基于改进算法的粗糙集理论故障选线装置研究  

Research on improving algorithm of the fault line selection device based on the rough set theory

作　　者：张凯[1] 袁兆强[1] 范李平[1] 

机构地区：[1]三峡大学电气信息学院,湖北宜昌443002

出　　处：《电力系统保护与控制》2010年第4期15-20,共6页Power System Protection and Control

基　　金：湖北省自然科学研究项目(D200513009);湖北省自然科学基金项目(2006ABA202)~~

摘　　要：针对KA系列模糊智能选线装置中小波包多频带选线算法所出现的误选问题,提出了基于粗糙集理论的小波包多频带的选线算法。该算法主要是应用粗糙集理论对故障样本的数据挖掘能力来确定低频衰减信号的增强比例,并在此基础上进一步利用粗糙集理论的信息融合功能对单相接地故障电流信号中所蕴藏的各种暂态及稳态信息进行融合。动模仿真实验结果表明基于粗糙集理论的选线算法能更有效地解决小波包选线算法所出现的误选问题,且较模糊选线算法精度更高、抗干扰能力更强、运行更稳定。Aiming at the error selection issue of multi-band wavelet packet algorithm, which appears in view of the KA series of intelligent fuzzy line selection device, this paper puts forward a wavelet packet fault line selection algorithm based on the rough set theory, which uses rough set theory to conduct the data mining and knowledge discovery of fault sample set to determine the enhanced ratio of low-frequency attenuation signal. On the basis of that, it furtherly introduces the information fusion function of rough set theory to make full use of all the steady-state and transient-state information of single-phase-to-earth fault current. Dynamic simulation shows that the intelligent line selection algorithm based on the rough set theory, are not only more effective to solve the issue of the wavelet packet algorithm, but also much better than fuzzy theory in precision, anti-interference ability, and stable and reliable operation.

关 键 词：模糊理论 粗糙集理论 故障选线 信息融合 小波包分析 

分 类 号：TM755[电气工程—电力系统及自动化]