A new government-backed R&D project involving Costain, TerOpta and the University of the West of England aims to harness AI and building sensors to help businesses optimise their energy efficiency

A system to determine the energy and cost efficiency of commercial buildings is being developed as a part of a government-backed research project involving the University of the West of England (UWE Bristol), engineering firm TerOpta and smart infrastructure specialist Costain.

Data gathered from a network of small building sensors will help experts to paint an accurate picture of energy consumption in a range of test sites with a view to eventually setting up a service that offers energy efficiency advice to businesses.

Called i-REAP, which stands for IoT-enabled Real-time Energy Analytics Platform, the two-year, £1.5m collaborative research and development project is being funded by the Department for Business, Energy & Industrial Strategy.

The aim of i-REAP is to develop a one-stop solution to measure, predict and optimise energy consumption in commercial buildings using the latest AI and Internet of Things-based hardware to create a “whole building approach” to driving down the cost of improving thermal efficiency.

Advertisement

It is being led by TerOpta, an engineering firm formed in 2010 with a team who had previously worked in central R&D for companies including Ericsson, Marconi, GEC and Plessey, giving it expertise in both optical telecommunications and embedded controls, as well as general engineering. TerOpta is developing IoT-enabled sensors for i-REAP.

Another project partner, Costain, one of the UK’s biggest smart infrastructure companies, is providing five test sites across the UK in buildings belonging to the firm or its subcontractors.

Researchers from UWE Bristol’s Big-DEAL (Big Data Enterprise and Artificial Intelligence Laboratory) will initially carry out a feasibility study in the buildings, assessing the heating layout, staff sitting arrangements, They will then install up to 80 IoT sensors inside and four outside each of the buildings. The sensors deployed inside will measure temperature, humidity and ambient light intensity, while the external sensors will gauge temperature, humidity, wind speed and solar radiance.

By collecting data over a six-month period, the researchers will be able to gather enough intelligence on each building to give the client advice on how the current systems are functioning and how they could be improved by retrofitting the premises, for example by making partitions double-glazed or upgrading heating, to make them more energy efficient – and cost-effective.

The case for i-REAP

The adoption of thermal efficiency measures has been slow – and the use of low carbon measures in existing buildings accounts for only 1.6% of total heat use, according to the UK’s Committee on Climate Change.

The main barriers to change are a lack of awareness about the benefits of greater energy efficiency and financial constraints. The i-REAP project aims to address this through its whole-building approach, providing an affordable solution to create a personalised and optimal energy consumption regime.

It will also contribute to “fast-forwarding” the adoption of AI and IoT solutions for energy savings, and help the building sector move from reactive approaches to predictive ones.

A key element of i-REAP is that it will leverage the characteristics of each individual building to optimise its heating and cooling operations, and will develop guidelines for ideal retrofitting actions and low carbon heating technologies.

This will be achieved through:

  1. Measure: i-REAP will use off-the-shelf IoT sensors to capture real-time changes in temperature, energy consumption and user occupancy. It will then interrogate the data acquired with historic and current weather data, heating and cooling equipment specifications, architectural designs and building fabric characteristics.
  2. Predict: Using the compiled data, i-REAP will leverage machine learning techniques (such as deep neural networks and long short-term memory neural networks) to predict future energy consumption patterns 24 hours a day, 365 days a year.
  3. Optimise: Novel evolutionary and deterministic optimisation models will use the predicted energy consumption patterns to optimise and control the operation of the heating and cooling equipment. For example, the energy demand for heating units could be dynamically adjusted based on weather changes and the number of users at any given time in the building.
  4. Ideal retrofitting and low carbon heating technologies: Once the i-REAP systems have been running for a while, it will have the necessary data to identify which elements of the building fabric are responsible for the greatest thermal losses. This data will be used to develop guidelines for the ideal retrofitting of the building. In addition, i-REAP will identify the most beneficial low carbon heating technologies given the specific context of an individual building.

This combination of IoT and AI will facilitate the development of truly predictive energy consumption models, prescriptive operation plans and personalised retrofitting and low carbon technology guidelines, which are not currently available.

The main outputs include:

  1. The i-REAP application is the interface in which the user can check the status of the IoT devices, view predicted energy demands, monitor and adjust heating and cooling operations, and track the achieved energy savings.
  2. An i-REAP personalised simulation audit platform will run “what-if” scenarios to determine which parts of the building fabric will improve thermal efficiency the most if retrofitted. It will also identify which low carbon heating technologies are the most effective. office structure, orientation of buildings and building facades, materials and insulation.

In the long term, the overall aim of i-REAP is to contribute to the UK’s 2050 net-zero strategy and provide enough information to feed into policy formulation for commercial buildings.

Professor Lukumon Ovedele is principal researcher at UWE Bristol on the project and assistant vice-chancellor and chair professor of enterprise and project management.

He said: “What makes i-REAP unique is that we are able to analyse energy efficiencies in different sections of the building, at various times of the day and ultimately we want to see how commercial buildings can contribute to carbon neutrality.”

Another potential impact of the scheme is that it could lead to businesses generating their own electricity but, rather than pushing it back to the grid, using it for their own purposes.

“Pushing self-generated electricity to the grid gives rise to a cost, which is ultimately passed on to other consumers through their electricity bills,” Professor Ovedele said.

“If buildings are more efficient, the organisations may be able to generate electricity that only they use, thereby avoiding the extra cost.”

 

Professor Lukumon Ovedele

Director

Big Data Enterprise & Artificial Intelligence Laboratory, UWE Bristol

Tel: +44 (0)11732 83443

L.Oyedele@uwe.ac.uk

www1.uwe.ac.uk/bl/research/big-deal.aspx

Twitter: @uwebristol

LEAVE A REPLY

Please enter your comment!
Please enter your name here