基于灰色理论与BP神经网络组合模型的中长期负荷预测

The Prediction of the Medium and Long-term Load Based on Combined Model of the Grey Theory and BP Neural Networks

灰色系统预测模型是中长期负荷预测的一种有效方法,但是,此模型存在未考虑经济因素对负荷发展的影响及难以满足高精度要求的缺陷,构建了考虑经济因素影响的灰色BP神经网络组合预测模型,通过灰色关联分析方法确定影响负荷的主要经济因子。主要经济因子的引入,使预测模型更符合实际、更合理。应用此组合模型对某省全社会用电量进行了中长期预测,结果表明,该模型具有更高的精度和更好的实用性。

Although the grey system model is an efficient method for medium and long-term forecasting, but it does not take the influence of industry factor on the load growth into account and can not meet the demand of higher accuracy. This paper fills the gap by establishing the con-bined forecasting model of the grey BP neural network that takes the industrial factors into account. The key industrial factors are verified by means of the grey associating analysis, making the forecasting model more practicable and reasonable. The proposed combined forecasting model is applied to the medium and long-term load forecasting of electricity consumption in a province in China. The results show that this combined model is more precise and practicable.