Intelligent Agents for Home Energy Management
Lead Research Organisation:
University of Southampton
Department Name: Electronics and Computer Science
Abstract
Meeting the challenge of cutting UK greenhouse gas emissions by 80% by 2050, and ensuring energy security in the face of dwindling oil and gas reserves, requires a radical change in the way energy (and particularly electricity) is generated, distributed and consumed . Central to delivering this change, is the need to support domestic consumers (who have the least visibility regarding their energy use, but who generate approximately 25% of total UK carbon emissions) in both reducing their demand for energy and improving the efficiency with which they use it. This proposal will do both by applying novel artificial intelligence approaches to the development of intelligent agents that will be transformational in empowering domestic consumers to visualise, understand and manage their energy use.These home energy management agents will collect real-time data from smart gas and electricity meters, and simple low cost temperature and occupancy sensor, and they will learn both the thermal characteristics of the building in which they are deployed and the day-to-day behaviour and energy demands of the home's occupants. In the short term, these agents will provide personalised support to householders by (i) visualising, analysing and comparing energy consumption (e.g. providing itemised energy use information, performing energy audits and comparisons across similar homes), by (ii) autonomously modelling and advising householders of the potential impact of various energy saving practices, and by (iii) tracking, providing feedback and motivating progress toward energy and carbon reduction goals. Such agents will go beyond the simple energy displays of today, and will act as persuasive technologies informed by a cognitive model of behaviour change. In the medium term, they will directly interface with network enabled appliances and will actively manage the delivery of heat and the deferral of electrical loads whilst making efficient use of shared and private variable renewable generation. In doing so, they will provide autonomous and intelligent demand management, whilst satisfying the individual householders' preferences regarding comfort, cost and carbon. Finally, in the long term, these agents will integrate with electric vehicles (EV) and plug-in hybrid vehicles (PHEV), giving the home's occupants visibility and control of their total energy and carbon use, proactively managing electricity storage within these vehicles, and facilitating the delivery of individual carbon budgets and allowances . In essence the project will enable occupants to make appropriately relevant behaviour decisions on their energy consumption and generation, relating the impacts of these decisions on carbon emissions. Beyond the immediate confines of the home, these agents will also have a profound impact at the macro level. They will be developed with a future outlook to facilitate a smart grid in which electricity is bought and sold through short-term dynamically negotiated contracts with local, community-owned and national energy providers in response to real-time pricing and carbon intensity signals.To achieve the goals outlined above, the project brings together an interdisciplinary team comprising world leading experts in the fields of intelligent agents and multi-agent systems (School of Electronics and Computer Science), renewable energy and energy efficiency in the built environment, and human factors in the design of automated control and feedback systems (Sustainable Energy Research Group and Transportation Research Group in the School of Civil Engineering and the Environment) at the University of Southampton. The home energy management agents will be evaluated and demonstrated within two live deployments: one using an existing test-bed of 9 homes in Havant, and one using 25 homes currently undergoing a social house redevelopment programme in Southampton.
