Research On Parameter Identification Of Asynchronous Motor

The induction motor with simple structure and reliable operation is widely used in the industrial and agricultural fields.The induction motor has always occupied an important position in the field of AC variable transmission.The field-oriented control of induction motor can theoretically realize the complete decopling control of the torque component and the motor excitation component,thereby achieving the same excellent control performance as DC motors.However,the motor parameters required for field-oriented control always change with factors such as motor temperature and excitation current,making the actual control effect difficult to predict.In severe cases,it even leads to loss of control.Therefore,the parameter identification problem of the motor is one of the key technologies to further improve the motor control performance.Firstly,the paper establishes the mathematical model of three-phase induction motor.On this basis,the basic principle of field-oriented control of induction motor is introduced,providing a theoretical basis for the parameter identification method described later.The mathematical model of induction motor in the three-phase stator reference frame is not clear about the physical nature of torque generation,which is not conducive to the realization of specific control.Vector control based on rotor field orientation in synchronous rotating coordinate system,changing the nonlinearity of torque control,using coordinate transformation,making induction motors have good control performance.The parameter identification of induction motor is divided into offline identification and online identification.The purpose of offline identification is to provide a good initial value for online identification.The results of online identification can better adapt to the real-time transformation of parameters and ensure the control performance of the motor.No-load experiment and stall test are traditional offline parameter identification methods for induction motors,but this method requires specific external load conditions and is not suitable for widely applications.Therefore,this paper proposes a new method: Using the inverter's own resources to control the application of specific excitation signals to the induction motor through space vector pulse width modulation(SVPWM)technology,the response signal is processed to obtain more accurate motor parameters,using the principle same to the traditional experimental method.In terms of online identification,this paper proposes a model reference adaptive identification method,the rotor resistance is identified based on the reactive power model,the stator resistance is identified based on the rotor flux squared model,which can achieve complete decoupling of the two identifications.However,in practice,the digital delay will deviate from the recognition results.After quantitative analysis,this paper proposes a simple and easy compensation scheme,and verifyes the correctness of the analysis through simulation experiments.