基于BP神经网络的短期负荷预测建模仿真

Modeling and Simulation of Electricity Short-Term Load Forecast Based on Neural Network

通过BP神经网络与Matlab相结合,建立起三层四功能单元的BP神经网络短期负荷预测模型,并采用某条线路1年的历史负荷波动数据对模型进行"学习"训练。预测日24 h负荷数据的Matlab仿真及误差分析结果表明,所构筑的BP神经网络模型具有较高的可靠性和准确性,误差率控制在2%以内。BP神经网络模型大大提高了短期负荷预测数据的处理效率与可信性,为研究电力系统经济调度提供了一种新的非线性仿真建模模型。

Combining back propagation （BP） neural network with Matlab, the three layer and four-function BP neural network short-term load forecasting model was built. One year load fluctuations history data was used to train the model. The daily load predicting data by Matlab simulation and error analysis result show that the error rate can be effectively controlled in less than 2% which verifies the reliability and accuracy of the constructed BP neural network model. The BP neural network model greatly improves the efficiency and accuracy of short-term load forecasting data processing, which provides a new nonlinear simulation model for researching on power system economic dispatch.