Instantaneous Power Analysis And Research On Fault Diagnosis Of Mine Motor Rotor

In the coal industry,the motor as a driving device is widely used in crushing machine,screening machine,scraper machine,boring machine and other large equipment.Therefore,it is very important to timely monitor the running status and fault type of the mining motor,it not only related to the safe operation of coal mining coal system,more related to the personal safety of staff.In this thesis,the rotor broken bar and eccentric fault,which are the most common fault of the motor in the mine,as an example.When the motor rotor fault occurs,the corresponding fault characteristic frequency will be generated in the line current at both ends of the stator,the change of stator characteristic frequency is reflected in the change of rotor currentcharacteristic.But the motor slip is small,the fault characteristic frequency of the rotor current will be submerged by the fundamental frequency.Therefore,the line voltage corresponding to the rotor current at both ends of the motor is calculated,and then the instantaneous power of the motor rotor is obtained.Using the wavelet packet analysis method to analysis the variation law of instantaneous power for motor rotor,the wavelet packet decomposition and reconfiguration are applied to the instantaneous power spectrum of motor rotor,then the extracted energy values of different frequency bands are as the characteristic quantity to identify the fault of the motor rotor.In this thesis,some extracted fault characteristics quantity about different frequency bands of motor rotor may exist redundancy,repetition and uncertainty,these fault characteristics quantity not only affect the diagnostic speed,but also reduce the accuracy of fault diagnosis.Therefore,in order to reduce the redundant data to get the best data samples,the feature samples should be pretreated by the reduction theory of rough set theory before the fault diagnosis and classification of data samples.Because mine crusher motor bad working environment,the data can not be extracted in large quantity.In order to improve the accuracy of fault diagnosis,the rough set algorithm is introduced to preprocess the characteristics quantity of the sample,then data samples are reduced.Therefore,in this thesis,the support vector machine(SVM)algorithm which can be used to diagnose and classify small samples and nonlinear data based on the principle of structural risk minimization is proposed to realize fault diagnosis and classification of motor rotor.The simulation results show that the fault characteristic quantity which has been treated by rough set is more accurate.