Medium-Load Forecasting Based on Improved GM(1,1) Model

The traditional grey model has been successfully utilized in the field of forecasting system. However, many results have shown that the model still could be improved. This paper presents an improved grey model (GM) algorithm for the medium-term load forecasting. The traditional GM(1,1) forecasting model has a parameter a which is constant and it is not accurate for medium-term load forecasting. The improved GM(1,1) model utilized genetic algorithm to search the optimum a value. Meanwhile equal dimension new information is applied to modify the traditional GM(1,1). A load forecast example is used to test the traditional GM(1,1) and improved GM(1,1), which shows that the improved GM(1,1) had a better performance.