Day ahead a electricity market analysis through a neuroevolution algorithm

The ongoing changes in deregulated electricity markets creates a need of new tools for the analysis and simulation of electricity markets. The current paper proposes a new methodology based on neuroevolution algorithm for the analysis of a day-ahead electricity market. In this work, we consider a multi-period electricity market where market agents (suppliers and consumers) have to submit their offers for each hour of the trading day. A deregulated electricity market is an example of a stochastic game, market agents have to compete among themselves, in an uncertain environment, in order to maximize their profits from the day ahead transactions. A case study based on IEEE30 bus system is used to demonstrate the ability of neuroevolution algorithms as a powerful tools for the analysis of deregulated electricity markets. The proposed approach combines the ability of neural network to learn from a stochastic environment, and the optimization capability of coevolutionary algorithms to find the best strategies for market agents.