ice for electron microscopic imaging and analysis. A challenge in creating samples for this technique is controlling the thickness of the ice film. If the film is too thick the electrons will not be able to penetrate it and there will be no signal. If the film is too thin there will be holes in it and thus there will be no sample for the electrons to interact with. Mathematical and statistical modelling will be used to address this challenge.
A primary obstacle in CryoEM is the precise control of the ice layer thickness during sample preparation. A detailed description of the process of forming the thin films can be found in the literature see for example [1]. Current methods, predominantly manual plungers, yield inconsistent and non-repeatable results even when operated by skilled technicians. This inconsistency is a major barrier to the broader adoption and effectiveness of CryoEM as sample quality directly impacts the accuracy and reliability of the microscopic analysis.
To address this challenge a data-driven method to create samples with the optimal ice layer thickness will be developed, ensuring robust high-quality biological samples. This innovation aims to broaden accessibility and application of this critical technique.
A comprehensive dataset, designed by Dr Smith was collected at the National Physics Laboratory from Linkam Scientific Instruments Ltd.’s Cryogenium robotic system under the Analysis for Innovators programme, and will be used to develop a control system to autonomously optimise ice layer thickness.
Collaboration with Linkam, who are at the forefront of electronic measuring and testing equipment innovation, adds expert guidance enhancing the project’s scientific underpinnings and ensuring it aligns with the latest advancements in CryoEM methodologies. In addition, the supervisory team have a rich history of collaborative work and expertise ideally suited to the proposed analysis, a combination of mathematical and statistical modelling.
Mathematical Modelling: A fluid mechanical model of the formation of the ice film will be formulated as a system of partial differential equations describing mass, momentum, and energy balances. These equations will be analysed using nondimensionalisation and asymptotic methods to develop a reduced model which will be solved numerically to predict the profile of the film. These results will be used to inform statistical modelling.
Statistical Modelling: Statistical models will be developed incorporating the process parameters to predict the quality of samples, their accuracy and reproducibility. Multivariate regression models and classifiers will be generated using a combination of data driven and physical modelling methodologies. Exploratory data analysis of imaging data will be conducted to identify input parameter distributions, correlations, explanatory ability of variables and the dependency of the intensity of the transmitted electron beam on their value. The number and formation of any clusters, with respect to sample quality, will be determined and used to inform subsequent modelling of the process.
[1] Koning, R.I., Vader, H., van Nugteren, M. et al. Automated vitrification of cryo-EM samples with controllable sample thickness using suction and real-time optical inspection. Nat Commun 13, 2985 (2022). https://doi.org/10.1038/s41467-022-30562-7
Knowledge of data analysis, mathematical modelling, differential equations, and programming. This project would suit a candidate with a background in mathematics, statistics, engineering, physical science, or computer science.
Project Code |
EPSRC_2024_13 |
Supervisory Team |
Main Supervisor: Dr Graeme Greaves Co-Supervisor: Professor Konstantina Lambrinou |
This project will investigate the effects of radiation effects in innovative silicon carbide composites for applications in next generation fission and fusion reactors. Cutting-edge composites designed & produced in collaboration with European partners will be assessed in terms of their radiation tolerance using the unique MIAMI facility (https://en.wikipedia.org/wiki/MIAMI_Facilities). The student will perform in-situ ion irradiations within an electron microscope, monitoring in real-time at the nanoscale the dynamic evolution of radiation damage in various composites. This project is aligned with the European SCORPION project (https://cordis.europa.eu/project/id/101059511) and thus there will be the opportunity to travel and present at the various project meetings.
SiC/SiC composites are considered candidate fuel cladding materials for advanced fission reactors and candidate structural materials for specific applications in fusion reactors. SiC/SiC composites combine excellent refractory properties with high strength and pseudo-ductility, however, they are characterised by various (nuclear-system-specific) technical shortcomings, such as their poor compatibility with water/steam and their conditional (e.g., temperature-dependent) response to radiation swelling, which need further investigation.
Materials for advanced fission and fusion reactors are expected to perform reliably upon prolonged bombardment by energetic neutrons at high temperatures. Neutrons cause atomic displacement, creating point defects that can migrate and form larger defects, such as voids and dislocations. Neutrons might also trigger transmutation reactions producing inert gases that can form nanometre-sized bubbles. Combined with high temperatures, these processes can degrade mechanical and physical properties and, ultimately, cause material failure.
The successful candidate will be trained to independently operate various pieces of scientific equipment within the MIAMI facility, including SEM, TEM, and the in-house particle accelerator systems. These will be systematically used to form ion beams to act as neutron surrogates, which will enable us to assess the radiation response of different SiC/SiC composite material variants, each tailored to the requirements of the targeted (fission/fusion) application. The irradiation experiments, performed over a wide range of irradiation conditions (e.g., ion/proton beam energy, temperature, ion/proton flux, damage dose) are expected to simulate the material’s in-service lifetime within mere hours, and provide invaluable insights into the complex material degradation processes, thereby driving the further optimisation/evolution of these innovative nuclear. This PhD project is also expected to lead to a greater understanding of the fundamental physics governing the mechanisms of radiation-induced damage in SiC/SiC composites intended for nuclear (fission/fusion) applications.
The candidate must hold a bachelor’s or master’s degree in Materials Science, Physics, Chemistry,
Nuclear Engineering or related discipline.
Additionally, knowledge and understanding of any of the following would be desirable:
Project Code |
EPSRC_2024_14 |
Supervisory Team |
Main Supervisor: Dr Hussam Muhamedsalih Co-Supervisor: Professor Liam Blunt, Dr Dawei Tang, Professor Jane Jiang |
The University of Huddersfield’s team from CPT will develop a novel surface metrology instrument to investigate thin film quality of conductive coating produced by roll-to-roll manufacturing processes. This project based on the awarded “Responsive Manufacturing of High Value Thin to Thick Films” project EP/V051261/1. The EPSRC project will focus on the enabling roll-to-roll production lines to be responsive to disruptions in the substrate or coating material being processed.
Surface science has become an important part of the wide engineering and physical science landscape and has received significant funding from the EPSRC since 2011. This investment reflects the importance of surface science in many areas of which ‘digital manufacturing’ is one. Nowadays, the roadmap of manufacturing surface metrology aims to integrate in-line sensors within of manufacturing processes to guarantee the specification of every part. As such, direct interaction between measurements and manufacture is essential in production process, for delivery of industry 4.0 compatible processes, and leads to significant savings in both time and cost.
However, most commercial-use sensors are still laboratory-based and need further development to be used for in-line measurement applications. Real topography measurement for precision surfaces is usually achieved optically by capturing multiple frames to cover a vertical range of several microns or more. The scanning time for methods such as coherence scanning interferometry (CSI) and wavelength scanning interferometry (WSI) is considered a limiting factor for using such interferometry systems for in-line measurement of moving surfaces3-4. For example, roll-to-roll processing, which is a fundamental approach in meeting the demand for very large numbers of functional flexible electronic devices, needs metrology systems that can cope with the manufacturing throughput (typically > 80 meters/min) and have the ability to handle and process very large amounts of data. There are many optical systems that can only inspect the surfaces visually without producing surface metrics but none that deliver quantitative information.
In addition to the aforementioned challenges in sensors’ development, one of the main challenges for in process metrology is how to assess large and multiple measurement data sets. Many measurement files (typically larger than 800 Mbyte) will be produced over a single 500 mm sheet width. Therefore, a new paradigm in data handling methods is necessary for roll-to-roll process.
The aim of this proposal is to provide a metrological technology for large area precision surfaces produced by roll-to-roll processes. The technology will consist of a novel fast in-line optical instrument, a method to handle large amounts of data without interaction from the inspector, and a calibration routine for the instrument. The instrument will be used for detecting and measuring micro/nano-scale defects and critical dimensions on large-area thin film layers, for example flexile photovoltaics (PV) and aspects will be demonstrated on a laboratory R2R test rig at the University of Huddersfield.
Project Code |
EPSRC_2024_15 |
Supervisory Team |
Main Supervisor: Professor Hyunkook Lee Co-Supervisor: Dr Duke Gledhill |
Extended Reality (XR) systems enable users to interact with virtual beings and/or objects superimposed onto the real or/and virtual worlds. This project aims to (i) create an innovative XR system utilising advanced 3D virtual acoustic and visual simulation technologies, allowing users to seamlessly sing or play a musical instrument alongside real/virtual musicians integrated into the XR world, and (ii) explore the impact of acoustic and visual congruencies within the system on the user's perceived self and social presence, as well as various psychological and biomedical attributes.
Extended Reality (XR) is a transformative technology that encompasses Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR), offering a spectrum of immersive experiences that blend the real and digitally simulated virtual worlds. Spatial computing technologies for XR are advancing rapidly, and XR applications find utility in diverse industries like gaming, music, education, healthcare, and more. One major advantage of XR technologies is their ability to enhance social presence by seamlessly merging virtual and physical worlds, fostering shared experiences, and facilitating more immersive and interactive communication between individuals.
This PhD project centres around the theme of interactive musical performance in XR environments. Achieving a truly realistic and immersive experience necessitates plausible simulation not only of visual elements in the virtual world but also acoustic features from either the virtual (in VR/MR) or real world (in AR). Despite progress in VR, there is a notable absence of explorations into the possibilities presented by AR and MR systems in musical interaction applications. Furthermore, no research to date has delved into the relative importance of accuracy in visual and acoustic simulations and their congruency in enhancing the sense of social presence, well-being, and other psychological attributes, highlighting the need for in-depth research in this area.
The project is multidisciplinary, involving audio and visual computing, sound engineering, as well as psychophysical and physiological measurements. The aims of this project are as follows:
(i) To develop an XR system that enables users to collaboratively create music with virtual counterparts, whether they be other users or virtual musicians seamlessly integrated into the real or virtual world.
(ii) To explore the threshold of accuracy required in visual and acoustic simulations to deliver a highly immersive experience in interactive XR musical performances.
(iii) To examine and compare the immersive experiences offered by VR, AR, and MR, evaluating their relative effectiveness in providing users with a heightened sense of engagement and presence.
This project has the potential to create significant societal impacts. The ability to sing or play music together with virtual counterparts will transform musical collaboration and education, while providing a sense of social presence in such collaborations also has the potential to enhance mental health, such as reducing stress or overcoming loneliness.
