Characteristic Component Amplitude Analysis And Fault Severity Evaluation In Induction Motors Using The Model-based Technique

Induction motors,especially three-phase squirrel cage induction motors(SCIMs),have found wide applications because of their low cost,simple and rugged construction,wide range of speed,and high reliability.Therefore,the online condition monitoring of induction motors becomes very important.Motor current signal analysis(MCSA),a non-invasive method for condition monitoring and fault diagnosis by capturing the motor stator current at fault-related frequency,has attracted wide attention in the field of induction motor fault diagnosis.However,the interpretation of the characteristic current amplitude is not clear,and the evaluation of the motor condition is lack of scientific basis,which make MCSA can not be widely used in practical industry.Using model-based technique,the induction motor model under radial vibration is established,an approximate expression of characteristic component amplitude is derived,and corresponding evaluation method is proposed to solve the above problems.Accurate modeling of induction motor under radial vibration is the key to solve the studied problem.Herein,the rotor motion model of induction motor under various mechanical fault types is presented and the model of induction motor under fault is established under the background of the existing classical winding theory,winding function method,air gap permeability method and other motor modeling methods.For the convenience of analysis,this paper simplifies the multiple coupled circuit model(MCCM)based on the classical winding theory,and derives the general expression of the stator characteristic current generated by rotor vibration.Based on this expression,the influence of supply voltage level,voltage harmonics,load level and other factors on the amplitude of characteristic current is analyzed and verified by simulation.Based on the analysis results,two methods of fault degree evaluation are proposed in this paper: the evaluation method based on vibration severity and the evaluation method based on characteristic current.The experimental results presented for condition monitoring in a real YB-180M-2induction motor clearly validate the two evaluation methods.In the evaluation method with vibration intensity as the index,this paper uses particle swarm optimization(PSO)algorithm to identify the parameters of motor fault degree,and the identified equivalent vibration severity is evaluated according to the existing mechanical vibration standard.In the evaluation method with characteristic current as the index,this paper converts the existing mechanical vibration standard into characteristic current amplitude standard through the proposed model,and directly evaluates the collected characteristic current amplitude.Both methods are highly consistent with the results measured by the vibration measurer,which indicates that the fault model and evaluation method proposed in this paper are reasonable and effective.