粗集理论在电力系统负荷预测中的应用.pdfVIP

第28 卷 第3 期 电 网 技 术 Vol. 28 No. 3 2004 年2 月 Power System Technology Feb. 2004 文章编号：1000-3673 （2004 ）03-0029-04 中图分类号：TM715 文献标识码：A 粗集理论在电力系统负荷预测中的应用 罗治强，张 焰，朱 杰 （上海交通大学电气工程系，上海200030 ） APPLICATION OF ROUGH SET THEORY IN ELECTRIC POWER LOAD FORECAST LUO Zhi-qiang ，ZHANG Yan ，ZHU Jie （Department of Electrical Engineering ，Shanghai Jiaotong University，Shanghai 200030，China ） ABSTRACT: The accuracy of power load forecasting is 适应性和预测精度。这些方法包括：模糊预测法[1] 、 significant to the reasonable planning and construction of power 灰色预测法[2] [1] 以及人工神经元网络预测法 等。但 system. The main factor influencing the accuracy of load 应用这些方法处理不确定性因素时往往需要作特 forecasting is the uncertainty of original data. A newly developed 定假设和外部参数的配合。 mathematical approach, i.e., the rough set theory, is used to 本文引入粗集理论以解决电力系统负荷预测 research the mid- and long-term load forecasting considering the 数据源中最常遇到的缺失数据的补齐问题，实现 influence of data uncertainty. On the basis of practical examples 从数据源中优选出最有影响的因素，排除联系较 and by use of rough set theory, a complete procedure of power 弱的因素，最后以决策规则形式实现负荷预测。 load forecasting including the modeling of power load forecasting, supplement of all lost data, preparation of effective 2 粗集理论简介 data, generation of decision-making rule, and finally, educing [3] 粗集理论 是1982 年由波兰科学家Z ．Pawlak forecasting results is presented. The analysis of the calculation 提出的。它是一种用于研究不完整、不确定性以 examples shows that it is feasible to forecast the power load 及模糊知识数据的表达、学习和归纳的理论。该 und