Short Term Load Forecasting using SVM based PUK kernel

Accurate load forecasting performs a vital role in energy management and electricity market of electric systems. In this paper, Support Vector Machine (SVM) model based on Pearson VII universal kernel known as PUK kernel has been proposed for Short Term Load Forecasting (STLF). The proposed model can predict the 24-hour electrical output of distribution feeder in a distributed grid station. Machine learning algorithms have been applied in two parts. First the best kernel is selected among the three available kernels (RBF, Polynomial and PUK) in SVM model. Then kernel with optimum results are compared with the widely used Artificial Neural Network (ANN) machine learning model. The performance is verified with simulation on a case study based on 6 weeks' load pattern of distributed grid station in Pakistan. The forecasting results of PUK based SVM model are proved to be better than the other models (RBF, Polynomial and ANN) and indicate that the forecasting precision of the proposed methods is much efficient to other methods.