For the safe operation of power system, short-term load forecasting is essential. The load of the power system is usually with time showing a certain nonlinear wave range, according to the variation of load characteristics in power system, a method by introducing Kalman filtering algorithm with a modifying factor to achieve the short-term load forecasting. By forecasting short-term load in the Chengdu region, it shows that this method has a higher prediction accuracy compared with the traditional Calman filter. At the same time, compared with other new intelligent algorithms, it has the advantages like fast convergence and short time consumption. The MATLAB simulation shows that the improved algorithm provides a new way for short-term load forecasting.
