基于灰色关联分析和BP神经网络的用电量预测

Electricity Consumption Forecast Based on Grey Relational Analysis and BP Neural Network

随着电网的高速发展,针对目前年度用电量预测的样本数据在新常态下电力负荷变化趋势和过去有一定的差异的问题,文章采用灰色关联分析与BP神经网络相结合的方法对湖北省年用电量进行预测。根据1997-2016年湖北省年用电量及其10个影响因子的数据,采用灰色关联分析法对数据进行处理,选取确定3个关联度较大的影响因子作为BP神经网络的输入参数,建立BP神经网络用电量预测模型,最后对湖北省未来几年用电量进行了预测。预测结果表明,灰色关联分析及BP神经网络方法在用电量预测上精度较高,计算方便,可为电力部门提供参考。

With the rapid development of power grids,the sample data of the current annual electricity consumption forecast has a certain difference in the trend of power load under the new normal state.The article uses the method of gray correlation analysis and BP neural network to predict the annual electricity consumption in Hubei Province.According to the data of annual electricity consumption and its10impact factors in Hubei Province from1997to2016,the data was processed by grey correlation analysis method,and three influencing factors with large correlation degree were selected as input parameters of BP neural network,the BP neural network electricity consumption prediction model was established and finally the electricity consumption in Hubei Province in the next few years was predicted.The prediction results show that the gray correlation analysis and BP neural network method have higher accuracy in power consumption prediction and are convenient to calculate,and it can provide reference for the power sector.