Research On Cracking Fault Identification Of Piezoelectric Ceramics Based On Mathematical Morphology Analysis

As a new type of motor,ultrasonic motor is often used in modern sophisticated equipment because of its special mechanical structure and operation principle.However,as the piezoelectric ceramic in the motor is a brittle element,it is easy to crack when working under high frequency excitation for a long time.In order to avoid the occurrence of serious accidents,this paper firstly proposes a degradation state recognition algorithm of piezoelectric ceramics for ultrasonic motor based on the boundary span analysis of mathematical morphology.This algorithm reduces the dependence on the state eigenvalues in the traditional fault recognition algorithm,so that the algorithm has a high accuracy of degradation state recognition when the state eigenvalues discrimination is small.However,the algorithm can not completely avoid the dependence on the state eigenvalues,so this paper proposes a segmented fractal dimension sparsity representation classifier,this algorithm mainly uses the characteristics of signal change law to recognize the degradation state of piezoelectric ceramics.The algorithm effectively solves the problem of false recognition brought by the extremely small discrimination of state eigenvalues.However,there is no rule to follow for the fault characteristics of piezoelectric ceramics in the failure state,so the algorithm has the phenomenon of false recognition in the failure state.In order to solve the disadvantages of the two algorithms mentioned above,this paper proposes a gray association selector algorithm based on information entropy,which can effectively distinguish the advantages and disadvantages of the two algorithms.Therefore,in the process of degradation state recognition,the effective degradation state recognition results can be dynamically selected as the final recognition results.Finally,the experimental analyses shows that the algorithm is effective,and the algorithm still has a very high accuracy of degradation state recognition in the case of high noise interference.