Remaining Life Prediction Of Ultrasonic Motor Based On Elman Neural Network Optimized By Particle Swarm Optimization

Based on the inverse piezoelectric effect of piezoelectric ceramics and the frictional motion of the stator and rotor,ultrasonic motor converts electrical energy into mechanical energy to generate driving torque.The useful life of the ultrasonic motor is related to the safe operation of the motor under various working conditions.Accurate prediction of the life of the motor can provide reliable data support for the predictive maintenance of the ultrasonic motor and ensure the long-term stability of the equipment.At present,the existing methods of ultrasonic motor life mostly focus on the prediction of friction material life of the motor,there is a lack of research on the life evaluation of ultrasonic motor.In view of the problems existing in previous studies,this paper establishes an ultrasonic motor life prediction method based on motor state monitoring data and IPSO-Elman neural network.Related research contents are as follows:1.Based on the principle of PMR60 traveling wave rotating ultrasonic motor,the influence of motor performance degradation and remaining life reduction on motor state monitoring data is studied,and the degradation characteristic parameters that can represent motor life are determined by analyzing the state monitoring data.2.According to the requirements of ultrasonic motor life test experiment,the ultrasonic motor life test device is designed,and the upper computer of instrument control and data acquisition is developed based on Lab VIEW,so as to realize the loading,control of ultrasonic motor and collection,processing and storage of state monitoring data.3.The life prediction model proposed in this paper takes Elman neural network as the core and uses IPSO algorithm to optimize the neural network.The prediction model takes the characteristic parameters of motor performance degradation as input and the prediction time of remaining life as output.Two prediction models,Elman and PSO-Elman neural network,is introduced to verify the model prediction effect.The experimental results show that the prediction model proposed in this paper has good accuracy in predicting the remaining life of ultrasonic motor,which verifies the effectiveness of the method.To sum up,this paper takes the state monitoring data of ultrasonic motor as a new breakthrough point and integrates the artificial intelligence algorithm framework.Compared with the traditional prediction method,the proposed new life prediction method can realize the life prediction of the whole ultrasonic motor,which has practical engineering significance.