Qualification: a taught MSc degree (Distinction), an MSc by Research degree or a BSc degree with a First grade in relevant subject areas, including computer science, game engineering, visual computing, audio engineering and electrical/electronic engineering.
Skills: Programming for XR content creation (C++), statistical analysis and signal processing tools (e.g. R, Matlab, Python), games engine (preferably Unreal Engine) and audio workstation
Experience: The ideal candidate will possess prior experience in programming game content using Unreal Engine, and/or conducting or assisting with research projects in a related field.
Knowledge: The candidate should have solid basic knowledge in game programming, 3D visual computing and audio/sound engineering.
Project Code |
EPSRC_2023_16 |
Supervisory Team |
Main Supervisor: Professor Jonathan Hinks Co-Supervisor: Professor Konstantina Lambrinou |
The deployment of Gen-IV lead-fast reactors (LFRs) demands a good compatibility between structural/fuel cladding steels and candidate liquid metal (primary) coolants, i.e., liquid lead (Pb) and lead-bismuth eutectic (LBE). Unfortunately, most structural/functional candidate steels for Gen-IV LFRs suffer degradation due to liquid metal corrosion (LMC) and liquid metal embrittlement (LME). This PhD project will attempt to understand the effect of irradiation on the degradation of different Gen-IV LFR candidate materials (steels, advanced ceramics) exposed to liquid Pb/LBE. The judiciously designed in-situ ion irradiation campaigns will employ the MIAMI-2 facility to study the synergy of irradiation and LMC on material degradation.
The newest ‘Microscopes and Ion Accelerators for Materials Investigations’ (MIAMI-2) facility at the University of Huddersfield allows the in-situ observation of radiation-induced defect formation and microstructural changes in candidate nuclear materials at elevated temperatures (up to 1300°C) that are specific to the targeted nuclear energy system. The MIAMI-2 facility is one of only 2 of its kind in Europe and around 12 globally that can provide valuable insights into the degradation of candidate nuclear materials on the nanoscale as result of the extreme operation conditions of advanced nuclear systems such as Gen-IV LFRs. This PhD project will use the MIAMI-2 facility to study the radiation-induced degradation of candidate nuclear materials (e.g., commercial stainless steels, lab-made compositionally complex MAX phase solid solutions, etc.) that have been pre-exposed to liquid Pb or LBE. The PhD student entrusted with this ambitious project will be called to design and conduct liquid metal corrosion tests, in-situ ion irradiation experiments on the MIAMI-2 facility, and detailed post-irradiation examinations (PIEs), thereby acquiring an appreciable range of scientific/technical skills.
This PhD project will use static autoclaves to understand the liquid metal corrosion (LMC) behaviour of reference structural materials (e.g., stainless steels) and innovative material solutions developed specifically to mitigate LMC in contact with liquid Pb/LBE (e.g., MAX phase solid solutions). LMC testing will mostly expose the candidate materials to aggressive conditions (i.e., high temperatures, low concentrations of dissolved oxygen in the liquid metal, etc.) in order to (a) understand the material-specific LMC behaviour, and (b) identify the most efficient LMC mitigation strategies. Testing in contact with oxygen-poor liquid metals aims at suppressing material oxidation that delays the occurrence of aggressive LMC mechanisms such as dissolution corrosion and pitting corrosion. Corroded matter will be characterised by many analytical techniques, such as scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), electron backscatter diffraction (EBSD), focused ion beam (FIB), (scanning) transmission electron microscopy (S)TEM, etc. The best-performing materials will be subjected to in-situ ion irradiations on the MIAMI-2 facility to further investigate the material response to irradiation before and after their LMC testing. The MIAMI-2 ion irradiations require the lift out of thin foils from different areas of interest using FIB; the PIE of irradiated matter will be primarily conducted by means of (S)TEM, selected area electron diffraction (SAED), etc.
Bachelor’s or master’s degree in Physics, Materials Science, or a related field.
Additionally, understanding and/or prior experience with any of the following would be desirable:
Project Code |
EPSRC_2024_17 |
Supervisory Team |
Main Supervisor: Professor Paul Scott Co-Supervisor: Dr Shan Lou, Dr Wenhan Zeng, Dr Xiao Chen |
X-ray Computed Tomography (XCT) revolutionises non-destructive industrial inspection by offering 3D insights into components' internal structures, outperforming traditional tactile and optical methods. It excels in analysing complex geometries, crucial for additive manufacturing, without damaging the component. However, XCT's precision is challenged by issues like focal spot and detector blur, scattering, and beam hardening, which impair the quality of projection images. This project seeks to overcome these obstacles with advanced iterative reconstruction algorithms, enhancing image quality and measurement accuracy by addressing these root causes of degradation and refining the reconstruction process for better inspection outcomes.
XCT has emerged as a pivotal tool for the non-destructive inspection of industrial components, offering distinct advantages over traditional tactile and optical metrology techniques. By providing detailed 3D visualizations of an object's internal structure without physical dissection, XCT allows for comprehensive analysis of complex geometries inherent in components, especially those fabricated through additive manufacturing. This capability to inspect intricate internal features without compromising the integrity of the component sets XCT apart, enabling precise evaluations that are unattainable with surface-only inspection methods, e.g. tactile and optical measurement techniques.
This project targets enhancing the efficiency and fidelity of XCT for inspecting industrial components. Despite XCT's crucial role in providing detailed internal structures of components non-destructively, its effectiveness is often compromised by factors such as focal spot blur, detector blur, scattering, and beam hardening. These issues can significantly degrade image quality, resulting in a decrease in the accuracy of XCT inspections. Focal spot blur, stemming from the finite size of the X-ray source, and detector blur, due to the limitations of the XCT detector, both reduce image sharpness. Scattering, the deviation of X-rays caused by interactions with the component, introduces noise and artifacts, while beam hardening, the preferential absorption of lower energy X-rays, leads to misleading artefacts in images.
This project aims to mitigate these challenges by implementing advanced iterative reconstruction algorithms. These techniques, by iteratively refining the image reconstruction process, promise to enhance the clarity and accuracy of XCT projection images. The project will explore new algorithms that account for the physical phenomena causing image degradation, including corrections for focal spot and detector blur, as well as compensations for scattering and beam hardening effects.
Project Code |
EPSRC_2024_19 |
Supervisory Team |
Main Supervisor: Dr Qasim Ahmed Co-Supervisor: Professor Pavlos Lazaridis, Dr Salman Bashir |
HRI are increasingly common in warehouses, production facilities and elderly care environments. To this end, robots must navigate autonomously to avoid collisions and efficiently perform tasks in dynamic environments, ensuring physical safety during interactions with humans. Effective communication and interpretation of human actions enable robots to understand and respond to human intentions, enhancing both physical and emotional safety in HRI.
The goal of this project is to design integrate sensing and communication (ISAC) solutions that enhance safe interactions between robots and humans by designing wider bandwidth (UWB, mm-Wave, THz) radar solutions that can be installed on a mobile robot to act simultaneously as a radar system, a communication channel, and a contextual interpretation tool (e.g., for activity recognition), thereby creating an integrated framework that allows safe and effective HRI.
The design of adaptive algorithms for sensing and context awareness for safe HRI, requires adjusting RF settings depending upon the available context. This will result in designing of RF angle-of-arrival (AoA) solutions for antenna array configurations. Novel design of physical layer and antenna reconfiguration algorithms that will be able to switch between radar, communication, and localization by exploiting optimal settings for long-range sensing and communication versus settings more granular radar resolutions at short range. Finally, AI algorithms for simultaneous human detection, activity recognition, and localization will be developed.
Degree/speciality: Electronic or electrical engineering with focus towards wireless communications
Skills:
1) Excellent mathematical skills and background.
2) High proficiency in Matlab, Mathematica, Maple, R, or similar programming software.
3) Solid background on wireless communications (sensing, localization, V2X is a plus).
4) Excellent written and verbal communication, including presentation skills.
5) Excellent organisational skills, attention to detail and the ability to meet deadlines.
6) Ability to think logically, create solutions and make informed decisions.
7) Willingness to work collaboratively in a research environment.
8) A strong commitment to his/her own continuous professional development.
Project Code |
EPSRC_2024_21 |
Supervisory Team |
Main Supervisor: Dr Shan Lou Co-Supervisor: Dr Chen Xiao, Professor Andrew Ball, Professor Jane Jiang Industrial Supervisor: Dr Habib Murtaza (Manufacturing Technology Centre) |
Batteries power electric vehicles, portable electronics, and solar storage systems, making their performance, safety, and reliability crucial. Yet, cell defects can compromise these aspects, posing risks. To ensure battery quality, effective defect detection is vital. X-ray computed tomography (XCT) stands out as a non-destructive method for detailed internal examination, offering 3D insights into internal flaws. This research project will focus on leveraging XCT for in-depth defect analysis in batteries, enhancing their safety and sustainability.
Green battery technology is pivotal for achieving sustainability and advancing clean energy initiatives. These batteries offer environmentally friendly energy storage solutions, which help diminish our dependence on fossil fuels and lessen ecological harm. The use of XCT in examining green batteries presents significant advantages for both manufacturers and researchers. XCT's non-invasive approach allows batteries to be inspected without inflicting any harm, thereby preserving them for subsequent use or further investigation. This method is particularly valuable for detecting various geometrical distortions (such as internal short circuits, electrolyte voids, welding burrs, delamination, and electrode misalignment) and material defects (e.g., particles, cracking, porosity, and metal contamination). These flaws can adversely affect the battery's performance, durability, and safety. Early identification of such issues enables timely corrective measures, significantly improving battery quality and dependability. Furthermore, XCT's capacity to detect flaws at early production stages leads to swift problem-solving, enhancing battery safety and reliability. Additionally, XCT facilitates research into novel battery chemistries and materials, propelling advancements in energy storage technologies.
This research initiative aims to assess the efficacy of XCT as a non-destructive evaluation technique for spotting defects in green batteries. The project is structured around three principal goals: firstly, to evaluate XCT's effectiveness in measuring internal geometries and identifying material defects within battery cells; secondly, to develop two-dimensional and three-dimensional image processing methods for analysing the internal geometries and material anomalies; and thirdly, to develop case studies that track the progression of geometric distortions and material defects through various battery charging and discharging cycles.
