Short-Term Power Load Forecasting Based on LS-SVM

Abstract

In order to solve the Short-term Load Forecasting problems in Power Systems, this article puts forward the Least Squares Support Vector Machine’s improved model by selecting the appropriate Gauss kernel function and proposing the error calculation analytical method, thus reduces the computational complicate problems when large amount of data is input in Short-term Power Load Forecasting. An example is given to prove the validity of the algorithm.