Short-term load forecast based on decomposition of daily load curve

Short-term load forecasting is a part of power dispatching automation system. Accurate load forecasting can ensure and stable operation of power system, and reduce the cost of power generation. The load of different industries was influenced by different factors; the impact of special factors will be considered respectively when the loads were separated into different industries. Thus, the result of the load forecasting will be more accurate. In this paper, the characteristic of each substation was analyzed, on the basis of the load composition equations the typical daily curve of each substation can be separated, and the load of each substation can be forecasted, also. Then, the total forecasting result was obtained. Improved gray forecast model was carried on when forecasted. According to this model, not only the handling ability to the fluctuation load data series was strengthened, but also the error was eliminated. The example shows that the method proposed by this paper can gain a better result.