The project will be supported by the Manufacturing Technology Centre, a UK's leading High Value Manufacturing Catapult centre, offering both technical assistance and access to advanced XCT facilities.
Project Code |
EPSRC_2024_22 |
Supervisory Team |
Main Supervisor: Dr Tianhua Chen Co-Supervisor: Professor Mike Doyle |
The project leverages AI, particularly deep neural networks, to improve university students' mental health by forecasting potential moments of distress for early intervention. A mobile app, which is actively collecting data, has been developed that monitors students' behavioural patterns through smart devices, offering psychological assessments and self-help resources. The PhD project focuses on devising novel neural networks to analyse time series data, exploring techniques suitable for mental health context. The model's real-world effectiveness and its impact on mental health will be evaluated in collaboration with clinical experts. The developed work may be seamlessly adaptable for wider use in clinical practice.
The initiative employs Artificial Intelligence (AI) to bolster the mental health of university students by forecasting moments of distress, facilitating early detection and intervention to avert and lessen risks. This AI tool integrates seamlessly with mobile devices, monitoring individuals' engagement in educational activities and changes in behaviour—like academic performance, physical activity, sleep patterns, and unique risk indicators—hinting at mental state shifts.
A prototype app has been developed, offering features such as gathering personal traits, extracting behavioural data from smartwatches, enabling self-administered psychological assessments, journaling, and access to a wealth of self-help materials. This app is currently being used by students for data gathering. By identifying potential mental health challenges early and providing self-help tools, the project aims to empower students to recognize warning signs, anticipate what comes next, and initiate timely measures to maintain their mental well-being before it worsens.
This PhD project aims to leverage Artificial Intelligence (AI), specifically deep neural networks, for predicting outcomes from time series data collections. While the project has a strong practical application focus, it will also conduct in-depth methodological research. Initially, the project will review a wide array of contemporary AI methods suitable for time series analysis, identifying the distinct features of mental health data relevant to deep neural networks. This foundational work sets the stage for designing and implementing a neural network model tailored to the specific traits of the collected data. The effectiveness of this system will be assessed using real-world data, followed by a detailed evaluation of the system's potential to enhance student mental wellbeing, in partnership with experts in mental health from clinical and medical fields.
An ideal candidate would have a good honours degree in computer science, mathematics or a related scientific discipline. The candidate will have demonstrable knowledge in machine learning and possibly experience working with health data. The candidate should also have the proficiency in the use of mainstream data analytics platforms such as Python (preferred). Good professional communication skills (written and verbal) are also expected.
Project Code |
EPSRC_2024_23 |
Supervisory Team |
Main Supervisor: Dr Wenhang Zeng Co-Supervisor: Professor Zhijie Xu, Dr Wenbin Zhong, Dr Shan Lou |
Integrating AI-driven computer vision with ultra-precision machining represents a frontier in manufacturing technology, offering significant improvements in component quality and production efficiency. This research aims to bridge the gap between theoretical AI models and practical manufacturing challenges, paving the way for the next generation of manufacturing technologies. By enhancing the capabilities of UPM through AI and computer vision, this project will contribute to the advancement of manufacturing precision and efficiency across various high-tech industries.
Ultra-precision machining (UPM) is critical for producing components with nanometre-level precision, essential in industries like aerospace, optics, and medical devices. However, achieving and maintaining such high precision levels is challenging due to factors like tool wear, material inconsistencies, and environmental variations. Integrating Artificial Intelligence (AI) with computer vision into UPM processes offers a promising solution to these challenges by enabling real-time monitoring, analysis, and adaptive control of the machining process. This research proposes to develop and validate an AI-driven computer vision system that enhances the precision, efficiency, and reliability of on-machine ultra-precision machining operations.
Objectives
To investigate the current challenges and limitations in ultra-precision machining that can be addressed by AI and computer vision technologies.
Methodology
The research methodology will encompass:
Expected Outcomes
The CPT has the world’s most advanced ultra precision machine and state-of-the-art metrology instrument that can support this project.
Entry RequirementsAn ideal candidate for the proposed project should possess a recent undergraduate or master’s degree (1st class or distinction) in mechanical engineering or computer science or electronic engineering. Competence in modelling tools like MATLAB (or LabView), programming languages such as Python (or C++), and experience in independently developing software products using 3rd-party libraries like OpenCV are expected.
Project Code |
EPSRC_2024_24 |
Supervisory Team |
Main Supervisor: Dr Wenxian Yang Co-Supervisor: Professor Andrew Ball |
Wind farm digitisation enables real-time monitoring and optimised operation and maintenance of wind turbines, thereby playing a crucial role in reducing the cost of wind power. The present wind farm digitalisation relies on the use of numerous physical sensors, introducing complexities in hardware, escalating costs, and giving rise to reliability issues. This undermines the achievements of the UK's climate change mitigation and energy security efforts. To overcome this issue, this project aims to develop advanced virtual sensor technology, estimating crucial turbine parameters and health conditions without physical sensors, thus promising a more economical and reliable path towards wind farm digitisation.
This research project encompasses a comprehensive set of activities towards advancing the digitisation of wind farms. The overarching goal is to offer wind farm operators a more insightful comprehension of their wind turbines' operations and health conditions. This, in turn, enables optimised approaches to operation and maintenance, ultimately leading to substantial cost reductions of wind power. The key research activities include:
By addressing these research activities, the project aims to make significant strides in wind farm digitalisation, deepening our understanding, optimisation, and management of wind turbine operations under varying wind conditions and contributing to the advancement of renewable energy technologies.
Essential qualifications:
A Master degree (Merit or above) in a relevant field such as Mechanical Engineering, Electrical Engineering, Renewable Energy, Computer Science, or a related discipline.
Essential skills:
Desired skills:
Desired experience:
Project Code |
EPSRC_2024_25 |
Supervisory Team |
Main Supervisor: Professor William Lee Co-Supervisor: Dr Ann Smith |
Measuring flow rates in pipes is of importance to a number of industries including water treatment, and carbon capture and storage. This project will develop mathematical models to improve the accuracy of pipe flow measurement, focussing on techniques inferring flow rates from pressure measurement and ultrasonic flow metres. The project will focus on: (1) Modelling flow regimes in which measurements are challenging, (2) Improving algorithms for flow measurement.
The accurate measurement of pipe flow is of importance to traditional industries such as water treatment and oil and gas transport. It is also of critical to modern applications such as carbon capture and storage and enhanced oil recovery. Examples of flow regimes in which measurements are challenging include two-phase flow such as wet gas flow, non-Newtonian fluids such as drilling mud or flow at the transition between laminar and turbulent flow. In this project one or at most two of these regimes will be chosen for detailed investigation. Collaboration is anticipated with the company Friction Flow Measurement. The supervision team and Friction Flow Measurement have collaborated on a number of joint projects funded by the Analysis for Innovators Programme.
A mathematical model will be formulated to describe pipe flow under these conditions as a system of partial differential equations. These will be based on fundamental equations of mass, momentum and energy balances supplemented by constitutive laws as needed e.g. to describe a non-Newtonian rheology. These equations will be analysed using nondimensionalisation and asymptotic methods with the aim of identifying negligible terms which can either be dropped from the models or included perturbatively developing reduced models. The reduced models will then be solved using numerical methods, for example the method of lines or possibly a finite element or finite volume method. The resulting simulation will predict the flow state within the pipe including the velocity and pressure fields.
The results of these simulations will be used to aid the interpretation of experimental data and thus to develop improved algorithms for predicting flow rates. Areas of investigation may include:
This project will require a researcher with some familiarity with the formulation, analysis and numerical solution of mathematical models consisting of systems of partial differential equations. It would be suitable for candidates with a background in applied mathematics, engineering or the physical sciences.
Project Code |
EPSRC_2024_26 |
Supervisory Team |
Main Supervisor: Professor Jane Jiang Co-Supervisor: Dr Wenhan Zeng, Dr Wenbin Zhong, Professor Liam Blunt |
Hybrid machining represents a significant leap forward in manufacturing technology, offering unparalleled flexibility and efficiency. This research will contribute to the advancement of hybrid machining technologies, providing valuable insights for their implementation in various industrial sectors. Through experimental, theoretical, and applied research, this study aims to pave the way for next-generation manufacturing processes.
The manufacturing industry continuously seeks innovations to improve efficiency, precision, and the range of materials that can be effectively machined. Hybrid machining, combining additive and subtractive manufacturing processes, offers a promising avenue to meet these demands. This research aims to investigate the potentials and optimizations of hybrid machining techniques, focusing on their applications in aerospace, automotive, and medical industries.
Objectives
Methodology
The research will employ a mixed-methods approach:
Expected Outcomes
The CPT has the world’s most advanced ultra precision machine and state-of-the-art metrology instrument that can support this project.
An ideal candidate for the proposed project should possess a recent undergraduate or master’s degree (1st class or distinction) in mechanical engineering, control engineering or electronic engineering. Experience in CAD/CAM and CNC programming are preferable.
Project Code |
EPSRC_2024_27 |
Supervisory Team |
Main Supervisor: Professor Zhijie Xu Co-Supervisor: Dr Wenhan Zeng. Dr Shan Lou, Professor Richard Hill |
This PhD project focuses on investigating digital twin technologies for modelling Additive Manufacturing (AM) processes and its asset management within a collaborative metaverse space. Metaverse is a technology that leverages Extended Reality (XR) to enable users to interact with virtual and/or physical systems within a persistent, shared, 3D virtual space. The project starts with workflow definition for AM digital twins. It then develops heterogeneous data integration techniques that ensure compatibility and seamless interaction among digital twin models. Standardised data structure will be defined for configuring and operating the industrial metaverse to bridge the virtual and physical systems and processes.
AM digital twins, crucial for efficient asset management and process simulation, can be categorised into component twins, system twins, and process twins. These categories encompass a variety of data sources, including those from electrical, mechanical, and optical engineering. Integrating these heterogeneous data sources into a collaborative and interactive digital environment is essential for AM productivity and efficiency. Through identifying and developing suitable metaverse hosting platforms and digital twin modelling tools, this project aims to enable collaborative workspaces that ensure data interoperability among different digital twin models and allow simulation and interaction on the interconnected AM assets.
