Electricity Market Price Assesment Utilizing Hybrid PSO – ANN Algorithm

Abstract

Prediction of cost is the most imperative task and the reason for settling on choices in competitive bidding strategies. Reliability, Robustness and optimal benefits for the market players are the fundamental concerns which can be accomplished by a point value anticipating module constitute of diminutive prediction errors, reduced complexity and lesser computational time. Thus in this work, a coordinated methodology dependent on Artificial Neural Networks (ANN) prepared with Particle Swarm Optimization (PSO) is proposed for momentary market clearing costs anticipating in pool based electricity markets. The proposed methodology overcomes the difficulties like trapping towards local minima and moderate convergence as in existing techniques. The work was speculated on territory Spain electricity markets and the outcomes obtained are compared with hybrid models presented in the previous literature. The response shows decline in forecasting errors that are recognized in price forecasting. The total research may help the ISO in finding the key factors that are fit for expectation with low errors.