Research On The Sound Signal Fault Detection System Of Micro-motor

The motor is the most widely used power equipment.The motor quality inspection is an important part of ensuring the smooth operation of the motor and the mechanical system.Traditional motor testing relies on workers’ listening methods to identify whether the motor is faulty,with low efficiency and poor consistency.The test results are greatly affected by human subjective factors.Because there is no uniform objective evaluation standard for motors and reasonable diagnostic testing methods,it is difficult to realize automatic detection.This article takes the micro-motors manufactured in batches on the motor manufacturer’s production line as the research object.Aiming at the actual engineering problems of the automatic detection of the motor,the sound signal is collected,the feature extraction and fault detection are carried out.The main research contents of this paper are as follows:(1)Analyze the internal structure characteristics of the micro-motor,and obtain the signal cycle law of the micro-motor.A micro-motor abnormal sound detection platform was built,and the working signal of the motor was collected.Comparative experiments verified the feasibility of using sound signals to achieve non-contact measurement.(2)According to the experience of artificial listening,combined with the human ears perceive the loudness,pitch,and timbre of the sound,a REFD combined evaluation model based on signal feature extraction is proposed,using the RMS,short-term energy entropy,and wavelet packet Decompose the objective data of the sub-band peak value and energy distribution characteristics,and distinguish the mechanical abnormal sound fault and electromagnetic abnormal sound fault of the micro-motor.(3)Using the traditional SVM classification model as a benchmark for comparison,the PSO-SVM classification model and the random forest classification model are used to distinguish the types of motor faults.Random experimental tests show this model performance of the classifier has improved significantly.(4)Based on the motor feature extraction method and the fault classification prediction model,using the Lab VIEW software platform,a micro-motor fault detection system integrating acquisition,analysis and diagnosis functions was developed to realize the sound signal collection and fault diagnosis classification of the micro-motor.