夏季最大负荷发生时间的预测

夏季最大负荷发生时间的预测是电力部门十分关注的问题,它关系到该地区的负荷调整方案以及购电计划制定。北京近年来夏季用电需求增长尤为突出。准确地估计夏季最大负荷发生时间,十分必要且迫切。为此,收集了北京市1990~2002年的夏季最大负荷发生时间数据,发现它是一个波动的、含有灰信息量的序列,故采用灰色系统理论进行预测建模分析。但常规的GM(1,1)模型在适应波动的数据方面,预测精度未能达到要求,故采用GM(1,1)残差周期修正预测模型来解决。经过计算发现,该模型不仅可以提高原始数据的拟合精度,而且用于预测时,与2003年夏季最大负荷发生时间的实际值进行比较,预测偏差也在允许的范围内。

Prediction of Occurring Date of Summer Maximal Power Load

The prediction of occurring dates of summer maximal power loads is a very important issue for electric utilities as the dates directly influence the scheme of regional load adjustment at the demand-side for the next year and the plan of purchasing electricity from a power pool.The electric power demands in the Beijing region during summer periods have grown rapidly in recent years.This makes it desirable and urgent to predict accurately the dates on which the maximal summer loads occur to ensure the secure operation of the power grid.This paper analyzes the pattern of dates from 1990 to 2002 on which maximal summer power loads occur in Beijing region.A fluctuant sequence with grey information has been detected.A grey system modeling method is then adopted to simulate the date sequence.Based on the findings from the experiments,this paper proposes a prediction model for the dates with residual periodic correction on the basis of GM(1,1).This prediction model is implemented and the comparison of predicted and real date on which the maximum summer load occurred in 2003 shows that the predicted date is substantially accurate.