电力负荷短期预测的模糊专家系统修正方法  

Updating of short-term load forecasting employing fuzzy expert system

作　　者：招海丹[1] 余得伟 

机构地区：[1]华南理工大学电力学院,广东广州510641 [2]广东省电力中心调度所,广东广州510600

出　　处：《广东电力》2001年第1期1-4,共4页Guangdong Electric Power

摘　　要：针对人工神经网络短期负荷预测方法的不足 ,考虑天气中的日平均气温、天气状况以及特殊事件等影响负荷变化的主要因素 ,利用专家经验 ,模仿专家处理问题的方法 ,设计了一个模糊专家系统 ,对负荷预测结果进行修正 ,以提高负荷预测精度。通过合理选择模糊推理规则的形式 ,有效地减少了规则的数目 ,使得人工总结专家经验并确定模糊推理规则成为可能 ,减少了计算量 ,提高了算法速度。In this paper, one fuzzy expert system is designed according to the main reasons of daily average temperature and special events which would affect the change of the load, utilizing expert experience and imitating the method by which the experts deal with the problem, in allusion to the shortcoming of the short term load forecasting employing artificial neural network. The load forecasting result can be updated, then the accuracy can be improved. The number of the rules is reduced via the reasonable choice of the fuzzy reasoning rules, which make it possible that the expert experience is summarized artificially to the fuzzy reasoning rules. Calculating quantity is reduced and the algorithm speed rises,too.

关 键 词：模糊专家系统 短期负荷预测 电力系统 人工神经网络 人工智能 

分 类 号：TM715[电气工程—电力系统及自动化]