基于小波神经元网络模型的网损预测方法研究

提出了一种电力系统网损预测的新算法-小波神经元网络预测模型，它以非线性小波基为神经元函数，通过优化伸缩因子和平移因子确定对应务神经元的小波基函数，从而合成小波神经元网络，达到全局最优的拟合效果，克服了普通人工神经元网络学习速度慢、难以确定网络结构、存在局部极小点等方面的缺点，仿真结果表明，该方法准确，并可使学习速度大大提高。

A New Power Loss Forecast Method Using Wavelet Neural Network

A new power loss forecast method is proposed. The method is based on wavelet neural network in which the nonlinear wavelet basis function is used as the excitation functions of neurons. The configuration and the parameters of the wavelet neural network are determined by optimizing both expansion and contraction factors and achieve the global best approximation effect. By comparison with the general artificial neural network model, the new method is fast in learning, easy in configuration decision and can overcome the local minimal solution. Simulation results show that the new power loss forecast method is fast and accurate.