基于人工神经网络的日负荷预测方法的研究

AN APPLICATION OF SHORT-TERM LOAD-FORECASTING BASED ON ARTIFICIAL NEURAL NETWORK

基于人工神经网络原理，设计了一个三层的ＢＰ网络模型。充分利用了神经网络高度非线性建模能力，实现电力系统的短期负荷预测。文中对样本数据进行了预处理，以及在算法中引入附加冲量项，以提高训练速度。预测仿真结果证明使用人工神经网络方法进行短期负荷预测是可行的。

Based on the theory of artificial neural network(ANN),a three-layer back propagation(BP) network is proposed. The idea is to predict short-term load using the ability of ANN to model arbitrary nonlinear systems. In order to improve training speed, the training data are pretreated and an additional impulse term is introduced into BP algorithm. In order to improve the precision, the selection of the hidden nods is studied in this paper, and a best network is generated finally. Simulation results show that the effectiveness of the proposed method based on ANN.