Load Forecasting In Power Systems Based On Artificial And Fuzzy Neural Network

After analyzing the meaning and methods of power system load forecasting, the paper explains the general theory and meaning of artificial neural network(ANN) and fuzzy inference system(FIS), and studies load forecasting methods based on artificial neural network and fuzzy neural network(FNN).The paper adopts two ways to improve traditional artificial neural network. The first category involves the development of heuristic techniques which include such ideas as varying the learning rate, using momentum and rescaling variables. Another category of research has focused on standard numerical optimization techniques which are mainly the conjugate gradient algorithm and the Levenberg-Marquardt algorithm. The results show these methods can get better improvement than traditional backpropagation network.Then, the paper studies adaptive fuzzy neural network for short-term load forecasting. This network combines fuzzy inference system with artificial neural network and forms fuzzy neural network. The results show it can also get better improvement than traditional backpropagation network.The two methods give new ways for load forecasting and also a new thought for using them, and have certain theoretical and practical value.