基于改进FCM算法的电力数据异常检测方法  

Power abnormal data detection method based on improved FCM algorithm

作　　者：王璞 谢晓娜[2] WANG Pu;XIE Xiaona(Guodian Dadu River Drainage Area Hydroelectricity Development Co.,Ltd.,Chengdu 610041,China;School of Automation,Chengdu University of Information Technology,Chengdu 610225,China)

机构地区：[1]国能大渡河流域水电开发有限公司,四川成都610041 [2]成都信息工程大学自动化学院,四川成都610225

出　　处：《电子设计工程》2023年第24期33-37,共5页Electronic Design Engineering

基　　金：四川省自然科学基金资助项目(2022NSFSC0500)。

摘　　要：电力系统中的异常数据会极大降低电力系统运行状态估计的准确性,提出一种改进FCM聚类算法应用于电力异常数据检测。基于距离测度公理化定义给出一种新型距离测度计算公式,并根据新型距离测度建立相似性矩阵提出一种改进FCM算法;结合萤火虫算法在全局寻优方面的优势,利用萤火虫算法优化改进FCM算法的初始化聚类中心。通过加噪人工数据集实验验证了该算法与其他方法相比,类别划分更清晰、噪声鲁棒性更强,并依据3σ原理利用该算法对某电厂进行电力异常数据检测,实验结果表明,文中算法能够准确检测出电力异常数据。Abnormal data in power system can greatly reduce the accuracy of power system operation state estimation,an improved FCM algorithm is proposed to detect abnormal data in power system.Based on the axiomatic definition of distance measure,a new distance measure calculation formula is given,and an improved FCM algorithm is proposed by using the new distance measure to establish the similarity matrix.Combined with the advantages of firefly algorithm in global optimization,firefly algorithm is used to optimize the initialization cluster center of the improved FCM algorithm.Experiments result of artificial data sets with noise show that this algorithm is clearer in classification and more robust than other methods.Based on the 3σprinciple,this algorithm is used to detect abnormal data of a power plant.The experimental results show that the algorithm can accurately detect abnormal power data.

关 键 词：距离测度 异常数据 FCM 萤火虫 鲁棒性 

分 类 号：TN98[电子电信—信息与通信工程] TP393[自动化与计算机技术—计算机应用技术]