Application of fuzzy neural network to power system short-term load forecast

According to the features of short time power load, the influence of temperature, weather and day type are considered in this paper. A fuzzy neural network approach which combined the neural model and fuzzy logic method for short-term load forecasting is presented. In this method, the temperature, date and weather data are translated into fuzzy set firstly. Then use them as the inputs of the neural net works model. At last, the method was simulated by using MATLAB, and took the data from an area in Baoding as an example. The results show that it has higher prediction accuracy than normal BP algorithm.