基于BP神经网络算法的短期电力负荷预测研究

Research on short⁃term power load forecasting based on BP neural network algorithm

针对电力调度机构对短期负荷预测的精度日益提高的问题,文中应用BP神经网络算法对电力负荷进行了短期预测,该方法模拟了人类大脑神经的功能,对收集到的样本数据进行处理和储存。在已经建立完成的负荷预测模型的基础上,分别收集预测区域的温度、气象等相关数据信息,进而确定所需训练的样本;对上述数据进行归一化处理,经过神经网络正向与反向传递得到期望值;应用BP神经网络算法优化实现了负荷预测模型,以某区域典型日负荷曲线为例进行实验测试,预测数值分析比对结果验证了所提算法的有效性和精确度。

Aiming at the problem that the accuracy of short⁃term load forecasting of power dispatching organization is improving day by day.In this paper,BP neural network algorithm is applied to short-term forecasting of power load.The method simulates the function of human brain nerve and processes and stores the collected sample data.On the basis of the established load forecasting model,the temperature,weather and other related data information of the forecasting area are collected respectively,and then the training samples are determined.The above data are normalized,and the expected value is obtained through forward and backward transmission of neural network.The BP neural network algorithm is applied to optimize the load forecasting model,and the typical daily load curve of a certain area is taken as an example to carry out the experimental test.The numerical analysis and comparison results verify the effectiveness and accuracy of the proposed algorithm.