上下游行业用电变化的因果关系及预测方法

Causality and forecasting methods of electricity consumption changes in upstream and downstream industries

通过分析区域内行业用电量的关联关系,可以挖掘行业间的上下游关系,并进一步对行业用电量建模,从而对未来数月的用电量进行预测,以此来应对可能发生的突发情况。利用关联性分析模型,可以对行业用电量的时间序列数据进行深入分析,通过格兰杰因果关系检验方法验证行业间的上下游关系,然后利用向量自回归模型对行业用电量进行建模。实验结果表明,使用关联性分析模型,可以很好地预测用电量趋势,验证了关联性分析模型的有效性。

By analyzing the correlation between industry power consumption in the region,we can explore the upstream and downstream relationships between industries and further model the industry power consumption,so as to predict the electricity consumption in the next few months,in order to deal with unexpected situations that may occur.Using the correlation analysis model,it is possible to conduct in-depth analysis of the time series data of industry power consumption.The Granger causality verification method is used to verify the upstream and downstream relationships between industries,and then the vector autoregressive model is used to model the industry’s electricity consumption.The experimental results show that the use of the correlation analysis model can predict the electricity consumption trend well,which verifies the effectiveness of the correlation analysis model.