The research involves four main objectives:
By harnessing the collective capabilities of XR, which encompasses Augmented Reality (AR), Virtual Reality (VR) and Mixed Reality (MR), metaverse facilitates multimodal interactions with digital objects, virtual systems, and other individuals. The project plans to develop a set of collaborative activities specific to AM simulation within the developed metaverse system. The sandbox will serve as a prototype to incorporate features like digital twin integration, immersive experiences, collaborative tools, and seamless data management, all tailored for effective AM asset management.
The expected outcome also includes a rich list of metaverse and digital twin development tools that supports case studies featuring lifelike graphics, authentic audio, and sensory feedback for AM operations. The evaluation criteria include measures on function fidelity, system scalability, operational and data security, and compatibility with digital twin standards, namely a few, ISO/IEC 23005 – Guidelines for the design, development, and deployment of Metaverse; IEEE 2888 – Standardises interfaces for the physical and virtual worlds, data formats and APIs; and IEEE P2048 – Terminology, definition, taxonomies, and sustainability of Metaverse.
An ideal candidate for the proposed project should possess a recent undergraduate or master’s degree (1st class or distinction) in computer science or electronic engineering. The candidate must have a strong foundation in computer science, general engineering, and software development. Competence in modelling tools like MATLAB, programming languages such as Python or C++, and experience in independently developing software products using 3rd-party libraries like OpenCV and game engines are expected.
Project Code |
EPSRC-2024-28 |
Supervisory Team |
Main Supervisor: Prof Pavlos Lazaridis Co-Supervisor: Dr Qasim Ahmed, Dr Evangelos Vassos |
Using optically transparent conductive materials like Indium Tin Oxide (ITO), Fluorine-doped Tin Oxide, Aluminum-doped Zinc Oxide (AZO), and/or Silver Nanowires (AgNWs) in combination with optically transparent dielectric materials like glass or quartz, this project aims to develop optically transparent antennas and metasurfaces for 5G/6G communication systems and the Internet of Things (IoT). These cutting-edge components can enhance the aesthetics and functionality of next-generation wireless networks by offering multiple functions, dependable connectivity, and cutting-edge materials and design strategies. Furthermore, studies can be conducted on applying transparent or semi-transparent tuning techniques like liquid crystals and vanadium dioxide (VO2).
Specifically intended to improve the integration and performance of 5G/6G communication systems and Internet of Things applications, this project suggests an investigation into optically transparent antennas and metasurfaces. Through the utilisation of the distinct characteristics of optically transparent conductive materials, such as glass or quartz, in combination with indium tin oxide (ITO)[1], [2], fluorine-doped tin oxide (FTO), aluminium-doped zinc oxide (AZO), and silver nanowires (AgNWs), we plan to create components that provide superior electromagnetic performance and visual transparency.
In order to address the difficulty of preserving signal integrity in visually delicate environments, integrating these materials will be optimised to achieve a balance between electrical conductivity, optical transparency, and electromagnetic functionality[3]. Additionally, the project will investigate the possibilities of novel tuning techniques, such as the use of liquid crystals and phase-change materials like vanadium dioxide (VO2), in order to dynamically modify the characteristics of the antennas and metasurfaces in response to communication requirements. This adaptability is vital because efficient and flexible communication channels are critical for meeting the dynamic demands of next-generation wireless networks.
Furthermore, by examining how transparent antennas and metasurfaces can be incorporated into commonplace objects and environments without sacrificing design or functionality, the project will look into how these technologies may affect the development and implementation of Internet of Things devices and systems.
The optically transparent antennas and metasurfaces will be designed using CST Microwave Studio and an in-house mathematical optimisation approach.
The outcomes will be published in prestigious Q1 high-impact journals such as IEEE Transactions on Antennas and Propagation (IF = 4.4), IEEE Access (IF = 3.7), etc. Also, as the specific project is directly related to commercial applications, we will file a patent based on the produced novel prototypes.
Early research results.
[1] S. Chalkidis, E. Vassos et al, “Design of Unit Cells for Intelligent Reflection Surfaces Based on Transparent Materials,” 2021 10th International Conference on Modern Circuits and Systems Technologies, MOCAST 2021, pp. 8–11, 2021, doi: 10.1109/MOCAST52088.2021.9493335.
[2] E. Vassos, P. I. Theoharis, S. Chalkidis, F. Tubbal, R. Raad, and A. Feresidis, “A Comparative Study of a Reflectarray Antenna Based on Optical Transparent Materials,” in 2023 12th International Conference on Modern Circuits and Systems Technologies, MOCAST 2023 - Proceedings, 2023. doi: 10.1109/MOCAST57943.2023.10176544.
[3] S. Chalkidis, E. Vassos, and A. Feresidis, “Polarization independent dual function metasurface using transparent materials,” 2022 3rd URSI Atlantic and Asia Pacific Radio Science Meeting, AT-AP-RASC 2022, no. June, pp. 2022–2024, 2022, doi: 10.23919/AT-AP-RASC54737.2022.9814286.
Project Code |
EPSRC_2024_29 |
Supervisory Team |
Main Supervisor: Dr Paul Bills Co-Supervisor: Professor Liam Blunt |
Patients that have suffered from bone metastases are left with large defects that need to be accounted for and replaced or scaffolded as part of any surgical implantation process. In such cases an implant needs to be manufactured that has a patient specific geometry aligned with the removed defective tissue so the design and manufacture is a time critical process to ensure that the defect is effectively replaced. In addition such patients are uniquely vulnerable to infection so it is critical that any designed fixation surface must be resistant to biofilm formation.
The use of generative design tools that optimise component topology based on function is rightly seen as the key enabling tool to develop customised components that are tailored to their individual demands and environment. The manufacture of complex functionally structured/meta surfaces using additive manufacturing technologies has opened up the possibility of implants functionalised for osseo-integration properties along with optimised mass distribution to better replace the original tissue. In the case of a patient with metastatic bone loss the design, manufacturing and clinical challenges are increased. The functionality of topologically optimised surfaces must be assessed against the potential for biofilm formation and consequent infection in such immuno-compromised patients.
Furthermore the design, which is based on the size of the bone defect to be replaced, to implantation pipeline of such implants further needs to be optimised due to the time critical nature of the need for surgery. The definition of this pipeline and integration of the elements therein will form the key components of this project
1st class honours degree or Masters degree in Mechanical Engineering, Medical/Biomedical Engineering, Physics or similar
Demonstrable aptitude or knowledge of additive manufacturing
Demonstrable aptitude in design
Project Code |
EPSRC_2024_30 |
Supervisory Team |
Main Supervisor: Dr Haiyan (Helen) Miao Co-Supervisor: Professor John Allport |
Gas turbines have been used widely in power generation. To secure future power needs, it is important to find renewable fuels for high performance gas turbines with ultra-low emissions. The effects of fuel constitute, combustor design and operation conditions on gas turbine performance and emissions will be investigated both experimentally and numerically in this project, which create key knowledge for the development of next generation net zero gas turbine. This project is in collaboration with a gas turbine manufacturer in UK and contributes to the efforts of developing renewable energy innovative hub at Yorkshire.
.
Micro gas turbines are smaller-scale versions of gas turbines that are used in both power generation and aviation. They are compact in size with much higher energy density than battery and fuel cell and can run multi-fuels with little change of their design. Because of these advantages, the interest in micro gas turbine has been increased steadily, with many attentions focused on design optimisation or innovative combustor concepts to achieve high performance with low emissions.
On the other hand, searching for alternative fuels for combustion engines stems from high costs and unstable supply of petroleum fuel, global climate change and air quality that relates closely to public health. Among those alternative fuels, biofuels, green hydrogen, and synthetic fuels produced from biomass stand out due to their carbon neutrality which enables achieving net zero emission in power generation.
This project focuses on studying the effect of different renewable fuels (such as biofuels and fuels produced locally from biomass by thermochemical processes, such as Fischer-Tropsch (F-T) synthesis and pyrolysis), on gas turbine performance and exhaust products. The effect of the changing fuel constitute on combustion process inside the combustor shall be investigated systematically to understand how it affects gas turbine performance and emissions. The effects of combustor design and operation conditions will also be investigated. The understanding and knowledge created in this project is vital for developing next generation high performance net-zero gas turbines.
During the PhD study, you will develop a computational fluid dynamics (CFD) model for gas turbine when fuelling with different fuels. Huddersfield University has established partnership with an UK gas turbine manufacturer. As such, you will have the opportunity to work with gas turbines provided by our partner and then use experimental results to validate the CFD model. The validated model will be applied not only to optimise the design of gas turbines but also to propose new combustion concepts that leads to innovative design, which would bring in further collaboration with industry partners securing funding from EPSRC or Innovative UK.
The project is multidisciplinary in nature, involving testing, signal processing, chemical kinetic analysis, and CFD modelling. Specific objectives and timeline of the project are:
Essential
Desirable
Project Code |
EPSRC_2024_31 |
Supervisory Team |
Main Supervisor: Dr Haiyan (Helen) Miao Co-Supervisor: Professor Wenxian Yang, Professor Andrew Ball |
To meet UK’s net-zero transport target, it is important to decarbonise hard-to-electrify propulsion systems for heavy-duty usages such as in railway, marine, and off-road applications. Renewable fuels (such as hydrogen and biofuel) are carbon neutral and therefore, provide an attractive solution for net-zero high-performance heavy-duty engines with sustainable resources. This project aims at investigating the effects of fuel constitute, engine design, and operation conditions on engine performance and emissions, which create key knowledge for the development of next generation net-zero engine for transport. It contributes to the efforts of developing renewable energy innovative hub at Yorkshire.
Renewable fuels (such as hydrogen and biofuels) provide a promising potential solution for decarbonisation of hard-to-electrify propulsion systems, i.e. heavy-duty combustion engines for railway, marine, and off-road applications. This is because they are carbon neutral and have energy density that is much higher than battery and fuel cells. However, those fuels have not found their way to be widely used in transportation yet. For example, the previous studies on hydrogen show two major limitations for pure hydrogen engines, which this proposed project aims to address: (1) misfire at idle or low engine speeds and (2) super knock at high loan. On the other hand, the heating values of biofuels vary depending on their sources as well as conversion methods. For biogas, its methane content typically ranges from 40% to 75% by volume, making it very difficult to be used directly in transport propulsion systems.
