Systems - Unlocking Domestic Demand-side Flexibility Using Artificial Intelligence

Number of awards:

1

Start date and duration:

1 October 2018 for 3 years.

Overview:

This PhD project will explore the use of machine learning to predict household appliance power profiles and usage patterns for smart grid applications.

Equiwatt has developed an Internet of Things (IoT) platform for management of the UK’s energy demand by optimizing household energy usage, by Demand Side Response (DSR). The platform runs cloud-based predictive algorithms that pre-plan for peak-power events by modulating the power usage across thousands of households. The successful candidate will explore machine learning to develop appliance models and energy grid conditions that empower this technology. The models will predict the power consumption of appliances, accounting for usage patterns and trends in customer behaviour; and predict grid stress events and residential DSR service deployment.

The project has a strong industrial focus that will drive a revolutionary grid-scale demand management technology from the laboratory to commercial viability. The successful candidate will join the Electrical Power Group at Newcastle University and have weekly academic supervision meetings. They will join a team of around 20 researchers to discuss current research topics, refine methods to tackle research challenges, present findings and share best practice for dissemination of knowledge. The successful candidate will attend international conferences and workshops at which they will present their work to their peers and meet other researchers. Industrial advisors from Equiwatt will provide mentorship and training on the industrial sides of the project, helping the successful candidate develop their commercial awareness and entrepreneurship skills.

Sponsor:

This project is part-funded by European Regional Development Fund (ERDF) (https://bit.ly/2MuNSHO) and Equiwatt Limited.

Name of supervisor(s)

Dr Neal Wade (https://bit.ly/2vFGnEi), Professor Phil Taylor (https://bit.ly/2KRNduN), Dr Johnson Fernandes.

Eligibility Criteria:

Only UK or EU applications will be considered.

Applicants should have a first-class degree, or a combination of qualifications and/or experience equivalent to that level. Ideally, students should have a BSc or MSc degree in engineering, computer science or a suitable quantitative field.

Applicants should be strong programmers, and experience in machine learning will be greatly valued Familiarity with the Python programming language and its ecosystem of tools and machine learning frameworks, will be welcomed.

How to apply:

You must apply through the University’s online postgraduate application system. To do this please ‘Create a new account’ (https://bit.ly/2nDS75G).

All relevant fields should be completed, but fields marked with a red asterisk must to be completed. The following information will help us to process your application. You will need to:
•insert the programme code 8060F in the programme of study section
•select ‘PhD Electrical and Electronic Engineering (full time) - Electrical and Electronic Engineering’ as the programme of study
•insert the studentship code ENG033 in the studentship/partnership reference field
•attach a covering letter and CV. The covering letter must state the title of the studentship, quote reference code ENG033 and state how your interests and experience relate to the project
•attach degree transcripts and certificates and, if English is not your first language, a copy of your English language qualifications.