Research On Fault Diagnosis Method Of Induction Motor Based On VMD And SVM

In order to meet the needs of the current industrial development,asynchronous motor equipment towards large-scale,large-scale,automated development.Once the fault occurs,the traditional intrusive troubleshooting method needs professional to identify the fault type through experience,and it may lead to new fault when evaluating the fault point.Therefore,it is very important for industrial development to explore non-invasive intelligent fault diagnosis method for induction motor.Secondly,due to the non-linearity and non-stationarity of the fault signal of the induction motor,and the characteristics of less retention and small sample,the traditional methods of spectrum analysis and fault diagnosis can not accurately extract the fault features of asynchronous motor fault signals and make fault diagnosis.Aiming at the above problems,this paper studies the current signals of two kinds of common electrical faults in induction motors.In this paper,Variational modal decomposition(VMD)is combined with an improved Support vector machine(SVM)to construct a fault diagnosis model for induction Support vector machine.In this paper,the mechanism of stator winding inter-turn short circuit and rotor broken bar fault in induction motor is analyzed,and two kinds of electrical fault simulation models are established,the current fault characteristics of induction motor with two kinds of faults are analyzed.Secondly,the experiment platform is set up to collect the stator current signals of induction motor when two kinds of faults occur,and the measured data of two kinds of electrical faults are decomposed into many intrinsic mode functions by VMD,the fault characteristic frequency band is obtained by FFT analysis after signal reconstruction.Then,the energy of 5 characteristic frequency bands is extracted as fault features and compared with different classifiers.At the same time,the diagnosis effects of different fault feature extraction methods on stator winding inter-turn short circuit fault and rotor broken bar fault are compared.According to the characteristics of small sample of fault data of induction motor,SVM is used as the classifier of fault diagnosis based on the previous research,but SVM is greatly affected by the hyperparameters,therefore,a multi-population intelligent optimization algorithm is used to optimize the super-parameters of the Support vector machine,and the optimal diagnosis model is obtained.After analyzing the advantages and disadvantages of the algorithm,the chaotic Tent mapping and the adaptive weight factor strategy are introduced to optimize the algorithm to obtain the IAO algorithm,the performance of the IAO algorithm is tested on the benchmark function.Then IAO is used to optimize the hyperparameters of SVM,and the IAO-SVM fault diagnosis model is obtained,finally,the fault characteristics of 13 states under different operating conditions,such as normal state,stator winding inter-turn short-circuit fault and rotor broken bar fault,are input into the model.The test results show that the average classification accuracy can reach 94.62%,which verifies that the proposed diagnosis model has good recognition ability.Compared with other diagnosis models,the results show that the method has high classification accuracy under the same condition,different fault,same fault,different condition and so on.Figure [47] table [17] reference [82]...