基于模糊小波网络的电力系统短期负荷预测方法

提出一种基于模糊小波网络的短期负荷预测模型。模糊小波网络结合了小波变换良好的时频局域化性质、模糊推理和神经网络的学习能力,因此函数逼近能力大大提高。模糊小波网络由一组模糊推理规则和若干小波子网络组成,其中模糊规则的结论部分与某一特定尺度的小波子网络相对应。在学习过程中通过同时调整小波基函数的平移因子和隶属度函数的形状,使得模糊小波网络的精度和泛化能力大大提高。实例计算表明,这种模型是切实可行的。

A method of power system short-term load forecasting based on fuzzy wavelet neural networks

A novel short-term load forecasting model based on fuzzy wavelet neural networks(FWN) is presented in this paper. Because FWN combines the time-frequency localization ability of wavelet, fuzzy inferring and the education character of ANN together,its ability to reach the global best results is greatly improved. The FWN includes a set of fuzzy rules and several sub-WNNs. Every sub-WNN, corresponding to a certain fuzzy rule, consists of wavelets with a specified dilation .By adjusting the translation parameters of the wavelets and the shape of membership functions, the accuracy and generalization capability of FWN can be remarkably improved. The calculation result shows that the presented model is effective.