Multi-area load forecasting for system with large geographical area

Abstract

In a power system covering large geographical area, a single model for overall load forecasting of the entire area sometimes can not guarantee satisfactory forecasting accuracy. One of the major reasons is due to the load diversity, usually caused by weather diversity, throughout the area. Multi-area load forecasting will be a feasible and effective solution to generate more accurate forecasting results, as well as provide regional forecasts for the utilities. However, the major challenge is how to optimally partition/merge the areas according to the load and weather conditions. This paper investigates the electricity demand and weather data from an electric utility in Midwest US. Based on the data analysis, we demonstrate the existence of weather and load diversity within its control area, and then develop a short-term adaptive multi-area load forecasting system based on support vector regression (SVR) for day-ahead operation and market. The proposed multi-area forecasting system can find the optimal area partition under diverse weather and load conditions, and finally achieve more accurate aggregate load forecasts. The proposed forecasting system has been tested by using the real data from the system. The numerical results obtained for different area partition schemes validate the effectiveness of the proposed multi-area forecasting system. The detailed discussions on the forecasting results have also been given in this paper.