Abstract:  Based on the peak load data and the daily meteorological observational data from 2007 to 2010 in the southern Hebei province, the annual and daily variations of the peak load were analyzed. The results show that the trend curve of the peak load has three peaks and two valleys. Three peak loads appear in the spring irrigation period, the summer high temperature period and before and after the winter heating period, while two low valleys are in the spring festival of China and from October 1 to 7. The three irrigations of winter wheat had a significant effect on the peak load. The regional strong wind and rainfall could make the peak load decreased obviously. The grades of daily peak load are divided based on its variation amplitude. The relationship between peak load and meteorological factors was analyzed using methods of a correlation analysis and a multiple regression analysis, and a forecast model is established. This model is favorable to adjust the routine electric power according to the practical test.

Key words: Spring irrigation, Daily peak load, Amplitude, Forecast model