| Permanent Magnet Synchronous Machine(PMSM)has the advantages of high efficiency and power density,and it has been widely used in industrial servo drive,electrical vehicle,and renewable energy generation system.Due to the influence of temperature rising and magnetic circuit saturation,the parameters of PMSM including stator resistance,rotor flux linkage and d-,q-axis inductance are changed with the time.These parameters need to be acquired instantly in order to control PMSM better.In this thesis,the core problem of parameter estimation-"rankdeficient" is discussed and two solutions to it are proposed.A motor drive platform with ARM as the controller is set up,and the control algorithm is verified by experiments based on this platform.The two methods of parameters estimation are as follows:Firstly,a parameter estimation method based on step-pulse current injection is proposed.The stator resistance,rotor flux linkage and d-,q-axes inductance were identified by forgetting factor recursive least squares algorithm(FFRLS)without obtaining any nominal parameters.The change of inductance caused by current injection makes the estimated parameters difficult to converge.At the same time,online parameter estimation brings heavy burden to the operation of the control system.To solve these two problems,the inductance model is linearized.Moreover,the step-by-step parameter estimation method is adopted and orders of feedback matrix are reduced,which improves the computation efficiency.The minimum amplitude and appropriate frequency of step-pulse are determined by the parameter error and convergence analysis.In order to achieve the fast and accurate convergence of the identification algorithm,the condition of Persistence of Excitation(PE)is discussed.At the same time,the nonlinear factors of inverter are compensated to increase the precision of parameter identification.At last,a simulation model based on step-pulse current injection was established,and the effectiveness of the identification method was verified.Secondly,a parameter estimation method based on square-wave angle injection is proposed.By using FFRLS,the rotor flux linkage and stator inductance of SPMSM are identified without obtaining any nominal parameters.Then,the principle of eliminating the influence of inverter nonlinear factors on parameter identification is analyzed,and the influence of magnetic circuit saturation is also considered.The parameter identification method based on square-wave angle pulse injection and that of current injection method would cause inductance change during injection,and the compensation method of changed inductance is proposed.Then,the minimum amplitude and appropriate frequency of square-wave angle are determined by the parameter error and convergence analysis.At last,a simulation model based on square-wave angle injection was established,and the effectiveness of the identification method was verified.Finally,the hardware of the power and control circuit of the driver control system based on ARM is designed,and the overall structure of the control software is constructed.The motordriver experimental platform was built and debugged.In addition,the effectiveness of the proposed estimation methods based on step-pulse current injection and square-wave angle injection are verified via the experiment platform. |
PDF Full Text Request