Impact of demand response contracts on short-term load forecasting in smart grid using SVR optimized by GA

Abstract

In a Smart Grid environment the performance measure of the grid is calculated by considering the fact that how accurately and precisely a load forecasting (LF) is done. A true Load Forecasting is vital to make a current grid smarter and more reliable when it comes to its performance. Demand Response (DR) contracts is a type of program in smart grid where the customer is free to select a type of contract which is given by the utility and is one of the growing factor which affects the load forecasting results in the Smart Grid, therefore in order to do a complete evaluation of smart grid performance and to accomplish an accurate load forecasting results the different types of contracts should also be studied. The purpose of this study is to accomplish two goals. The first one is to develop a suitable model which can incorporate various factors that can affect the load forecasting results. The subsequent goal is to identify the impact of the demand response contracts on the load forecasting results. In the proposed study, Support Vector Machine-Regression (SVR) is selected as the base methodology to perform a Short — Term Load Forecasting (STLF) under smart grid environment.