| Dual-Motor Planetary Coupling Driving System(DMPCDS) used in electric buses has a novel propulsion topology. Two motors are connected to the sun gear and the ring gear of a planetary gear train respectively, and the power from the motors output through the carrier to drive the vehicle. When the vehicle speed is low, only the motor connected to the sun gear(Sun-Motor for short) works, and the Sun-Motor output torque is enlarged by the planetary gear train to meet the torque requirements when the bus is climbing or launching. When the vehicle speed is high, the motor connected to the ring gear(Ring-Motor for short) begins to work, and the two motors’ speeds are coupled by the planetary gear train to drive the bus. To make sure the DMPCDS working steadily, smoothly and efficiently, the coordinated control methods of two motors coupling driving, the control methods of mode switching, and parameters optimization were studied.Dual motors planetary coupling drive is a multi-stream transmission, and the motors have the four-quadrant operating characteristics, thus the power flows in different operating statuses are different. Based on the analysis of the power flows in different statuses, the bond graphs were drawn, and the dynamic mathematical model of the DMPCDS was built. According to the mathematical model and the motors’ working characteristics, the control model of the DMPCDS in two-motor mode was established with two motors’ commands as inputs and two motors’ speeds as outputs. Through measuring the interaction degree of two inputs and two outputs of the system by Relative Grain Array(RGA), it was revealed that the interaction degree was weak when the Sun-Motor speed was greater than the rated speed. So the two motors’ rotation speed can be separately controlled and did not need to be decoupled. Therefore the coordinated control methods were formulated: the Sun-Motor speed was controlled by proportional regulator in the closed-loop and the Ring-motor was controlled in the open-loop.In the closed-loop of the Sun-Motor speed control system the data is exchanged through CAN bus network, Time-delay is the main problem in the control system. After analyzing the lag of the system, it was found that the lag was approximate to the period of transmitting message in CAN bus. By the means of computer simulation, the stability of the system, which is a constant delay nonlinear system, was analyzed and the effects on the stability of the parameters of proportional coefficient and reference speed were obtained.To avoid cyclic shifting, the mode switching schedule based on the two motors’ speeds was designed. Based on it, the control methods including the power reduction of the Sun-Motor and the adjustment of the reference speed were proposed. These methods could realize one-motor mode to two-motor mode switching smoothly without power interruption. This study also proposed the control methods including torque control for Motor1 and delayed stopping the work of Motor2 which could realize switching from two-motor mode to one-motor mode with a minor jerk but power interruption. Then the integrated control methods including the two motor couping drive coordinated control methods and the mode switching control methods were validated by the combined models from AMESim and Simulink.The on-board measurement system based on CAN was developed, and the integrated control methods were test on the vehicle. The experiments showed that the integrated control methods could coordinate the two motors to work steadily, and can successfully manipulate mode switching without cyclic shifting. There was no jerk and no power interruption when the system switched from one-motor mode to two-motor mode; however there was a jerk but it was smaller than the German standards limit when the system switched from two-motor model to one-motor mode. The energy consumption economy of DMPCDS was test on the bench. The results showed that: the area of high efficiency(≥80%) exceeded 50% of the whole work range, and the efficiency was low when the system output torque was small and the system speed was near the mode switch speed.The dimensionless model of the motor efficiency was built based on Willans line model. After that the scalable model with the variation components parameters and control parameters of DMPCDS was established. Then by applying the improved adaptive genetic algorithm, the component parameters and the control parameters were jointly optimized simultaneously with the constraints of vehicle dynamic performance requirements. It’s worth noting that the China typical city bus driving cycle was adopted in the optimization process and the energy consumption was selected to be the minimization target. The simulation results showed that the optimized parameters could make the bus dynamic performance meet the requirements, and compared with the former parameters they could improve the energy consumption performance by 14.3%. |
PDF Full Text Request