Study On The Torque Control Of 4WD Electric Vehicle Drive System With Permanent Magnet Synchronous In-wheel Motors

With energy and environmental issues becoming increasingly prominent, pure electric vehicle is widely considered as one of the effective solutions to the transport due to its energy-saving, pollution-free characteristics. Compared with other types of pure electric vehicle, Four-Wheel-Drive Electric Vehicle, thanks to its compact structural attribute, higher drive system efficiency and better drivability has recently became a hot topic for domestic and international researchers. It points a direction for future automobile technology study as well.As the sole power source of the four wheel drive electric vehicle, Permanent Magnet Synchronous in-wheel Motor has a significant impact on vehicle’s power, economy and security performance. Therefore, research on the Permanent Magnet Synchronous Motor(PMSM) torque control and torque distribution method of the drive system has profound theoretical and practical significance. This paper studies the fast and accurate torque control method of Permanent Magnet Synchronous hub Motor to meet the Four-Wheel-Drive Electric Vehicle’s performance demand, through building PMSM mathematical model with iron loss took in account. Besides, shaft torque distribution method aiming to acquire best system efficiency is studied based on the proposed motor system loss model, thus achieving overall optimal torque control of four-wheel drive PMSM motor electric vehicle.The main contents are as follows:(1) Offline identification of PMSM parameters.In order to achieve precise control of the motor output torque, it is necessary to establish an accurate model with the actual motor torque parameters. Equivalent PMSM parameters with iron loss in consideration, including the exact magnitude of the permanent magnet flux, iron loss equivalent resistance, as well as relatively low accuracy of dq-axis equivalent inductance and stator equivalent resistance,are acquired through typical motor tests(stall and idle) and some special methods(voltage integral method, RLC bridge measurement method). In addition, a motor parameter identification method based on offline genetic optimization algorithm is proposed to improve the accuracy of parameter identification. Also this paper does the dead zone delay compensation through characteristic analysis of equivalent inverter parameters, thus reducing the effects of dead-band delay under inverter supply conditions on the parameter identification results. Then, the offline optimization identification of motor parameters is completed in Matlab using test data from typical steady conditions. Besides, a parameter validation program at the rotor field-oriented coordinates is designed to verify the accuracy of the motor parameters obtained from identification. The result shows that identified PMSM parameter can accurately reflect the motor response in different conditions. Therefore, the parameter identification results can be used for following researches on PMSM torque control algorithms and loss analysis of motor system.(2) Fast and accurate torque control of PMSM.Firstly, analyze the demands of four-wheel drive vehicles for permanent magnet synchronous in-wheel motor torque performance, mainly including vehicle dynamics, braking performance and the yaw moment control of the in-wheel motor torque responsive to the needs of time. In the analyzing of the yaw moment control performance requirements for motor, established a transfer function of the yaw moment control system, through the frequency domain analysis calculated the demand indicators for the hub motor torque response time, and verified by the simulation about the correctness of mentioned transfer moments response time indexes. Targeting at the defined in-wheel motor torque performance demands and as the designing goal: aiming at the accuracy demands of torque control proposed the torque model within the consideration of the factors of iron loss, and combined with the permanent magnet flux amplitude of online identification to increasing the motor torque control accuracy; for the response time requirements of torque control, proposed that through the improvement of the motor current to meet the demand of the torque response, firstly basing on the PI parameter tuning method to improving the current response under conventional PI controller, in order to further reducing the current response time proposed the current control method based on the feedforward + PI feedback, meanwhile, to enhancing the robustness of feedforward procession, adopted the method of motor online parameter identification to achieving the adaptive feedforward of current control.(3) Loss modeling and optimal braking torque analysis of In-wheel Motor drive system.Focusing on a single in-wheel motor, this paper builds motor loss model and inverter loss model based upon PMSM model involved with iron loss and the loss models are verified under various conditions. According to the motor loss model’s parameter sensitivity analysis, calculate motor losses using the result of motor parameters online identification in order to improve the accuracy of the loss model with changing motor parameters. Finally, on the basis of the system loss model, this paper studies optimal braking torque allocation to achieve the maximum brake energy recovery under certain motor speed and analyzes optimal braking torque variation pattern when motor parameters changes occur.(4) Research on optimal efficient torque distribution method between axles in driving system.Based on four-wheel driving system, firstly torque distribution situation is studied when parameters of front-axle and rear-axle motors are of no difference. We analyzed the optimal torque distribution by proposing extreme point of driving system consumption and consumption function irregularities of motor system, concluding that while single motor consumption model meets the concave function characteristics, the average torque distribution between front and rear axles can achieve optimal efficiency under linear condition without yaw-control. Besides, the conclusion can also be popularized to steering condition without yaw-control according to vehicle steering model. Secondly, when front-axle motor parameter differs from rear-axle, we proposed an optimal torque distribution coefficient calculation method according to motor system parameters, and theoretical analysis was made for the in-wheel PMSM driving system controlled with 0di ? method. Thirdly, for driving system without accurate motor parameter, torque distribution optimization algorithm based on efficiency MAP is proposed. The algorithm improved the objective optimization function by utilizing motor no-load consumption, making it better reflect the real system consumption. Simulation comparison validated the correctness of improved torque optimization algorithm. Lastly, this algorithm was also validated by bench test and real vehicle test.