电力负荷的模糊粗糙集预测方法研究

通常空间电力负荷的影响因素较多 ,小区划分较小时模糊推理规则将成倍增加 ,为此引入了粗糙集理论的属性约简方法简约预测因子 ,即采用粗糙集理论的不可分辨性关系约简可能影响小区用地决策的距离和环境等相关属性 ,得出决定小区用地类型的决策推理规则库 ;为克服模糊理论确定权重系数的主观性 ,应用粗糙集理论中属性重要度分析计算了各条件属性对各用地类型的不同权重。重庆江北某区域空间电力负荷预测的结果显示 ,粗糙集与模糊集理论相结合能节约存储空间且能提高预测速度

Electric Load Forecasting Method Based on Fuzzy-rough Set

Because geographical, environmental and man-made influence makes load forecasting of small areas complex in spatial electrical load forecasting, fuzzy reasoning rule triply increase while small areas is deal with more influential factor. In order to solve above problems, author introduces attribute of distance and environment etc likely affecting small area land-use decision-making by indistinguishable relation of rough set, which achieve decision-making reasoning rule database of small area land-use type. At the same time, author analyzes and counts different weights of condition attribute to land-use type by attribute importance tolerance of rough set. The result shows that combination of rough set and fuzzy set may not only saves memory space, but also improves forecasting rate by spatial electrical load forecasting of some area at Jiangbei in Chongqing.