论文部分内容阅读
利用广义误差反向传播算法生成了一种应用于电力系统短期负荷预报的神经网络模型,用以克服传统BP网络所存在的易陷入局部极小点和对初值要求较高的缺陷。模型同时考虑了影响短期负荷预报的若干重要因素,从而增强了模型的精确性和适用性。
A generalized error back propagation algorithm is used to generate a neural network model for short-term load forecasting of power system to overcome the defects of traditional BP network that it is easy to fall into a local minimum and has a high initial value. The model also considers several important factors that affect short-term load forecasting, thus enhancing the accuracy and applicability of the model.