A New ANN Optimized By Improved PSO Algorithm Combined With Chaos And Its Application In Short-term Load Forecasting

Abstract

A new hybrid particle swarm optimization (PSO) algorithm with adaptive inertia weight factor (AIWF) is proposed. By incorporating chaotic local research method, it proposed the PSO which combined with chaos (CPSO), and applied it in evolving the artificial neural network (ANN). Then, based on the actual load data provided by a regional power grid in the south of China, the proposed method is used in the load forecasting. Results and comparisons with the PSO-ANN and the GA-ANN algorithms show that the CPSO can effectively enhance the searching efficiency and greatly improve the searching quality. While being used in the short-term load forecasting, the CPSO-ANN is better than the other algorithms in both forecasting effect and network function, such as PSO-ANN, GA-ANN and so on.