Broken Rotor Bar And Eccentricity Fault Diagnosis Method Based On Parameter Identification Of Induction Motor

Induction motors are used in many aspects of industry and agriculture,national defense,transportation and life.At present,the society is vigorously advocating the use of highefficiency and energy-saving motors,the failure or defect of motors will not only cause abnormal operation of production equipment,resulting in suspension of the production process,but also cause the motor to run in a low and medium energy efficiency state before it turns into a fault and shutdown,resulting in additional energy loss.If the abnormality of the motor can be found in time and future state of the motor can be predicted to repair the motor or adjust the production plan as early as possible,the loss can be minimized.Statistical data on the occurrence probability of various types of faults in induction motors shows that the total probability of broken rotor bars and air gap eccentricity is about 22%.Based on the idea of parameter identification,this paper derives the broken rotor bars and air-gap eccentric fault models of induction motors,which are used for early fault diagnosis of motors.The multi-loop model of induction motor provides convenience for broken rotor bars and air gap eccentric fault modeling.In this paper,the multi-loop model is used to analyze the steady-state characteristics of the stator and rotor current under normal induction motors,broken rotor bars and air gap eccentricity.In order to solve the problems of multi-loop model with many parameters,large amount of calculation,and insignificant parameter change characteristics in the early stage of the fault,the transformation matrix is constructed based on the idea of coordinate transformation to realize the transformation of the multi-loop model.The broken rotor bars and air gap eccentric fault models obtained through transformation are derived in detail.Based on the fault model,the equation form is unified with the normal model,and the relational expression between the equivalent parameter and the input three-phase voltage initial phase angle is analyzed,and the equivalent parameter characteristics in the fault state are extracted to diagnose motor operating conditions.A 11 kW motor with a constant torque load and a 18.5k W motor with a constant power load are used for example verification.The motor fault data is obtained through the multi-loop model,and the particle swarm-simulated annealing alternating algorithm is used to apply the motor fault data to the fault model for parameter identification.The relationship between equivalent parameters of fault characteristics and fault types is analyzed in detail,and the feasibility of parameter identification based on the motor fault model derived in this paper to diagnose early faults of induction motor is proved.Finally,the experimental verification is carried out through the 22 k W motor data.