Motor Fault Diagnosis Based On Neural Network

As an important device in the electrical and mechanical system,motors are widely used in various fields of industrial production with therapid development of science and technology as well as the continuousimprovement of factory automation. Motor failure will has largeinfluence on the order of production, and even cause damage to the safetyof persons and property. Therefore, it is particularly important to ensurethat the motor running in the normal state. Our country and abroad havealways attached great importance to this aspect. How to find the fault inearly stage has always been very important research topic.Generally, when the motor is running, the faults will display in someways, such as the change of the current of the motor, vibration on thesurface of the machine, the abnormal temperature rising, the changes ofamplitude and frequency in audio and etc. These signals are acquired andprocessed in some ways. And then, with the application of the expertknowledge and various analytical methods, we can get the conclusionwhether the motor faults exist.This dissertation presents an idea based on the combination of wavelet packet and neural network in order to identify the specific motorfailure and defects of the components. Firstly, we use FFT to process theaudio signal, get the different motors`spectrum of the same type. At thesame time, a preliminary judgment is made on the basis of experience.Next we apply the wavelet transform on the signal and observe the lowband spectrum, then make the faulty judgment and select the featuresignal based on the expert knowledge. Furthermore, we use improvedwavelet packet transform to deal with the audio signal, collect the dataaccording to characteristic band. Finally, we do the training of the neuralnetwork with the data, achieve the function of detection the specificmotor failure.