Foreseeing New Control Challenges in Electricity Prosumer Communities

This paper is dedicated to electricity prosumer communities, which are groups of people producing, sharing and consuming electricity locally. This paper focuses on building a rigorous mathematical framework in order to formalise sequen- tial decision making problems that may soon be encountered within electricity prosumer communities. After introducing our formalism, we propose a set of optimisation problems reflecting several types of theoretically optimal behaviours for energy exchanges between prosumers. We then discuss the advantages and disadvantages of centralised and decentralised schemes and provide illustrations of decision making strategies, allowing a prosumer community to generate more distributed electricity (compared to commonly applied strategies) by mitigating over- voltages over a low-voltage feeder. We finally investigate how to design distributed control schemes that may contribute reaching (at least partially) the objectives of the community, by resort in to machine learning techniques to extract, from centralised solution(s), decision making patterns to be applied locally. First empirical results show that, even after a post-processing phase so that it satisfies physical constraints, the learning approach still performs better than predetermined strategies targeting safety first, then cost minimisation.