This PhD project aims to (1) create new knowledge about the effects of fuel constitute on internal combustion engine performance and emissions, especially at abnormal operating conditions such as misfire and knock, (2) establish numerical models for different renewable fuels, and (3) develop an engine design tool for optimising both engine design and operation conditions for high performance and low emissions heavy duty engines using renewable fuels. The project is multidisciplinary in nature, involving engine testing, signal processing, chemical kinetic analysis, and computational fluid dynamic modelling. Specific objectives and timeline of the project are:
This project has good potential to create significant impacts on knowledge, society and the economy. Knowing how the mechanisms of fuel constitute on engine performance, emissions and abnormal operations would help engine manufacturers, rail and marine operators, and government (especially Department of Transport) achieve UK’s net-zero target. The intelligent usage of renewable fuels could potentially create patentable intellectual properties and be commercialised as a product.
Essential
Desirable
Project Code |
EPSRC_2024_32 |
Supervisory Team |
Main Supervisor: Professor Parikshit Goswami Co-Supervisor: Professor Liz Towns-Andrews |
The project aims to develop bio-inspired sustainable bio-composites for garment applications. This project will focus on manufacturing fibre-reinforced bio composites using recycled fibres and optimising the shape and mechanical properly of the composites. Work will be done to fundamentally understand and improve the fibre/matrix interface. The developed bio-composites will be thoroughly characterised and optimised. These novel bio-composites can replace non-renewable materials in different garment applications, reducing environmental impacts and enhancing sustainability.
Nature produces numerous hierarchical composites (e.g. wood which combines hierarchically different micro and nano-scale components with different structures and functions) with excellent mechanical performance and diverse application potential. Design and development of such materials using waste products and renewable resources would be an important step towards a sustainable and greener world and can greatly replace non-renewable products in different apparel applications. The overarching aim of this project is to develop a framework to reduced environmental impacts and improved sustainability of the functional apparel sector.
The project has the following objectives/work packages:
Qualification:
First Class or Upper Second-Class Honours degree or equivalent in a textile and fashion-related discipline or material science.
Key Skills:
Textile Manufacturing; Garment construction; Textile characterisation
Project Code |
EPSRC_2024_33 |
Supervisory Team |
Main Supervisor: Professor Rakesh Mishra Co-Supervisor: Dr Shirsendu Sikdar |
The proposed project aims to develop a cutting-edge cyber-physical system for intelligent monitoring of pressure valves, utilizing Acoustic Emission, Ultrasonic Guided Wave Propagation, and Vibration analysis. It integrates digital twins through rigorous Multiphysics simulations, physics/mechanics-informed machine learning, and smart sensing systems. Aligned with EPSRC's priorities in engineering and energy decarbonization, the research ensures emissions-free operations in Hydrogen-energy and Oil & Gas sectors. Focused on real-time early-stage detection and predictive maintenance, this work contributes to the UK's energy sustainability goals, enhancing the reliability and environmental compatibility of critical infrastructure.
The proposed project aims to pioneer a state-of-the-art cyber-physical system for smart monitoring of pressure valves, employing AE/Ultrasonic-Guided-Wave(GW), and Vibration analysis. This initiative ensures emissions-free operations in Hydrogen-energy and O&G industries. Spearheaded by Prof. Rakesh Mishra, an expert in fluid mechanics and digital twining, and Dr Shirsendu Sikdar, a specialist in smart monitoring and prognostics, the research strongly aligns with EPSRC's priorities in engineering and energy decarbonization. Emphasizing real-time early-stage detection of valve degradation and predictive maintenance, this work actively contributes to the UK's energy sustainability objectives while advancing the reliability and environmental compatibility of critical infrastructure.
The project objectives include:
Objective-1. Development of a Comprehensive Monitoring System: Design and implement a sophisticated cyber-physical system integrating AE, GW Propagation, and Vibration analysis technologies for real-time monitoring of pressure valves. This system will incorporate digital twin models based on rigorous Multiphysics simulations to accurately capture valve behavior under various operating conditions, including variable loading and environmental factors.
Objective-2. Advanced Data Analytics and Machine Learning: Employ advanced data analytics techniques, including physics/mechanics-informed and data-driven machine learning algorithms, to analyze sensor data and extract valuable insights regarding valve health and performance. Develop intelligent prognostic models capable of predicting potential valve degradation and failures under variable loading and operating conditions, enabling proactive maintenance interventions.
Objective-3. Optimization of Maintenance Strategies: Utilize the information provided by the monitoring system and predictive models to optimize maintenance strategies for pressure valves. Implement proactive maintenance measures based on early-detection of degradation, minimizing downtime, and extending the operational lifespan of valves. Tailor maintenance schedules and interventions to specific environmental and operational factors, such as temperature, pressure, and vibrations, to ensure optimal performance and reliability.
Objective-4. Validation and Demonstration in Field Environments: To validate and demonstrate the effectiveness of the developed cyber-physical system in field environments. This objective involves deploying prototype monitoring systems in operational hydrogen energy and O&G facilities to assess performance under real-world conditions. Through comprehensive field testing and validation exercises, the project will demonstrate the scalability, reliability, and practical utility of the proposed monitoring solutions, including their ability to intelligently adapt to variable loading and operating conditions.
Through the achievement of these objectives, the project aims to enhance the efficiency, reliability, and safety of pressure valve systems in Hydrogen-energy and O&G industries. The research will contribute to the UK's energy sustainability goals and facilitates the transition towards a low-carbon future.
Qualifications:
A minimum of a Bachelor's degree in Mechanical/Civil/Electrical Engineering or a related field.
A master’s degree in a relevant discipline is preferred.
Skills, Experience, and Knowledge:
Knowledge in programming languages such as Python, MATLAB, or C/C++.
Strong analytical and problem-solving skills.
Experience with research or industry projects related to monitoring and prognostics.
Familiarity with Acoustic Emission, Ultrasonic Guided Wave Propagation, or Vibration analysis techniques.
Knowledge of machine learning algorithms and their application to predictive maintenance.
Understanding of engineering principles, particularly in fluid mechanics and mechanics of materials.
Familiarity with energy systems, especially in Hydrogen-energy and Oil & Gas industries.
Knowledge of relevant regulatory standards and safety protocols in industrial settings.
Candidates should demonstrate a strong commitment to research excellence, possess excellent communication skills, and be able to work effectively both independently and as part of a multidisciplinary team.
Project Code |
EPSRC_2022_34 |
Supervisory Team |
Main Supervisor: Professor Rakesh Sharma Co-Supervisor: Dr Frankie Jackson |
The proposed project aims to develop a cutting-edge digital twin system for intelligent monitoring of polishing process. Polishing is an essential operation in the manufacturing process of many engineering components. These components include those made of glass, metals, alloys, and fibre reinforced polymeric composites etc. Coupling macroscopic CFD with MD along with sensing and cyber-physical system allows for an efficient parallel hybrid system, where the strengths of each approach can be utilised. For this research, conventional CFD techniques would be used to solve the macroscale length scale phenomena, while MD would be used to solve microscale interactions.
The proposed project aims to pioneer a state-of-the-art digital twin system for smart monitoring of mechanical polishing process. In order to bring about a step change in the development of new tools for the industrial polishing processes, a micro-, meso- and macro-scale understanding of the flow mechanisms and solid-liquid-surface interaction are needed.. The project objectives include:
Through the achievement of these objectives, the project aims to bring about a step change in mechanical polishing process. This will also enhance the efficiency and reliability of polishing operation.
Qualifications:
Skills, Experience, and Knowledge:
Candidates should demonstrate a strong commitment to research excellence, possess excellent communication skills, and be able to work effectively both independently and as part of a multidisciplinary team.
Project Code |
EPSRC_2024_35 |
Supervisory Team |
Main Supervisor: Professor Alan Smith Co-Supervisor: Dr Anke Bruning-Richardson, Dr Muhammad Ghori, Dr Jessica Senior |
Glioblastoma is one of the most aggressive cancer types and is part of a group identified by Cancer Research UK as having an “unmet clinical need”. The only existing medication to prevent tumour recurrence after surgical removal are crude wafers containing a cytotoxic drug, which is placed randomly in the brain’s resection cavity. This method, developed in the early 1980s, is generally ineffective in preventing long-term recurrence. Therefore, there is an urgent need for new treatment options to replace this outdated technology. This project aims to design innovative intercranial drug delivery systems to combat the recurrence of glioblastoma post-surgery.
Glioblastomas can proliferate rapidly within the brain, a characteristic attributed to their swift growth rate and migratory nature. This swift and widespread growth renders glioblastoma especially difficult to manage. It is common for these tumours to recur and lead to patient mortality, even after an initial combined treatment approach involving surgery, chemotherapy, and radiotherapy. Following surgical removal of glioblastomas, a drug delivery system known as Gliadel wafers is often used to try to prevent recurrence of the tumour. These wafers, which were first developed in the early 1980s contain the cytotoxic drug carmustine and are currently the only available implantable treatment for newly diagnosed or recurrent glioblastoma. Gliadel wafers are inserted directly into the resected tumour cavity of the following surgical removal, bypassing the blood-brain barrier, allowing for higher drug concentrations of carmustine near any remaining tumour cells. The solid wafers then undergo a two-stage degradation process that releases the drug, with water penetration hydrolysing the polymer in the first 10 hours, followed by the erosion of the wafer into the surrounding aqueous environment. However, post-implantation, several side effects have been reported. These include seizures, intracranial hypertension, meningitis, cerebral oedemas, and compromised wound healing. Some of these complications are thought to be linked to the properties of the implant itself. It is thought that the rigid nature of the implant leads to micro-tears when it moves post-implantation leading to oedema. Furthermore, the drug release from the implant is often not continuous and is dependent on intracranial conditions. Moreover, recurrence often occurs at a secondary site within the brain due to the aggressive migratory nature of glioblastoma cells which is not addressed by the Gliadel wafers.
The aim of this project therefore, is to engineer a safer intercranial drug delivery system by developing an injectable hydrogel to facilitate ease of application for the surgeon and that solidifies and remains in place when applied into the resection cavity. The solidified hydrogel will also have mechanical properties matched to that of brain tissue to overcome issues associated with the rigid nature of Gliedel wafers. The hydrogel will be designed to contains an optimised combination of anti-migratory and cytotoxic drugs for their targeted and time-controlled release into the post-surgical wound to prevent tumour cell migration and enhance cytotoxic effect. This multidisciplinary project will integrate pharmaceutical science, material science and biology and will require investigations into hydrogel forming biomaterials, microspheres, nanoparticles, drug entrapment/release studies and in vitro testing on patient derived cells. Techniques used in this project will include 2D and 3D cell culture, rheology, drug release testing, particle sizing and zeta potential, microscopy, drug encapsulation and release studies.
