Short Term Load Forecasting Using Core Vector Regression Trained with Particle Swarm Optimization

Short term load forecasting is very essential to the operation of electricity companies. In this paper, we propose a new method for short term load forecasting trained by PSO and Core Vector Regression (CVR). The CVR algorithm extend Core Vector Machine algorithm to the regression setting by generalizing the underlying minimum enclosing ball problem. In this paper, we use particle swarm optimization (PSO) to optimize the parameters of the CVR. Experiments show that the PSO optimized method has comparable performance with SVR (Support Vector Regression), but is much faster and produces much fewer support vectors on very large data sets.