Maximum Torque Per Ampere Control Of Permanent Magnet Synchronous Machines Based On Inductance Identifications

In recent years,due to the increasingly severe problem of environmental pollution and energy shortage,new energy electric vehicles are widely concerned for the advantages of energy saving and high efficiency.Permanent magnet synchronous motor(PMSM)is commonly used in the field of electric vehicle drive,because of its small size,high power density,high efficiency and reliable operation.For making full use of the reluctance torque of interior permanent magnet synchronous motor(IPMSM),the maximum torque per amper(MTPA)operation is often adopted.However,the parameters of PMSMs change with the effect of temperature and magnetic saturation,which affect the accuracy of the control system.Therefore,the online identification of parameters based on adaline neural network(ANN)is put forward in this paper,which is used to achieve the d-axis inductance in real time,and this parameter is introduced to the MTPA control system based on virtual signal injection control(VSIC).The MTPA operation based on online identification of inductance is realized finally,and this method does not depend on the design parameters of the motor and has good dynamic tracking performance and anti-disturbance performance.Firstly,the research background and significance of this paper are introduced briefly.Then,the research status of MTPA operation and parameters identification are elaborated,and the advantages and disadvantages of each method are analized and compared.The research content of this paper is put forward at last.Secondly,the basic structure and electromagnetic properties of PMSM are analized.The coordinate transformation matrixs are deduced based on the coordinate systems which are common used in three-phase PMSM.Besides,the mathematical models of three-phase PMSM are established in different coordinate systems,which lay the foundation for the following control system building.Thirdly,the realization principle of MTPA is elaborated,and the MTPA operation based on VSIC is adopted.The realizing process of VSIC is introduced,and the accuracy of this method in different application occasions is analized.When VSIC is applied to the lowreluctance-torque IPMSM,it is essential to acquire d-axis inductance in real time.The online identification of inductance based on ANN is proposed considering the time-varying parameters of motor,and the MTPA operation based on online identification of inductance is realized.Finally,the external hardware circuits are designed and set up to constitute real-time simulation system together with controller DS1103.Then,the electrical characteristics of DS1103 as well as the drive and protection mechanism of hardware are clarified,and the experimental verifications are completed.The experimental results show that the proposed MTPA control baesd on online identification of inductance can identify the inductance in real time and find the optimal current angle in this paper.