Research On Predictive Control System Of Permanent Magnet Flux-switching Motors Based On Parameter Identification

Permanent magnet flux-switching motors(PMFSM),as a new type of permanent magnet motor.Because of its high power density,no demagnetization problems,and diverse control methods,in the past two decades,it has always been a research focus in the field of permanent magnet motors.Due to the short development time of PMFSM,in order to further enhance the control capability of PMFSM motors,this thesis takes motor model predictive current control(MPCC)as the research topic.The main research elements are as follows:This thesis first expounds the research status of PMFSM,establishes a mathematical model according to the basic working principle of PMFSM,and applies Ansoft software to analyze the electromagnetic performance of its ontology,and gives its basic control mode.After applying the Euler discrete method to obtain its prediction model,the traditional MPCC algorithm is further studied.The voltage vector of traditional three-vector MPCC mostly adopts traversal optimization,which is too large in calculation and the voltage vector selection range is too limited,which reduces the problem of steady-state performance of the control system.To solve this problem,this thesis proposes an improved three-vector MPCC algorithm,which determines the first preferred voltage vector by locating the sector where the desired voltage vector is located,Substitute the two effective voltage vectors of the sector into the value function to determine the first preferred voltage vector,and determine the second preferred voltage vector by locating the sector where the voltage error vector is located.and only four current prediction calculations can synthesize the optimal voltage vector.The simulation results show that the improved three-vector algorithm proposed in this thesis not only expands the selection range of the second optimal voltage vector,improves the stability of the control system,but also reduces the computational complexity and the requirement of the algorithm on the hardware system.Secondly,considering the mismatch between the actual motor parameters and the predicted model parameters and the PMFSM operating conditions,this will affect the system control performance.This thesis proposes to use model reference adaptive algorithm to identify parameters of PMFSM motor in real time,and then compensate and correct the prediction model in the three-vector MPCC.The simulation results show that the parameter identification algorithm based on model reference adaptive can accurately identify PMFSM motor parameters,which can effectively compensate for the prediction error and improve the robustness of the system.Finally,with the DSP chip TMS320F28335 as the core,an experimental platform was built to design related circuits and algorithm programs.The simulation and experimental results show that the proposed PMFSM three-vector strategy based on parameter identification can significantly improve the steady-state performance of the system and reduce the amount of computation,remove the adverse effects of parameter mismatch,and the sensitivity of the three-vector MPCC to parameters is suppressed.