应用数据融合的电网故障诊断  

Novel Fault Diagnosis Approach of Power Grid Using Data Fusion

作　　者：徐岩[1] 张锐[1] 卫波[1] 

机构地区：[1]华北电力大学电气与电子工程学院,保定071003

出　　处：《电力系统及其自动化学报》2014年第11期67-71,共5页Proceedings of the CSU-EPSA

摘　　要：电网发生故障时,由于保护及断路器的拒动、误动等因素使得诊断结果存在不确定性。首先提出了应用数据融合的电网故障诊断方法;再针对融合故障发生时的开关量和电气量信息,利用模糊Petri网对开关量信息进行分析,利用希尔伯特-黄变换HHT(Hilbert-Huang transform)对电气量进行分析,提取开关量故障度、幅值故障度、频率故障度和能量故障度,并利用D-S证据理论对各种故障表征予以融合,进而得故障诊断结果;最后,通过仿真验证所提方法能充分利用电气量信息,减小单一开关量的不确定性因素的干扰,提高故障诊断的精度。When fault happens in the power grid,the diagnosis result is uncertain sometimes because of the mal-operation and miss-operation.Firstly,the power gird diagnosis based on data fusion is proposed in this paper.Then,fused the digital and analogue data,the Petri fault degree is obtained by fuzzy Petri nets.By Hilbert-Huang transform （HHT）,the amplitude fault degree,the frequency fault degree and the energy fault degree are obtained.The D-S evidence theory is employed to fuse the degrees.Finally,the research is proved to be available by the application of some cases in which the uncertain factors could be eliminated and thus,the accuracy of the fault diagnosis could be improved.

关 键 词：数据融合 故障诊断 希尔伯特-黄变换 模糊PETRI网 D-S证据理论 

分 类 号：TM711[电气工程—电力系统及自动化]