基于BP神经网络系统的短期电力负荷预测

Short-term Power Load Forecasting Based on BP Neural Network System

以河南某地区电网2009年和2010年的电力负荷数据为基础,设计了一个由输入层、隐含层和输出层组成的三层BP网络模型,利用神经网络高度非线性建模能力,在不考虑温度影响情况下,采用BP神经网络对该地区短期电力负荷进行预测.探讨了负荷预测模型分类模式,对应用于实际的BP神经网络算法进行了具体处理.结果表明,基于BP神经元网络的短期电力负荷预测方法具有精度高的特点,取得了均方根误差小于3.02%的精度.

Based on the power load data in 2009 and 2010 of a certain area power grid in Henan,a three layer BP neural network model consisting of input layer, hidden layer and output layer is designed. Using the nonlinear modeling ability of neural network, the short-term power load in the area without considering the temperature influence is forecasted. The classification mode of the load forecasting models is discussed and the actual BP neural network algorithm is handled. The results show that the neural network for short-term load forecasting method based on BP has high accuracy, the root-mean-square error is less than 3.02% .