The realisation of fuzzy cognitive agents representing electricity market participants

Worldwide trading markets are becoming increasingly complex and competitive making it increasingly difficult for participants to optimise their trading actions. Electrical trading markets are no exception to this phenomenon and are faced with the additional and combined uncertainties associated with need to continuously balance supply and demand and trading decisions actions which are moving ever closer to real time. In response, many differing computer market simulation techniques have been investigated to develop either a deeper understanding of these evolving markets or to create effective system support tools. In this paper we present a novel new approach within which Simulink¹ - a graphical block diagramming tool and a customizable set of block libraries was used to model fuzzy cognitive maps encoding individual market trading participants, then encapsulated within AI-Agent type architecture. These fuzzy cognitive maps dynamically interact with each other by the communication protocols of the AI-Agent. This new paradigm is termed a Fuzzy Cognitive Agent, and multiple agents of this type a Fuzzy Cognitive Agency.