Study On The Approach Of Parameter Identification For ISG PMSM In Hybrid Electrical Vehicle

The contents this paper studying is from the project endowed of Jilin Scientific Department, the project is named as'Reach on the controller of the single axis motor in hybrid electrical vehicle for the 1st Automobile Works'. In the project the ISG motor control system adopts a permanent magnet synchronous motor with a direct control system. The purpose of this paper is to improve the operating ability of the ISG motor in order to fulfill the demand of the project, as the same time, it has academic value.In this paper, firstly the mathematical model of the PMSM is introduced, for easy static calculating and dynamic analyzing, a 2-axis graph is defined, and the transformation method between the 2-axis and 3axis graph is derived, on the principle of it, the mathematical model of the PMSM in 2-axis graph is deduced. Then it introduces the theory basic of the direct control system and the achievement of the ISG motor control system in the project. For the sake of improving the performing of the motor control system, the main factors which affect the motor performing are analyzed. It is pointed that the accuracy of the motor parameters one of the most important factor in this main factors, and theoretical derivation and experiment are prove the conclude. Consequently, for improving the performing of the motor, methods to improve the accuracy of the parameters shall be investigated, in another word, parameter identification shall be done.In researching the method of motor's parameter identification, the paper utilizes the classical Least Square Method according to the system identification theory, and designed 2 series least square parameter identification methods. The 1stone is for the identification of the rotor flux of magnet and inertia, it is derived from the accurate modal of the PMSM, and the iterative algorithm can be carried on in any motor operating condition. Simulation experiment proves that the methods has high convergence rate, and it can accurately identifies the actual value of the rotor flux of magnet and inertia in any motor operating condition. But this method needs complex data, some of them are not easily measured, sensors and circuit should be designed, it goes against the engineering actualization. The second method is for the identification of the stator resistance and inductance, it is derived from the approximate modal of the PMSM, the iterative algorithm needs motor speed arriving steady. It only requires motor's current and voltage, being propitious to the engineering actualization. Simulation experiment proves that the methods can comparatively accurately identifies the actual value of the stator resistance and inductance in the condition that motor speed is comparatively steady. But it has little accuracy than the 1st method, and the accuracy may be even decline when the motor speed goes less steady.In order to conquer the defect of the Least Square Method, designing a method which is propitious to the engineering actualization and can adapt the instability of the speed control, this paper studies the designing method of the model reference adaptive parameter identification system according to the Positivity and Hyperstability theory. On the principle of it, a model reference adaptive parameter identification system for stator resistance and inductance is designed. The system needs an initial value which accesses the actual value of the stator resistance or inductance, and requires motor's current and voltage as the input of the system when motor speed comparatively arrives steady. The output matrix of the system will be convergent at some function of stator resistance and inductance, and the identification value of the ISG motor'parameter will be solved from the equations. Simulation experiment proves the efficiency of the designed model reference adaptive parameter identification system, it can get the identification value of the stator resistance or inductance quite accurately in the condition that motor speed comparatively goes steady. The system has a longerconvergence time than the Least Square Method, but it accords with the engineering demand and is capable of an online parameter identification method.According to the designs of this paper, an experiment adopting the series least square parameter identification method for stator resistance and inductance is carried on the ISG motor experimental system. Then we made the identification values from the experiment as the calculated values which are calculated to work out the control signal in the control system. Comparing with the control system which made the parameter on the motor's nameplate as the calculated values, the ISG motor's response time and the capability of speed control are improved, especially in the low speed condition. The experimental result accords with the conclude which is made from the analysis of the factors which affect the performance of the PMSM, and it also proves that the parameter identification method which is designed in this paper will make improvement for the ISG motor.The research of this paper fulfills the demand of the project, besides that, it has academic value for the research of correlated problem. They mainly incarnate that, it made thorough investigate on the direct toque control system, pointed out the main factors that affect the performance of the system and its causation, it has theoretical value for the improvement of the direct toque control system. It extended the Least Square Method for induction motor to the PMSM, and made a design method according to the approximate model aiming at the engineering practicality, it made the LS Method to be used in accrual project. It designed a model reference adaptive parameter identification system adapt to the PMSM direct toque control system, it can be the basic of identification method of other parameter and the research of non-sensor technology.