Research on fault detection of small and medium-sized rotating equipment based on FCM algorithm
To solve the problems of slow convergence speed and low fault recognition rate in the fault detec-tion of small and medium-sized rotating equipment by existing algorithms,a fault detection scheme based on FCM fusion algorithm is proposed.The original fault set is denoised,and the membership degree of the fault vector is extracted under the multi-scale condition based on the fuzzy entropy value theory.The itera-tive performance and convergence performance of the FCM algorithm are optimized by the GA algorithm,and the fuzzy membership degree matrix and clustering performance of the fault eigenvector are updated re-spectively,all of which are used to improve the clustering accuracy and the fault identification rate.The ex-periment results show that the algorithm can obtain better clustering effect and higher convergence speed un-der the condition of different number of cluster centers and fault types,and the average fault identification of training set and test set can reach 99.19％and 98.23％respectively.
万方数据
维普
