Machine Learning Approach to Identify Environmentally Friendly Alternatives to SF6 for UK Electricity Networks

Manchester is home to one of the largest electrical and electronic engineering departments in the country. Our High Voltage Laboratory makes us a place renowned for applied research, a place ranked in the top three UK universities for research impact. *

That’s why industry looks to us to tackle real challenges, and why we’re currently supporting partners like National Grid.

Now, we’re looking for innovators interested in working as part of a pioneering team of five PhDs and two PDRAs to help develop a grid of the future.

The use of SF₆ in modern electricity networks needs to be phased out due to its high global warming potential 23,500 times higher than CO₂. SF₆ has excellent dielectric strength and chemical stability, making it challenging to find a like-for-like substitute. There are tens of thousands of known gases, which is impractical and impossible to experimentally investigate all candidates and their combinations. In this project, a machine learning based approach will be developed to evaluate known gases for identification of alternatives to SF₆. Based on the pre-determined properties, their simulation and/or experimental data, machine learning will be able to explore suitable candidates matching the required properties. Gases will be characterised into suitable representations based on their various properties. Combinatorial and hyper-volume based multi-objective optimisation techniques will be investigated first for identifying various combinations of candidates that could meet the requirements of SF₆ properties and environmental constraints. Deep learning approaches such as autoencoder and generative adversarial networks will be developed for extracting the forward and inverse mapping between possible candidates and required properties. Subsequently, suitable gases can be further evaluated and tested in a controlled laboratory environment. This is a cross-disciplinary project for a student to apply a machine learning based approach in tackling an environmental challenge facing the power industry. The project will be jointly supervised by two academics with specialised research background in SF₆ replacement and machine learning. 

We also have two additional projects available as part of this cohort - 'Investigation of Greenhouse Gas Alternatives for Electrical Insulation' (Investigation of Greenhouse Gas Alternatives for Electrical Insulation at The University of Manchester on FindAPhD.com) and 'Development of condition monitoring techniques for SF₆ alternatives in high voltage plant' (Development of condition monitoring techniques for SF6 alternatives in high voltage plant at The University of Manchester on FindAPhD.com)

Start Date: 2022/23 academic year (1^st September 2022 earliest start date and 1^st July 2023 latest start date) 

Entry Requirements: The standard academic entry requirements for a Doctorate level programme is an Upper Second UK honours degree (or international equivalent) in a relevant science or engineering discipline or a first degree with an additional pass in a UK Masters degree (or international equivalent)

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact. We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status. 

We also support applications from those returning from a career break or other roles. We consider offering flexible study arrangements (including part-time: 50%, 60% or 80%, depending on the project/funder). 

All appointments are made on merit.