Optimal Prediction in Local Electricity Markets

Over the next four years this Fellowship aims to build an internationally leading research team in stochastic modelling of local energy markets. The Fellow and researcher will develop theory and numerical methods to solve emerging mathematical problems in UK power systems, with a focus on modelling the community scale, in order to achieve the maximum welfare benefit from the design of these markets. The research programme will be carried out in the context of the Fellow's existing Energy group, which includes three PhD students working in power systems and energy storage. This work will involve fruitful interactions with other probabilists in the UK, power systems engineers in the UK, US and Canada, and UK industry experts working on such problems. It will have significant impact through the creation of algorithms and software, enabling the efficient numerical solution of planning and operational problems for local electricity markets. A longer term impact will be to further establish Manchester (and the UK) as a centre for talented researchers in cross-disciplinary applications of probability theory to power systems.

The ultimate impact from the application in future power systems will be to UK plc. The need for new and fruitful approaches to problems in future power systems, particularly at the distribution level and with respect to demand participation and storage, is evidenced by the recent Technology Stategy Board (TSB) 'Future grid: Smart power distribution and demand' competition for feasibility funding, and by the EPSRC Storage Grand Challenge. The Fellow currently has applications under consideration in both of these initiatives and, in order to maximise impact, will remain fully involved in such initiatives over the period of the Fellowship.

The Fellow's proposal to the TSB competition is to assess the feasibility of extending his 2011 knowledge transfer project with IBM UK Ltd (List of Publications [8, 9]) to an application on the Isle of Wight in partnership with the EcoIsland Community Interest Company. IBM UK Ltd have committed to making an in-kind contribution of time worth £2,000 assessing the feasibility of the communications architecture necessary for the project. Since this and similar industrial collaborations will be key pathways to impact, the Fellow has requested 10% of the time funded by this Fellowship to pursue engagement with the public and with industry. Knowledge transfer activity in this area is likely to generate intellectual property (IP) and the Fellow will work closely with UMIP, the University of Manchester's IP commercialisation company, to ensure that these issues of IP are dealt with transparently and professionally.

Electricity supply is a key public service, and its efficiency is crucial to the UK's economic competitiveness. Further, the productive efficiency of electricity generation is increased by efficient local power systems which can respond to price signals, for example through reduced spillage of wind generation and reduced use of inefficient thermal peaking plant generation. This leads to reduced carbon emissions and the associated environmental benefits. To maximise the impact of the Fellowship in this respect, results on local electricity markets will be made available to regulatory policymaking bodies as appropriate, in a format suitable for use as an evidence base.

Regarding timescales, the UK is committed to achieving its legally binding targets for integration of renewable electricity generation by 2020. At the same time technological advances have recently been made in the physical assets required for local electricity markets, and continue to be made. As an example in April 2011 Sony began bulk shipments of 1.2kW 'Ion Energy' battery storage modules suitable for local applications, with a lifespan of over ten years based on one daily charge/discharge cycle at room temperature. Since the application area of this Fellowship concerns the underpinning mathematics and welfare economics of local electricity markets, its impact can realistically be expected to be realised within 8-10 years.