Research On Parameter Identification And Decoupling Control Of Permanent Magnet Synchronous Motor

With the characteristic of high torque inertia ratio,high energy density and high efficiency,Permanent magnet synchronous motor(PMSM)has been widely applied in the field of a servo system with high accuracy and high dynamic response.In a permanent magnet synchronous motor servo control system,permanent magnet synchronous motor is a nonlinear,strong coupling control an object.Accurate detection of electrical parameters and optimal design of controller is key technologies for high-performance control of permanent magnet synchronous motor.Therefore,this paper focuses on the parameter identification and decoupling control of permanent magnet synchronous motor.Firstly,aiming at the problem of low identification accuracy in parameter identification of traditional PMSM,a teaching and learning optimization algorithm based on Levy flight is proposed,and applied for PMSM parameter identification.The introduction of adaptive teaching stage teaching factor,it can adjust adaptively the convergence speed;In the learning stage introduce the Levy flight random process improvement optimization strategy,what’s more it is the introduction of self-learning strategy,global search and local exploitation ability of the algorithm.Through simulation experiments,we compared particle swarm optimization,teaching and learning optimization algorithm and improved teaching and learning algorithm to PMSM’s identification effect.It is found that the improved teaching and learning optimization algorithm have smaller identification error and faster speed.Secondly,aiming at the coupling problem of permanent magnet synchronous motor vector control current loop,a complex vector decoupling controller based on improved sliding mode observer is proposed,and the controller is applied to the decoupling control of permanent magnet synchronous motor.Based on the traditional sliding mode observer,the introduction of Sigmoid as a switching function and variable sliding mode gain reduce the high frequency chattering.Followed by construction of complex vector PI current regulator,combined with the transfer function of zero pole configuration and stability theory,determine the parameters of PI regulator.And then,through the information of improved sliding mode observer to estimate the rotor speed,estimate the speed of information based on complex vector PI decoupling controller decoupling control.The simulation research is carried out in two aspects of speed change and load torque mutation.Simulation results show that the controller can effectively decouple control of PMSM.Finally,for verifying the effectiveness of the controller,two cases of the speed change and the load torque change are studied with the ds PACE hardware-in-the-loop simulation platform.Using the DS1007 PPC controller board,the Simulink control model is converted RTW to the executable real-time object code(OBJ)and system description file(SDF).To manage the dSPACE hardware devices,ControlDesk software is employed,where experimental data are monitored and collected in real time online.By analyzing the experimental data,it is verified that the controller is effective for decoupling PMSM control.