基于短期负荷打捆预测的售电公司偏差考核控制方法

Deviation assessment and control method for electricity sales companies based on short-term load bundling forecast

偏差电量考核是电力市场改革至关重要的一个过渡环节,部分售电公司缺乏针对偏差电量考核进行的负荷预测分析,导致偏差考核准确性较低。针对上述问题,提出了基于负荷打捆和反向传播(BP)神经网络模型的负荷预测方法,同时分析了负荷预测精度对偏差考核结果的影响机理。根据重庆地区某售电公司10家用户历史负荷数据进行负荷打捆预测并计算偏差考核结果,结果表明该方法可有效减少售电公司的偏差考核电量。

Assessment on electricity deviation is a crucial transitional link in the electric market reform.For lack of scientific and effective load forecasting analysis on electricity deviation assessment,some power sales companies are of low assessment accuracy.In view of the above situation,a load forecasting method based on the load bundling forecasting and BP neural network model is proposed.The influence of load forecasting on the accuracy of deviation assessment is analyzed.Ten power sales companies in Chongqing executed forecast on their bundled load based on their customers'power consumption data.According to the calculated results of the deviation assessment,this method can alleviate the deviated electricity in the assessment of power sales companies.