One day forth forecasting of hourly electrical load using genetically tuned Support Vector Regression for smart grid frame work

This paper is dedicated to the analysis and betterment of a very recognized field of research in the field of Electrical Power System, i.e. Short Term Load Forecasting (STLF) which is very essential for utilities to control, manage and schedule generator units. Various available approaches to STLF focus to minimize the training error. This paper presents two forecasting models viz. three-day-trained Support Vector Regression model and parameter optimized Support Vector Regression using Genetic Algorithm. Unlike existing models, these models accomplish accurate forecasting by optimizing the regularized risk function. The models make use of previous three days hourly load data for predicting next day hourly load. They give better results with only three days´ training data, and thus have potential to be used for STLF for smart grid in real time.