The Use of Fuzzy Cognitive Agents to Simulate Trading Patterns within the Liberalised UK Electricity Market

The world is becoming increasingly competitive by the action of liberalised national, regional, or global markets. Electrical power systems are not immune to such change, and many national power systems have are now subject to the influence of overtly neo-liberal market models. The resulting electrical markets are subject to increasingly complex rule bases put in place to ensure that individual consumer and national imperatives are met. Further, these markets are also increasingly moving towards real time trading, reducing the time available for critical decisions to be made. In an effort to develop a deeper understanding of these evolving markets and to create effective system support tools for market participants, many simulation techniques such as neural networks and expert systems have been applied. In this paper we report our efforts to develop an effective simulation platform using fuzzy cognitive agents. Our approach is based upon the encapsulation of fuzzy cognitive maps (FCM) generated on the Matlab Simulink platform within commercially available Intelligent Agent software. The approach permits the retention of the visually simple but rich relational complexity of the Matlab Simulink based FCM, whilst enhancing their domain applicability by integrating it with the relational and structural flexibility of the Intelligent Agents. We will describe a potential architecture for the fuzzy cognitive agent and report on our first attempts to integrate the Matlab Simulink based FCM with the jack intelligent agent toolkit will be presented