基于模糊聚类和相似度的电力系统可靠性原始参数预估  

Prediction of original reliability parameters of power system based on fuzzy clustering and similarity

作　　者：张勇军[1] 陈超[1] 许亮[1] 

机构地区：[1]华南理工大学电力学院,广东省绿色能源技术重点实验室,广东广州510640

出　　处：《电力系统保护与控制》2011年第8期1-5,共5页Power System Protection and Control

基　　金：国家自然科学基金重点资助项目(50337010)~~

摘　　要：以可靠性气候影响因素作为线路模糊集元素,通过模糊聚类将可靠性原始参数已知和待求的线路进行适当分类,引入β作为考虑其他影响因素的修正系数,在分析欧氏贴近度、模糊贴近度、灰色关联度各自反映模糊集特征的差异性和局限性的基础上提出相似度指标,在同类中以可靠性参数已知的线路的可靠性影响因素序列为基准向量,以可靠性参数未知的线路的可靠性影响因素序列为待检向量,使用相似度指标进行求取。实例证明该方法的可行性和预估结果的准确性,并讨论了不同的权重系数取值会对结果产生影响。The climatic factors that affect reliability are considered as fuzzy elements of lines,and all lines with known and unknown original reliability parameters are divided into several types by fuzzy clustering.β is introduced to consider as the modified coefficient of other factors.The similarity index is proposed based on the analysis of the difference and limitations of Euclid approach degree,fuzzy closeness degree and grey correlation degree in embodying fuzzy set features.The reliability-affecting factors sequence of lines with known reliability parameters is adopted as the reference vector,and that with unknown reliability parameters is adopted as the vector to be measured in the same types.The similarity index is used to obtain the original reliability parameters.Cases show that the proposed method is practicable and accurate.Finally the effects of different values of weight coefficients on the results are discussed.

关 键 词：电力系统 可靠性原始参数 模糊聚类 相似度 贴近度 灰色关联度 

分 类 号：TM732[电气工程—电力系统及自动化]