根据短期电力负荷预测的特点，提出一种负荷预测新算法——小波神经元网络负荷预测模型。它以非线性小波基为神经元函数，通过伸缩因子和平移因子计算小波基函数合成的小波网络，从而达到全局最优的逼近效果，同时有效地克服了人工神经元网络学习速度慢、难以合理确定网络结构、存在局部极小点的固有缺陷。经实例验证，该方法能有效地提高预测精度，可用于短期电力负荷预测 According to the characteristics of short-term power load forecasting, a new load forecasting algorithm-wavelet neural network load forecasting model is proposed. It takes the nonlinear wavelet base as the neuron function and calculates the wavelet network synthesized by the wavelet basis function by scaling factor and translation factor so as to achieve the global optimum approximation effect and overcomes the problem that the artificial neural network learning speed is slow and difficult to reasonably Determine the network structure, there are inherent shortcomings of the inherent defects. The example verified that this method can effectively improve the prediction accuracy and can be used for short-term load forecasting