Research On Crack Diagnosis Of Traveling Wave Piezoelectric Ceramics In Ultrasonic Motor Based On Time-frequency Analysis

As a new type of motor,the ultrasonic motor is widely used in high-tech machinery and equipment in high and new fields relying on its own structure and unique operation mode.Piezoelectric ceramics,as the transducer of ultrasonic motors,play a key role in converting electrical energy into rotor kinetic energy.However,due to the brittleness of materials,long-term operation under high-frequency driving power will lead to fatigue damage and cracking failure.Fault diagnosis of piezoelectric ceramics can prepare for the failure of the ultrasonic motor function and avoid major accidents.Therefore,the research on piezoelectric ceramic fault diagnosis for ultrasonic motors is of great significance to ensure the safe and stable operation of ultrasonic motors.Based on the working principle of ultrasonic motor and the cracking process of piezoelectric ceramics,a fault diagnosis method is proposed,and selects the monitor electrode voltage(MEV)as the condition monitoring signal of the piezoelectric ceramics.The signal processing and degradation feature extraction of MEV are mainly studied.The data collected by the ultrasonic motor test bench analyzed and verified the above research content.The main research contents and innovations include the following three aspects:1)The influence of piezoelectric ceramics cracking on the soliton voltage is analyzed,and an ultrasonic motor acquisition platform is built.By characterizing the MEV with different cracking degrees in the time and frequency domains,signal characteristics of different fault degrees are obtained.2)For deep-level mining of the fault information contained in the MEV,as well as analysis and processing of the fault information.An adaptive signal processing method based on Local Mean Decomposition(LMD)was proposed for the analysis and processing of the MEV,which utilizes its excellent adaptive separation characteristic in the frequency domain.The product function reflecting the vibration state of the stator is separated from the high-frequency band to the low-frequency band in turn.The product function is converted and calculated to obtain the piezoelectric ceramic fault characteristics in the time-frequency domain,which reflects the degradation degree of the piezoelectric ceramics,and it is applied to the analysis of the measured signals.The result show that the extracted degradation features can generally distinguish the different degrees of cracking failure of thepiezoceramics.3)Aiming at the fact that the MEV is not sensitive to the change of cracking failure of piezoelectric ceramics,a new piezoelectric ceramics fault diagnosis method is proposed.Discrete Cosine Transform(DCT)with energy accumulation characteristics is used to extract fault information and improve the correlation with cracks.In order to avoid the influence of human factors on the traditional DCT noise reduction method,a matrix composed of discrete cosine coefficients is constructed,and the singular values ??(SV)cross entropy of the matrix is ??used as the degradation characteristics of the ultrasonic motor.The numerical simulation noise is added to the measured signal to verify the anti-noise performance of the feature.The analysis of the experimental results shows that the proposed DCT-SV cross entropy is feasible and effective in indicating the degradation of piezoelectric ceramics in ultrasonic motors.