Coventry University Featured PhD Programmes
University of Reading Featured PhD Programmes

Energy prediction in buildings using Artificial Intelligence

Department of Computer Science

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
Dr G Cosma No more applications being accepted Competition Funded PhD Project (European/UK Students Only)

About the Project

Application details:
Reference number: CO/GC-CDT-2/2020
Start date of studentship: 1 July 2020
Closing date of advert: 17 April 2020

Primary supervisor: Dr Georgina Cosma
Secondary supervisor: To be confirmed

Loughborough University
Loughborough University is a top-10 UK university, consistently ranked by the Guardian and other league tables. Founded in 1974, the Department of Computer Science has been ranked 4th in the UK for Computer Science (Guardian University Guide 2018) and has an excellent research track record in A.I, construction and Building Information Modelling.

Project Detail
Predicting building energy is important in building energy management and maintenance as it can be used for tasks such as evaluating building energy efficiency, conducting building commissioning and detecting and diagnosing building system faults.
This project involves the development of A.I. and other algorithms for predicting energy ratings of buildings.

The project has been defined by Cundall, who will provide the data for developing and testing the prediction models. Cundall is an international multi-disciplinary design consultancy, recognised as a leader in sustainability across the world. Offering a full range of integrated engineering services, Cundall has delivered many exemplar low-energy buildings and has built up an international sustainable design expertise that is second to none. The world’s first consultancy to be formally endorsed as a One Planet Company, Cundall’s work is guided and governed by its Sustainability Roadmap which frames its targets, principles and action plans as industry thought-leaders in responsible and sustainable design.

The project involves:
• Developing methods to identify features for predicting building efficiency ratings.
• Developing Deep Learning prediction models using large data.
• Applying the models to various case studies and scenarios provided by the company.

The Candidate
The ideal student will be a fast learner, self-driven, hard-working, keen to embrace the project, deliver it successfully, and should be comfortable working with deadlines. The student should be a strong programmer (using the Python programming language).

Experience in developing combinatorial optimisation, feature engineering and machine learning algorithms, and the ability to analyse, interpret and communicate results from experiments are essential. Experience in Building Information Modelling is highly desirable.

The PhD student will be based at the Department of Computer Science at Loughborough University.

The student will be supervised by Dr Cosma who has experience in data science and A.I., and by industry supervisors with expertise in Building Information Modelling. The student will have access to a range of training, research support and computing facilities including HPC, and high-spec machines suitable for deep learning tasks.

Entry requirements
Applicants should have, or expect to achieve, at least a 2:1 Honours degree (or equivalent) in Computer Science or a Computer Science subject. A relevant Master’s degree and/or experience in one or more of the following subjects will be an advantage: Deep Learning, Artificial Intelligence, Data Science, Building Information Modelling.

Contact details:
Name: Dr Georgina Cosma
Email address: [Email Address Removed]

How to apply
All applications should be made online at Under programme name, select Computer Science.

Please quote reference number: CO/GC-CDT-2/2020.

Funding Notes

This 3-year studentship will provide a tax-free stipend of £15,009 per year, plus tuition fees at the UK/EU rate. This PhD studentship is only available to UK/EU nationals.
Search Suggestions

Search Suggestions

Based on your current searches we recommend the following search filters.

FindAPhD. Copyright 2005-2021
All rights reserved.