An Efficient Cuckoo Search Algorithm for System-Level Fault Diagnosis  

作　　者：XUAN Hengnong ZHANG Runchi SHI Shengsheng 

机构地区：[1]School of Information Engineering, Nanjing University of Finance and Economics [2]Department of Computer Science and Technology,Nanjing University

出　　处：《Chinese Journal of Electronics》2016年第6期999-1004,共6页电子学报（英文版）

基　　金：supported by the National Natural Science Foundation of China(No.90718008,No.61133015)

摘　　要：We propose a new efficient algorithm named Cuckoo search fault diagnosis(CSFD) to solve system-level fault diagnosis problem. KMP algorithm is proposed for initialization based on the K-means partition algorithm; a fitness function is designed according to the equation constraints satisfied by the test model; the binary mapping method is advanced by optimizing existing binary mapping algorithm. Experiments show that KMP algorithm significantly reduces the disparity between the initial solution and the actual solution, and CSFD algorithm improves the efficiency and correctness significantly compared with existing typical swarm intelligence diagnosis algorithm.We propose a new efficient algorithm named Cuckoo search fault diagnosis（CSFD） to solve system-level fault diagnosis problem. KMP algorithm is proposed for initialization based on the K-means partition algorithm; a fitness function is designed according to the equation constraints satisfied by the test model; the binary mapping method is advanced by optimizing existing binary mapping algorithm. Experiments show that KMP algorithm significantly reduces the disparity between the initial solution and the actual solution, and CSFD algorithm improves the efficiency and correctness significantly compared with existing typical swarm intelligence diagnosis algorithm.

关 键 词：System-level fault diagnosis Equation model Cuckoo search K-means clustering Cuckoo search fault diagnosis algorithm（CSFD） 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP277[自动化与计算机技术—控制科学与工程]