Planned Impact
The technologies developed within this project will help towards the UK's legislated goal of cutting greenhouse gas emissions by 80% by 2050. As such, it will contribute to the global effort to mitigate the worst effects of climate change which challenges the future growth, prosperity, and political stability of all nations. Improving energy efficiency will assist in ensuring UK energy security by reducing the need to import gas from overseas markets, and it represents an opportunity for the UK to be at the forefront of innovation in the area of digital technologies for energy efficiency and smart grids (a market which Siemens estimates to be worth up to 27B over the next five years). Given the innovative digital technologies proposed within this project, which will effectively combine a thermal model of the home, a model of the householders energy use, a principled cognitive model of behaviour change, and sophisticated prediction and optimisation algorithms, our aim would be to better the 20% reductions in energy demand that are seen in well informed and motivated individuals using simple real-time energy displays. If this reduction were delivered across all 25M homes that are to be equipped with smart meters, this could potentially save UK householders 4.6B annually in energy costs. To ensure that these impacts are realised, the project has enlisted the support of an industrial advisory panel to enable early engagement with end users, to guide research toward industrially relevant goals and to identify early exploitation opportunities. This board will be led by a panel of executives (specifically, the chairman, managing director, chief scientist and technology manager) of PRI (a UK-based smart meter manufacturer operating in 50 countries worldwide). Other panel members will be (i) Horstmann Controls (the UK's leading manufacturer of controls for domestic heating systems and an existing partner of PRI) who will advise on the integration of the home energy management agents with current and future home heating systems, (ii) AlertMe (a UK-based manufacturer of home energy monitoring and security systems) who will advise on the immediate commercial potential of the home energy management agents, will work with the project partners to commercialise early research results, and will provide the energy, temperature and occupancy sensing systems for the trial deployments, (iii) Virgin Media (who support the Havant trial homes through the provision of broadband internet) will provide advice on the potential of bundled web, media and energy efficiency packages that will enable or support low-carbon living, and, (iv) the Energy Saving Trust (experts in the public engagement with energy issues) who will advise on future social policy and the dissemination of the research results. The project will hold two public one-day workshops which will encompass additional dissemination to stakeholders and potential end-users. The first workshop will highlight to stakeholders the components of the agents, the data to be gathered and to iron out any ethical issues that may arise through the process, and will provide an opportunity to discuss the implications of smart metering and the use of smart meter data with potential end users. The second workshop will be held 3 months before the end of the project and will summarise the final research outputs to stakeholders and will capture appropriate feedback. Finally, an internship programme, which is intended to enthuse and attract the next generation of researchers, involving students from both the University of Southampton and the Winchester School of Art (specifically, those studying graphic and communication design at post-graduate level), shall engage in demonstration activities and explore innovative and visually engaging means to publicise the specific results of the research project and the need for energy demand reduction in general.
Organisations
Publications
Alex Rogers (Co-Author)
(2012)
Delivering the Smart Grid: Challenges for Autonomous Agents and Multi-Agent Systems Research
Alex Rogers (Co-Author)
(2012)
A Scoring Rule-Based Mechanism for Aggregate Demand Prediction in the Smart Grid
Alex Rogers (Co-Author)
(2012)
Non-intrusive load monitoring using prior models of general appliance types
Auffenberg F
(2015)
A personalised thermal comfort model using a Bayesian network
Auffenberg F
(2017)
A Comfort-Based Approach to Smart Heating and Air Conditioning
in ACM Transactions on Intelligent Systems and Technology
Ghosh S
(2015)
Modeling the Thermal Dynamics of Buildings A Latent-Force- Model-Based Approach
in ACM Transactions on Intelligent Systems and Technology
Kirsten M.A. Revell (Author)
(2013)
Using the notion of mental models in design to encourage optimal behaviour in home heating use
Maleki S
(2020)
The Shapley value for a fair division of group discounts for coordinating cooling loads
in PLOS ONE
Neville Stanton (Co-Author)
(2012)
Affective design and its role in energy consuming behaviour; part of the problem or part of the solution?
Description | This project has developed a number of core technologies to monitor, visualise and control domestic energy consumption. |
Exploitation Route | A number of core technologies were deployed for use by the general public in MyJoulo - a free service allowing households to evaluate their heating energy consumption (www.myjoulo.com). The core approaches used to model and optimise energy consumption are now being taken forward by the project's industrial partners. MyJoulo was spun out of the University of Southampton as Joulo Ltd. and later acquired by Quby Ltd. |
Sectors | Digital/Communication/Information Technologies (including Software) Energy |
URL | http://www.homeenergyagents.info |
Description | The finding of the project led to the founding of a spin-out company - Joulo Ltd - in 2014. Joulo was acquired in January 2015 by Quby Ltd. |
First Year Of Impact | 2014 |
Sector | Digital/Communication/Information Technologies (including Software),Energy |
Impact Types | Economic |
Company Name | Joulo |
Description | Joulo develops energy monitoring systems that enable customers to download their energy usage data and receive advice on ways to minimise their energy consumption. |
Year Established | 2013 |
Impact | Conducted trials with three UK energy companies in 2013 and 2014. |
Website | http://www.joulon.com |