Based Differential Evolution K-means Algorithm for Fault Clustering on Flight Control System
We have designed a differential evolution K-means (DEK) scheme for fault classification of the flight control system. This method combined K-means clustering method with differential evolution algorithm (DE)in the presence of initial value sensitiveness and prematurity with K-means and FCM algorithms. Differential evolution is a floating-point encoding evolutionary algorithm for global optimization over continuous space. Using the robustness and efficient global search capability of differential evolution algorithm, this method can overcome the shortcomings of k-means and FCM algorithms. Through fault clustering analyzing, this algorithm can accurately fulfill failure recognition. The proposed scheme is applied to flight control system failures of an aircraft altitude holding model. The results show that the proposed method can effectively classify failures.
differential evolution k-means clustering algorithm FCM fault clustering
GU Wei ZHANG Weiguo HUANG Zhiyi LI Lili
College of Automation Northwestern Polytechnical University, Xian 710072, China
国际会议
第八届国际测试技术研讨会(8th International Symposium on Test and Measurement)
重庆
英文
1586-1590
2009-08-01（万方平台首次上网日期，不代表论文的发表时间）