Successful applicants will have a very good first or upper second degree or Master’s degree in a relevant subject (pharmaceutical Science/chemistry, biochemistry, medical biology, pharmacy, pharmacology, materials science, bio/chemical engineering, chemistry/physical science, analytical science or related discipline).
Project Code |
EPSRC_2024_36 |
Supervisory Team |
Main Supervisor: Dr Hazel Girvan Co-Supervisor: Dr Kirsty McLean, Dr Paul Elliott |
The flavour and scent industry was estimated to have a global value of 35 billion euros in 2019, manufacture of these chemicals by enzymatic methods is of great interest for commercial application. Enzymes present the ability to carry out chemically difficult reactions in an aqueous environment adhering to the 12 principles of green chemistry. This project will push beyond the tools available to Nature engineering enzymes unseen in the natural world.
Enzymes present an exquisite tool for the manufacture of chemicals, they are fine tuned to catalyse reactions in a stereo and regio specific manner in a polar environment and at ambient temperatures which are often difficult for classical chemistry approaches. Cytochrome P450 (CYPs or P450s) enzymes are utilised in several biomanufacturing processes including for production of statins, biofuels, and flavour and scent compounds, catalysing reactions including chemoselective oxidations, decarboxylations, dealykylations and racemisations. The enzyme CYP102A1 is one of the most widely exploited P450 enzymes for synthetic biology applications; having been utilised for industrial biotechnology processes including the manufacture of flavour and scent compounds such as the grapefruit flavour nootkatone and drugs including the anti-malarial artemisinin. In the native enzyme CYP102A1 contains a heme group with an iron containing tetrapyrrole axially coordinated through a cysteine thiolate. This thiolate coordinated heme utilises a catalytic intermediate which allows it to insert an atom of oxygen from molecular oxygen into a C-H bond. Studies of heme proteins have shown that the central iron can be replaced by alternative transition metals yielding variants with altered catalytic capabilities. Preliminary research in our group has demonstrated that a variant of CYP102A1 can be produced where the central iron is replaced with cobalt. This project will build on these findings; establish the catalytic power of the cobalt variant enzyme before exploring further novel chemistry with alternative metals allowing us to harness the power of enzymes for new applications.
This project will allow the successful applicant to gain experience in molecular-biology approaches; recombinant protein overexpression and purification; analytical chemistry techniques including LC-MS, GC-MS and ICP-MS; spectroscopic techniques including UV-visible, fluorescence and NMR spectroscopy and X-ray crystallography.
This project would suit students with a degree in either biochemistry, chemistry, biophysics or similar at 2.i or first class level. Students with an interest in protein biochemistry, biotechnology, analytical biochemistry, structural biology or biophysics are invited to apply
Project Code |
EPSRC_2024_37 |
Supervisory Team |
Main Supervisor: Dr George Bargiannis Co-Supervisor: Dr Emmanuel Papadakis
|
Recent successes of Artificial Intelligence (AI) technologies have shown their limitless potential in many areas, including supply chains. However, their widespread adoption within supply chains is hindered by limited explainability of developed solutions, as well as an overall “fear of AI”. This project will directly tackle these issues by developing trustworthy, ethical and responsible AI approaches, frameworks and tools to identify, measure and recommend factors affecting resilience and support supply chain stakeholders in making bespoke and strategic decisions to improve the ability of supply chains to resist disruptions or swiftly return back to normal after disruptions.
Supply chains in the UK and abroad have significantly been impacted by events such as the 2023 Turkey-Syria earthquake, the 2022 Pakistan floods, the 2022 Russian invasion of Ukraine and the ongoing Covid-19 pandemic. These have repeatedly exposed the limited capabilities of supply chains, especially global ones, in managing risks and recovering from disruptions such as extreme supply/demand deviations, increased costs and staff unavailability.
The overall aim of this project will be to assist supply chains and stakeholders in detecting potential risks, recovering from disruptive events and monitoring supply chain “health”. Previous work in this area has primarily focused on methodologies that rely on purely “black-box” solutions which, while being performant, are quite difficult to be adopted by stakeholders and integrated within supply chain operations. To increase likelihood of adoption, the focus of this project will instead be on maximising interpretability and transparency of outputs, as well as trust and understanding of the developed intelligent technologies. The project will also seek to clarify to relevant SC stakeholders any misconceptions around AI technologies, in terms of their requirements and capabilities.
To achieve these aims, the following objectives are identified:
The envisaged impact of the project includes:
Master’s degree (merit or distinction) or an honours degree (2:1 or above) or equivalent in Computer Science, Artificial Intelligence or a related discipline
Excellent programming skills in a range of programming languages, both declarative and imperative
Practical experience in artificial intelligence projects that have involved one or more of the following areas: data analytics, machine learning, knowledge representation, automated reasoning.
(Desirable) An understanding of concepts related to supply chains, risk management, resilience and sustainability
Project Code |
EPSRC_2024_38 |
Supervisory Team |
Main Supervisor: Dr Hamidreza Faham Co-Supervisor: Dr Pritesh Mistry |
Using the IRR’s state-of-the-art industrial robot facilities, the aim of this project is to integrate a HEX sensor, which has 6 degrees of freedom to register force/torque readings as the robot performs operations. This integration will involve developing physical mechanical fittings, coupled with appropriate electrical wiring and a programming interface enabling the end user to use the data generated and provide feedback sensory information to the robot. This may be in the form of statistical inference or machine learning guided decision making. The overall objective of the project is to advance the capabilities of the KUKA by incorporating HEX sensors, renowned for their precision and sensitivity.
This research project will involve the meticulous process of integrating HEX sensors into the existing operational framework of our KUKA robots. Appropriate physical fittings will have to be designed, developed, and integrated into the robotic architecture, interfacing with the programmable software used to control the robots.
A subsequent phase will focus on using these new robotic capabilities to perform pressure sensitive tasks such as cleaning and/or polishing delicate surfaces such as glass. The emphasis is on the precise control necessary of the robotic arm for it to navigate and perform desirable operations on varying surface conditions. A proposed additional step would be to combine this with computer vision to allow the robot to perform semi-automated/automated tasks.
The anticipated outcomes of this research proposal include:
These novel contributions of this research study will illustrate the practical advantages of integrating such sensors to enhance the adaptability and precision of the robotic system. A capability which these robots do not currently have.
To disseminate the developed technology to the wider community, the appointed researcher will identify industrial applications where such capabilities are lacking and use their research to provide practical solutions to solve some of these problems.
This includes adapting the robotic control system to accommodate HEX sensor data and establishing communication protocols for real-time feedback. The subsequent phase will focus on developing an application that leverages force feedback for surface cleaning or polishing, emphasizing the precise control and adaptability of the robotic system in response to varying surface conditions.
The anticipated outcomes of this research include a successful integration of HEX sensors with KUKA robots, establishing a novel framework for force-sensitive feedback. The demonstrated application in surface cleaning or polishing will illustrate the practical advantages of this integration, showcasing the adaptability and precision of the robotic system.
The ideal candidate will come from an engineering/electrical engineering or computing background, with experience of robotic automation and control. Knowledge of a programming language is essential (e.g. MATLAB, Python). Working knowledge of programmable logic controllers (PLC) would be desirable.
Project Code |
EPSRC_2024_39 |
Supervisory Team |
Main Supervisor: Dr Huidong Wei Co-Supervisor: Dr Shan Lou, Dr Leigh Fleming |
Polymer-based clear aligners become the new generation of teeth straightening orthodontic devices, bringing an enhanced user experience for its high invisibility compared to metallic or ceramic braces. The teeth alignment relies on the biomechanical forces exerted from the aligner product whilst the stress decaying behaviour of polymer materials under oral environment is not fully understood. This research project will fill this knowledge gap by characterising and modelling the nonlinear visco-plasticity of biomedical polymers to gain deep insight into its biomechanical performance, leading to a controllable force exertion of clear aligners by optimising its design and manufacture with desirable material behaviour.
Compared to the traditional metallic or ceramic aligners with wires, the invisibility of the clear aligners from polymers attracts more users with expanding market. The complex mechanical behaviour of polymeric materials however brings a big challenge of predicting the biomechanical effect on teeth for its strong time and temperature dependence. This research will investigate the nonlinear visco-plastic mechanical properties of 3D-printed biomedical polymers, aiming for the manufacture of clear aligners in orthodontic applications. The primary tasks of this research will be (1) Characterise the mechanical behaviour of biomedical polymers under different loading conditions mimicking oral environment; (2) Develop a constitutive model to describe the nonlinear visco-plasticity of materials; (3) Investigate the influence of processing conditions on its mechanical performance; (4)Optimise the design and manufacture of clear aligners by finite element analysis.
This project is highlighted for its research on in vitro study of biomedical polymers under temperature and humidity-controlled conditions, which simulates the oral environment to investigate the potential biomechanical performance. The University of Huddersfield has advanced manufacture and metrology capabilities to support the related research activities. Through the Centre for Precision Technologies (CPT) and Precision Manufacture Hub, biopolymer specimens from 3D printing and environmental testing chamber by CNC machining will be prepared. The biomedical materials will be characterised on its macroscopic behaviour by employing tensile testing and surface profiling techniques. The microstructure of materials will be investigated by using modern imaging methods, such as scanning electron microscopy (SEM) and compute tomography (CT). The interdisciplinary nature of this project of combing engineering with material and dental sciences showcases a great novelty with integrated ways. Based on the novel output, the candidate will generate high-quality publications and potential patents with significant research impact.
The success of this project will contribute to a deep understanding of mechanical behaviour of biomedical polymers manufactured by 3D printing. A scientific expression on the nonlinear visco-plasticity of materials will be derived under variable manufacture conditions. The research output will aid in the development of next-generation clear aligners for tailored biomechanical forces on teeth straightening. This will lead to the design and manufacture optimisation for patient-specific device applications. The dental devices industry will be benefited to develop new innovative products with better user experience and market competitiveness.
