Short Term Load Forecasting improved by ensemble and its variations

The implementation of an effective Short Term Load Forecasting (STLF) method with minimal error has been one key goal of the power industry. By accomplishing the goal, electrical utilities can effectively plan economical scheduling of generating capacity, scheduling of fuel purchases, security assessment, and planning for energy transactions. Various developments have been done to improve the forecasting accuracy, of Neural Network, Auto Regression, and Multiple Regression, which are the major forecasting methods used for load forecasting. In this paper, Short Term Load Forecasting by Ensemble method is proposed and evaluated on the actual load and metrological data in PJM in the USA with encouraging results.