Process Induced Neural Networks for Environmental Data Analysis and AI for Data Adjustments

The University of Sheffield offers a PhD position in Process Induced Neural Networks for Environmental Data Analysis. This research focuses on developing AI techniques for analyzing environmental datasets and improving data accuracy. The role is part of the Centre for Doctoral Training in Green Industrial Futures, which includes training, international opportunities, and industry placements aimed at supporting UK industry in achieving net zero goals.
This advert will close once a suitable candidate has been found.

This PhD is one of a number of projects hosted by the Centre for Doctoral Training in Green Industrial Futures (CDT-GIF). We are offering pioneering research projects that will enable PhD researchers to explore key technologies and solutions that will support UK industry to reach net zero. https://greenindustrialfutures.site.hw.ac.uk/the-programme/training-programme/

Project: The field of environmental data analysis has seen significant advancements with the integration of Artificial Intelligence (AI) and Machine Learning (ML) techniques. This PhD research aims to explore the development and application of Process Induced Neural Networks (PINNs) for analysing complex environmental energy datasets and utilizing AI for data adjustments to enhance accuracy and efficiency.

Research Objectives:

Development of Process Induced Neural Networks (PINNs): The primary objective is to design and implement PINNs tailored for environmental data analysis. These neural networks will incorporate prior knowledge of underlying physical laws to generate physically consistent predictions, bridging the gap between classical and black-box AI models.

AI for Data Adjustments: The research will focus on leveraging AI techniques to adjust and refine environmental data, ensuring higher accuracy and reliability.

Application in Environmental and Energy monitoring: The PINNs and AI models will be applied to various environmental monitoring tasks, climate change impact assessment and energy used in the process lines. The goal is to improve decision-making processes and develop proactive strategies to address environmental challenges.

Lead supervisor/s: Prof. Mohamed Pourkashanian, Prof. Kevin Hughes, Dr Maria Fernanda, John Malpas, Magdalena Coventry and Andrew Willett

Project partner: Toyota Motor Europe NV/SA

CDT in Green Industrial Futures: The CDT is funded by the UK Engineering and Physical Sciences Research Council and is a partnership between Heriot Watt University, Imperial College London, and the Universities of Sheffield and Bath. The CDT is further supported by contributions from industrial partners. Bringing these leading universities together allows CDT-GIF students access to a wide range of academic expertise, resources and facilities.

The CDT-GIF has an exciting and challenging programme specifically designed for top performing junior researchers. Alongside the four-year research project, students will receive expert training and opportunities to contextualise their research within the wider net zero landscape, including:
• Residential taught courses at each of the partner universities in Years 1 and 2 that provide training in the systematic considerations for industry including: Life Cycle Analysis (LCA), technoeconomics, business models, policy & regulation, public engagement, plant operation.
• An international opportunity in Year 2 or 3 of the programme, including opportunities to visit a world-leading facility, conference or forum and explore the global context of industrial decarbonisation.
• A work placement with one of our industrial partners.
• A bespoke ‘net-zero leadership programme’, including regular exchanges with cohort members from the other universities, student-led activities, industry challenge sandpit, industrial site visits, and professional development opportunities.

Candidate requirements: As a minimum we require candidates to have a First-class or 2:1 MEng, MSc, or MA with merit (over 60%) in a relevant area i.e. Artificial Intelligence, Chemical Engineering, Process Engineering. Applicants who have a First-class BSc/BEng (Hons) and can demonstrate significant relevant industry/research experience may also be considered.

Candidates should be aware of and meet any additional entry requirements for the university hosting the PhD studentship.

Non-native English speakers must ensure they meet the English language requirements.

Enquiries and applications. Informal enquiries regarding the research project can be directed to the supervisor. Alternatively, if you have a question regarding the wider CDT-GIF programme, please contact cdtgreenindustrialfutures@hw.ac.uk .

Formal applications can be made here: https://greenindustrialfutures.site.hw.ac.uk/the-programme/how-to-apply/ . Applications are open until filled so we would encourage early application.

The initial application involves an application form (including a short personal statement and CV). You may select up to two CDT-GIF projects. Applicants will then be selected to interview. If successful at interview, candidates will be asked to complete the application process for the home university.

Funding notes. The programme is four years and starts in Oct 2025. Funding includes full UK fees, tax-free stipend (2025/2026 stipend is £20,780), plus budget for travel and consumables. The positions are in the first instance for UK students. There are limited spaces for overseas students.

Equality, Diversity and Inclusion. We warmly encourage applications from individuals of all backgrounds, experiences, and perspectives. Our programme values diversity as a cornerstone of innovation and collaboration, and we are committed to creating an inclusive environment where everyone feels respected, supported, and empowered to thrive. The CDT-GIF are committed to ensuring flexibility throughout our programme to support student’s needs and personal circumstances, for example those with medical conditions, caring responsibilities and other considerations. For example, we are open to exploring part-time options if appropriate for the nature of the research.