Qualification: A First Class or Upper Second-Class Honours degree or equivalent in mechanical engineering, biomedical engineering, material science, physics, or a relevant subject area. A master’s degree in a related field is a plus, although it may not be mandatory.
Knowledge, skills, and experience:
Project Code |
EPSRC_2024_40 |
Supervisory Team |
Main Supervisor: Dr Mohammed Nasser Al-Mhiqani Co-Supervisor: Professor Simon Parkinson, Dr Saad Khan |
Rising cyber-physical systems (CPSs) threats demand enhanced cybersecurity measures, given increasing attacks. Estimated 30.7 billion operational tech devices by 2023. Operational technology security incidents are up 250% in 5 years, averaging $1.8 million per breach. CPSs vulnerabilities attract sophisticated attacks, necessitating robust defence strategies. Current cybersecurity tools like IDSs, firewalls, and encryption lack effectiveness against insider threats in CPSs. Integrated CPSs domains pose risks of significant damage from insider manipulation.
The complexity of insider threats poses a significant security issue within CPSs, as they often evade detection and prevention measures commonly in place. Furthermore, insiders will be able to cause significant damage by manipulating tiny adjustments to manufacturing parameters in a fully integrated system structure of CPSs' digital and physical domains. Therefore, before CPSs fully manifest in the real world, insider threats need to be addressed and managed. However, important gaps and challenges remain to ensure the security of CPSs: Despite advanced research on insider threats, challenges in validating and refining the detection models remain due to the absence of real-world data from organizations. Despite the increase in the number of insider threat incidents, not all organizations report such incidents nor allow access to their data, typically due to ethical and privacy concerns. The issue of real data access is crucial for insider detection, which continues to be a significant obstacle to validating and refining effective and scalable detection systems. An additional limitation in current insider threat detection approaches is their narrow focus on either technical or social security aspects within the field of CPS. A significant of prior research has primarily conducted to detect insider threats by monitoring technical-related behaviours. Nevertheless, it is essential to recognize the significance of social behaviours in this context. For instance, the human factor can often constitute the weakest link in cybersecurity, particularly in OT environments. Insider attackers within OT settings possess intimate knowledge and expertise in critical system technology and software, along with a deep understanding of vulnerabilities and potential security weaknesses. These aspects require novel solutions, to be developed, and integrated in novel ways, to combine technical behaviour with the social behaviours of the CPSs. Additionally, exploring new methods, including privacy-preserving techniques, is essential to overcome the data access challenge.
The proposed project aims to explore and develop novel solutions for insider threat detection in CPSs. It will leverage artificial intelligence and privacy-preserving techniques, along with socio-technical approaches, to integrate socio-technical privacy-preserving insider threat detection.
Project Code |
EPSRC_2024_41 |
Supervisory Team |
Main Supervisor: Dr Quratul-ain Mahesar Co-Supervisor: Professor Andrew Crampton, Professor Mauro Vallati |
Argument mining is a research area within natural language processing. The aim of argument mining is the automatic identification and extraction of argumentative structure from real world textual resources such as legal documents. Argument mining can play a vital role in improving the overall effectiveness of legal decision-making processes by streamlining information extraction, enhancing case analysis, promoting consistency, and contributing to the efficiency and transparency of the legal system.
This project will involve a detailed research study of argument mining for text summarization of legal documents in the context of legal decision-making which involves the extraction and analysis of arguments from legal texts, such as court cases, statutes, and legal opinions. The goal of the project is to identify the reasoning and evidence presented in legal documents to aid legal professionals in understanding, summarizing, and ultimately supporting them in decision-making processes. Legal professionals often face information overload, and argument mining can help filter out irrelevant or redundant information. This will allow lawyers and judges to focus on the key arguments and critical aspects of a case, leading to more informed decision-making. The underlying techniques will be argumentation theory, natural language processing (NLP), machine learning (ML) and large language models (LLMs).
The work plan for the project with specific objectives is given below:
Project Code |
EPSRC_2024_42 |
Supervisory Team |
Main Supervisor: Dr Saad Khan Co-Supervisor: Professor Simon Parkinson |
The project aims to leverage Large Language Models (LLMs) for advanced analysis of Event logs, collected from one or more devices. Focussed on investigating an IT system targeted by cyberattacks, the project intends to revolutionise the way we perform security log management. Various analytical, machine learning, and statistical techniques exist, but they do not always provide precise and actionable insights for the detection, mitigation, and prevention of malicious activities, such as privilege escalation attacks, unauthorised logins, etc. LLMs offer a robust foundation that can be trained and fine-tuned to precisely align with the requirements and objectives of event log analysis
Event logs are the records of user, application, system, and network activities that take place within an organisation. As the system management has become log-centric, security professionals analyse event logs to identify and resolve suspicious/malicious activities, such as unauthorised access, account manipulation, anomalous network traffic, malicious processes, privilege escalation attacks, security policy violations, suspicious Registry modifications, etc. Most organisations have found that their systems generate more event log data than they can handle. This is why security professionals experience a heavy workload analysing large amounts of Event Logs. Even with the help of automated Security Incident and Event Management (SIEM) solutions, it is tedious to manage security, performance, and troubleshoot IT issues. SIEM solutions are also error-prone (an increasing number of security-critical events go unnoticed). It is time to integrate advanced AI/ML techniques to improve this process in terms of time and accuracy.
There are several challenges and gaps faced by the state-of-the-art methods, such as (1) lack of support for different event log formats, (2) unable to handle missing data, (3) unable to (cross) correlate activities from heterogeneous event log sources and endpoints, (4) lack of ease in search, debugging, comparison, and readability of event logs, and (5) incapable of providing a reasonable investigation plan or a security solution after identifying a suspicious activity. In recent years, large language models (LLMs) like GPT (Generative Pre-trained Transformer) and FLAN (Flexible Language Acquisition Network) have shown great promise in various domains; however, their potential is not yet exploited to improve the Event Log analysis process. This PhD research aims to develop a novel framework that leverages the capabilities of LLMs in addressing the challenges and gaps, along with reducing the need for manual intervention, speeding up processes, handling large volumes of data, and providing more meaningful and context-aware responses.
In this research, we will collect and pre-process relevant data, which will be used to train the LLMs and validate the effectiveness of the framework. Furthermore, empirical will be conducted to compare the performance and robustness of the developed LLM-based approach against existing manual and automated techniques commonly used Event log analysis. We will also seek collaboration with industry in applying the developed framework to real-world scenarios, enabling an assessment of its practical effectiveness.
Candidates should have a minimum of a UK first-class BSc or MSc degree in Computer Science or a closely related subject.
Project Code |
EPSRC_2024_43 |
Supervisory Team |
Main Supervisor: Dr Shirsendu Sikdar Co-Supervisor: Professor Rakesh Mishra |
This PhD research aims to pioneer real-time and autonomous condition monitoring systems for operational Hydrogen Energy components (pipelines/pressure vessels/valves). Integrating Acoustic Emission (AE), Vibration and ultrasonic sensing, the project targets rapid identification/characterisation of damage, including leakage/impact-damage/cracks/corrosion/localized-inhomogeneity, under variable loading and operating conditions. Through rigorous experimental, analytical, and computational efforts, including Finite Element Analysis (FEA) and Computational Fluid Dynamics (CFD) simulations, Multiphysics modeling, and Bayesian data assimilation, this interdisciplinary initiative seeks to develop intelligent condition monitoring technologies. Leveraging edge computing-based smart sensing technologies, the project aims to enhance safety, reduce maintenance costs, and redefine operational practices in the energy sectors
The project aims fast and efficient monitoring and maintenance of storage and transportation systems in hydrogen energy and Oil & Gas sectors through the development of a cyber-physical system. This system will utilize advanced sensing technologies and machine learning algorithms to enable real-time, proactive management of critical infrastructure, ensuring emissions-free operations and contributing to the transition towards a sustainable energy future. The project objectives include:
O1: Development of Multi-Sensor Fusion Techniques: To develop and implement advanced multi-sensor fusion techniques integrating Acoustic Emission (AE), Ultrasonic Guided Wave Propagation, and Vibration analysis. This objective seeks to enhance the accuracy and reliability of damage detection and characterization in storage tanks, pipelines, and valves, thereby enabling early identification of degradation and potential failures under variable loading and operating conditions.
O2: Establishment of Predictive Maintenance Framework: To establish a predictive maintenance framework leveraging machine learning algorithms and data analytics. This objective aims to predict the remaining useful life of components based on real-time sensor data, operational parameters, and historical maintenance records. By proactively identifying maintenance needs under varying operational conditions, this framework will minimize downtime, reduce repair costs, and optimize asset utilization.
O3: Development of Edge Computing Solutions: To develop edge computing solutions for real-time data processing and decision-making. This objective focuses on the design and implementation of edge devices capable of performing local data analysis and initiating timely responses, such as alert notifications or automatic adjustments to operational parameters. By reducing reliance on centralized data processing and adapting to changing operating conditions, edge computing solutions enhance system resilience and responsiveness.
O4: Validation and Demonstration in Field Environments: To validate and demonstrate the effectiveness of the developed cyber-physical system in field environments. This objective involves deploying prototype monitoring systems in operational hydrogen energy and Oil & Gas facilities to assess performance under real-world conditions. Through comprehensive field testing and validation exercises, the project will demonstrate the scalability, reliability, and practical utility of the proposed monitoring solutions, including their ability to intelligently adapt to variable loading and operating conditions.
By achieving these objectives, the project will advance the state-of-the-art in condition monitoring and predictive maintenance, enabling energy operators to optimize asset performance, minimize environmental impact, and ensure the long-term sustainability of critical infrastructure.
In this research, we will collect and pre-process relevant data, which will be used to train the LLMs and validate the effectiveness of the framework. Furthermore, empirical will be conducted to compare the performance and robustness of the developed LLM-based approach against existing manual and automated techniques commonly used Event log analysis. We will also seek collaboration with industry in applying the developed framework to real-world scenarios, enabling an assessment of its practical effectiveness.
Candidates for this PhD research opportunity should possess the following qualifications, expertise, knowledge, and skills:
Candidates with a passion for advancing research in intelligent monitoring and predictive maintenance, and who meet these requirements, are encouraged to apply.
