基于优化灰色理论和神经网络的电力负荷短期预测

Short Term Forecasting of Power Load Based on Optimized Grey Theory and Neural Network

为提高电力系统短期负荷的预测精度,确保电网安全和经济运行,提出一种基于优化的灰色理论和Elman神经网络混合方法。该方法充分考虑温度因素、周类型、天气状况等影响预测精度的不确定因素,通过数据模拟预测,该方法具有较高的预测精度和收敛速度,在电力系统短期负荷预测中具有一定的应用价值。

In order to improve the accuracy of power system short term load forecasting and to ensure security and economy of power grid, a hybrid method based on improved gray theory and Elman neural network is proposed. This method takes into account some factors affecting the prediction accuracy, such as temperature, week type and the weather condition. It has been proved that the method has higher prediction accuracy and convergence rate through data simulation. It has certain application value in the short term load forecasting of the power system.