Simulated Annealing Theory Based Particle Swarm Optimization for Support Vector Machine Model in Short-Term Load Forecasting

Support vector machine (SVM) is based on the statistical learning theory. It has recently been successfully used to solve nonlinear regression and time series problems and has been applied to predict values. The key problem of SVM is the choice of SVM parameters. Particle swarm optimization (PSO) algorithm has the ability of global optimization. This paper proposed an improved PSO algorithm based on simulated annealing (SA) theory. The strong searching ability of SA was employed to PSO algorithm to avoid the premature convergence with better stability and astringency. The SA based PSO algorithm was used to optimize the parameters of SVM model. The study used the new model to forecast load of electric power system. The simulation results show that the accuracy has been improved by using SA-PSO based SVM model than that of the traditional SVM load forecasting model. It provides an alternative for forecasting electricity load.