Project Code |
EPSRC_2024_44 |
Supervisory Team |
Main Supervisor: Dr Soran Parsa Co-Supervisor: Dr George Bargiannis |
Advancements in robotics have spurred the need for autonomous manipulation in intricate, entangled environments. Traditional systems face limitations in perception, hindering dexterous manipulation. To overcome this, the proposed project will integrate deep learning with coupled RGBD and tactile sensory data. By leveraging deep learning algorithms, meaningful representations can be extracted from sensory inputs, enhancing the robot's understanding and interaction with its surroundings. Extensive experimentation will validate the approach in terms of enabling precise manipulation in complex environments. This research project paves the way for applications in warehouse automation, manufacturing, healthcare, and agriculture, ushering in a new era of robotic capabilities.
In recent years, the demand for autonomous manipulation in complex environments has surged, posing challenges due to the intricate nature of entangled objects and the limited perception of traditional robotic systems. To overcome this, the proposed project will explore the integration of deep learning with coupled RGBD and tactile sensory data to extract meaningful representations from sensory inputs, enabling effective interaction with surroundings and enhanced dexterity.
By combining RGBD and tactile data, the robot gains comprehensive perception, facilitating precise manipulation. The deep learning model will be trained on a large dataset to predict object properties and plan manipulation strategies. This research presents promising prospects for robotic manipulation in various sectors, leveraging the synergy between deep learning and multimodal sensory inputs.
Qualifications: Degree in robotics, computer science, engineering, or a related field.
Skills: Proficiency in programming languages (e.g., Python, C++), experience with deep learning frameworks (e.g., TensorFlow, PyTorch), and strong mathematical skills.
Experience: Previous research experience in robotics, deep learning, and sensory data processing. Familiarity with robotic manipulation systems and sensor integration is preferred.
Knowledge: Understanding of robotic manipulation principles, deep learning algorithms, RGBD sensing, and tactile sensing. Knowledge of literature review methodologies and experimental design.
Strong communication and collaboration skills to work effectively within a research team.
Motivation and commitment to conducting cutting-edge research in robotic manipulation and deep learning, with a keen interest in addressing challenges in complex environments.
Ability to independently conduct literature reviews, design experiments, develop deep learning models, and analyse experimental results.
Project Code |
EPSRC_2024_45 |
Supervisory Team |
Main Supervisor: Dr Stavros Richard Christopoulos Co-Supervisor: Dr George Bargiannis, Dr Emmanuel Papadakis |
Every year millions of people and infrastructure globally are affected by earthquakes. In recent years, significant progress towards earthquake prediction has been made. One of the most promising approaches is Natural Time analysis, which has been applied on several scientific fields (e.g. statistical physics, cardiology, finance, mechanical engineering). This project aims to integrate Natural Time analysis based prediction, within an AI-based early warning system. This system will provide real-time analysis of data and available relevant information to inform whether a region of the solid Earth crust approaches a critical state that can generate a major earthquake.
The proposed project will focus on applying intelligent spatiotemporal modelling and reasoning to create several grid layers with different grain boundaries of the Earth’s surface using publicly available datasets from the United States Geological Survey (USGS). These layers will facilitate the study of the evolution of the K1 globally providing information on which areas approach a critical point. Subsequently, when we have this information, we can focus on the examined area creating smaller and more precise grid layers to further improve the location of the epicentre and the time of the occurrence of the forthcoming major earthquake.
Results of Natural Time analysis, along with analysis of additional contextual information through approaches such as machine learning will form the basis for an intelligent system that will issue early-warning alerts, along with a detailed explanation that can be used to both protect citizens and infrastructure and inform further research on earthquake predictions.
Evaluation of results will be based on the Receiver Operation Characteristic (ROC) methodology which is suitable for rare events, integrating tools such as VISROC [5], which evaluates the significance of binary diagnostic and prognostic tools for a family of k-ellipses which are based on confidence ellipses and cover the whole ROC space.
This project can lead to further exploration of additional aspects of Natural Time (e.g. variability entropy) that have the potential of increasing the effectiveness of the proposed methodology and to transfer this knowledge to assist in the prediction of other rare events, such as hurricanes and floods.
Master’s degree (merit or distinction) or an honours degree (2:1 or above) or equivalent in Mathematics, Physics, Computer Science, Artificial Intelligence or a related discipline
Excellent programming skills, preferably in Python
Experience in projects that have involved artificial intelligence and related aspects, such as data analytics, machine learning, knowledge representation, and automated reasoning.
Project Code |
EPSRC_2024_46 |
Supervisory Team |
Main Supervisor: Dr Gareth Parkes Co-Supervisor: Dr Karl Walton, Dr Katie Addinall |
Extrusion based additive manufacturing (EB-AM) is a cost-effective method of producing complex designs, which are either not possible in subtractive machining techniques, or incur high levels of waste. EB-AM methods are limited by current polymers, resulting in material properties which do not match those needed in application. With the development of high-loaded metal/ceramic polymers, it is now feasible to manufacture application ready artefacts using EB-AM. The project will apply advanced thermal processing techniques to the polymer de-binding and subsequent metal/ceramic sintering stages. Products will be analysed using mechanical and image based evaluation, to optimise processing conditions.
Recent developments of high-loaded metal/ceramic EB-AM filaments, are providing smaller companies and research groups with the ability to manufacture complex and robust products cost-effectively. Unlike conventional polymer only filaments, metal/ceramic filaments require two critical thermal processing stages: firstly, to initially debind the polymer matrix and secondly to sinter the metal to produce the desired artefact. Typically, this thermal processing is completed in kilns at set temperatures, selected empirically, that can lead to non-optimised products.
After modification of the TMRU’s current EB-AM printer to enable the use of high metal/ceramic-loaded filaments and initial familiarisation, the project will have three principal research sub-components:
Additionally, the thermal processing aspect of the project would be the use of simultaneous thermal analysis-mass spectrometry to analyse polymer break down products evolved during debinding. The composition of these products is likely to vary depending on the embedded metal (due to catalytic effects) and may well have health and environmental implications.
The project is envisaged as proceeding iteratively with the results from the metrological measurements being used to inform further thermal processing experiments.
A suitable candidate should have (or be expected to soon obtain) a degree in a STEM subject including Forensic and Analytical Sciences, Chemical Sciences, Physical Sciences, Materials Sciences, and/or any other Engineering degrees or related subject areas.
Project Code |
EPSRC_2024_47 |
Supervisory Team |
Main Supervisor: Dr Esta Bostock Co-Supervisor: Dr Katie Addinall, Dr Gage Ashton |
The use of household chemicals in criminal assaults is a wide issue within the UK, due to availability and ease of purchase. While research has focussed on analysing damage with caustic materials unavailable for general purchase (e.g. laboratory grade research chemicals) and damage caused to skin, there is a gap in knowledge with regards to household chemicals and their effects on apparel. This project aims to design a categorisation method for apparel damage with a wide range of household-based chemicals. This will ultimately result in an objective criminal investigative tool.
Currently there is no one accepted method for the categorisation and correlation of damage incurred in apparel fabrics with chemical attacks. This research aims to close the knowledge gap and provide an objective correlation tool for the effective investigation of this mode of attack, which is a known and escalating issue within the UK, with cases being reported weekly.
The project will be split into the following objectives:
Fabric composition of clothing apparel can have a wide variety: with natural, synthetic, blended, recycled and coated fibres used as mixtures in clothing. The mode of clothing manufacture can also be split into woven and knitted, thus there are a variety of variables that need to be analysed and understood. The first objective of this project will be to use both analytical and microscopic techniques to create a holistic understanding of apparel composition.
With the varying apparel compositions, there needs to be investigation into how damage with household chemicals compares across apparel mixtures. Therefore, once objective 1 is complete there will then be analysis regarding how each apparel variable reacts with various household chemicals. This will not only focus on acidic chemicals, but also known bases such as bleach and sodium hydroxide and heated sugars, commonly used in prison assaults.
A holistic approach of analysis will be required to determine damage patterns from the fabric level to individual fibres, creating a wealth of information which can be used to bolster investigation of chemical attacks. Image analysis will be used to compare patterns in damage with apparel type and chemical variables. This will ultimately result in creation of an objective mathematical categorisation method to determine mode of attack in unknown criminal investigation, which can then be utilised in correlation methods for mathematical matching of damage patterns.
Data gathered throughout the project will be stored in a reference database to allow for development of digitalised physical assets for comparison, disseminated through publication and sharing with interested parties to allow for a collaborative network, with the aim to aid criminal investigation and reduce chemical attacks, thus having a positive impact on society through changes in policy.
A suitable candidate should have (or be expected to soon obtain) a degree in a STEM subject including Forensic and Analytical Sciences, Chemical Sciences, Physical Sciences, Materials Sciences, and/or any other Engineering degrees or related subject areas.
Project Code |
EPSRC_2024_48 |
Supervisory Team |
Main Supervisor: Dr Ahmed Tawfik Co-Supervisor: Dr Paul Bills |
Additive manufacturing, or 3D-printing, is a disruptive technology enabling the manufacturing of complex structures as well as new material combinations and tailored material microstructures. However, process parameter development to enable printing of new materials with the desired properties may be extremely costly, time- and material consuming. To speed up the development, as well as to achieve a better understanding of the relationship between process parameters and the resulting material microstructure and hence its macroscopic properties, this project will focus on the use of advanced analytical methods to understand the impact of process parameters on magnesium alloys melt pool stability in laser beam (PBF-LB).
3D printing of biodegradable alloys (Mg- and Zn-based) is challenging, due to evaporation and alloy instability. These alloys are promising biomaterials for bone-healing implants due to their ability to be replaced by bone tissue while degrading within the human body. However, for degradable alloys, there is limited knowledge of how to manufacture these materials using additive manufacturing. The work duties within this interdisciplinary research project include conducting high-quality research, in collaboration with other team members, to design and implement different methods to analyse existing and new experimental data derived from powder bed fusion with laser beam (PBF-LB) of biodegradable alloys.
The work involves, among other things, to choose, adapt and implement thermofluidic analysis models for melt pool to predict a printed material’s properties as well as to develop processing parameters for additive manufacturing of improved materials. The focus will be on the numerical simulation of the melt pool and investigating the impact of process parameters on melt pool stability and printed part structural integrity and surface finish. The work will give the candidate good understanding of Additive manufacturing process, specifically for medical devices.
1st in mechanical engineering, good understanding of CFD and FEA analysis. Good generative design skill, very good analytical skills.