Application Of Least Squares Method In Parameter Identification Of Delay Circuit And Motor Model

As an important branch of modern control theory,identification has been widely used in many fields such as practical engineering,environmental science and biomedicine.In delay differential equation model,the nonlinear relationship between delay and other model parameters increases the difficulty of identification and reduces the control effect.Therefore,the research on parameter identification method of delay differential equation model has important theoretical significance and application value.The identification algorithm of delay and parameters for a class of linear delay differential systems is derived based on the nonlinear least square method.The parameter identification of the delay circuit model and the table-type permanent magnet synchronous motor model is studied.The main work of this paper can be divided into the following three parts:In the first part,the parameter identification algorithm of linear delay differential system with unknown delay is derived.The identification problems include time delay estimation and system parameter estimation.Firstly,the problem of delay and parameter estimation is transformed into a nonlinear least square problem with forgetting factor,then the delay estimation is obtained by Gauss-Newton method,and finally the system parameter estimation is obtained based on the least square method.In the second part,the derived identification method of delay and parameters is applied to the actual circuit model,and the input and state values in the actual circuit model are measured by oscilloscope.The voltage delay and the parameters of each electronic component in the delay circuit model are identified by the measured data.The experimental results show that the linear differential equation model with delay is scientific and the identification method is effective.In the third part,the parameter identification of surface-mounted permanent magnet synchronous motor is studied.Firstly,the state space model of SPMSM is established,and the random term is introduced into the model to make it closer to the real system.The Euler-Maruyama method is used to discretize the motor model.Then the numerical solution of the motor model is obtained to simulate the measured data.At last,the system parameters of SPMSM are identified online by recursive least square method with forgetting factor.Numerical results show that this method can identify d-q inductance,stator winding resistance and permanent magnet flux value in SPMSM system at the same time.