基于工况预测的电动汽车两挡AMT换挡过程控制

Shift process control of two-gear AMT for electric vehicle based on driving condition prediction

电动汽车两挡AMT换挡过程中,在驱动电机扭矩卸载和恢复阶段,驱动电机扭矩变化速率对冲击度影响最为明显,而冲击度和换挡时间是对立的,为了平衡换挡时间和冲击度,提升换挡品质,提出一种基于工况预测的扭矩控制策略。采用欧几里得贴近度识别车辆实际行驶工况类型,建立径向基函数(Radial Basis Function,RBF)神经网络预测未来行驶车速信息;设计模糊控制器识别驾驶员期望换挡类型,在驱动电机扭矩卸载和恢复阶段,输出相应换挡类型的驱动电机扭矩变化速率。仿真分析和实车验证表明:基于工况预测的扭矩控制策略可根据预测车辆的实际行驶工况实时调整驱动电机扭矩卸载和恢复阶段的驱动电机扭矩变化速率,平衡换挡时间和冲击度,满足了驾驶员对换挡平顺性和动力性的需求,有助于实现换挡品质的提升。

During the two-gear AMT shift of electric vehicle,the torque change rate of the drive motor has the most obvious influence on the impact degree in the torque unloading and recovery stage of the drive motor,while the impact and shift time are opposite.In order to balance the shift time and impact and improve the shift quality,a torque control strategy based on driving condition prediction was proposed.Euclidean closeness was used to identify the actual driving condition types of vehicles,and Radial Basis Function(RBF)neural network was established to predict the future driving speed information.The designed fuzzy controller identifies the type of shift expected by the driver and outputs the torque change rate of the drive motor for the corresponding shift type during the torque unloading and recovery stage of the drive motor.Simulation analysis and real vehicle verification showed that the torque control strategy based on driving condition prediction could adjust the torque change rate of the drive motor in unloading and recovery stages in real time according to the actual driving condition of the predicted vehicle,balance the shift time and impact,meet the driver′s demand for shift smoothness and power,and help to improve the shift quality.