自动变速器电子换挡系统挡位位置自学习控制算法设计

电子换挡系统 （Electronic Transmission Range Select System，ETRS） 的控制精度受自动变速器、换挡执行机构的零件自身误差及装配误差的影响，驻车挡、倒车挡、空挡、前进挡各挡位的理论位置与实际装配结果不可能完全匹配，这不仅会影响电子换挡系统的控制精度，而且长期使用后可能存在换挡功能失效的潜在风险。针对上述问题，研究设计了电子换挡系统相关硬件架构、挡位位置识别方法及关键自学习控制算法。控制算法集成直流电机匀速控制、H桥驱动电流数据读取、槽底挡位位置识别、多轮次槽底扫描迭代及挡位位置校验。仿真及实车试验表明，设计的挡位位置识别方法能够实现误差不大于0.15°，自学习控制算法能够实现实车试验大数据下均值与理论角度位置差距不大于0.3°，同时保证100%成功率，满足电控换挡系统长期工作的准确性及耐久性要求。 .

Gear Position Self-Learning Control Algorithm for Electronic Transmission Range Select System of Automatic Transmission

The control accuracy of the electronic transmission range select (ETRS) system is affected by the part dimensional errors and assembly errors in the automatic transmission and the gear box actuator. The theoretical positions of the parking gear, reverse gear, neutral gear and drive gear cannot match the actual results after assembly, which will not only affect the control precision of the ETRS system, but also have the risk of shift failure after long-term use. Therefore the hardware architecture was designed and the gear position recognition method and the key self-learning control algorithm were proposed for the ETRS system.The control algorithm achieves uniform speed control of the DC motor, sampling of H-bridge drive current data, gear position identification at the slot bottom, multi-turn scanning of the slot bottom and the gear position verification. The simulation and experimental results show that the error of the gear position identification is less than 0.15°. By applying the self-learning control algorithm, the difference between the mean value from the actual vehicle test results using big data and the theoretical angle is within 0.3°, meeting the long-term requirements on accuracy and durability for the ETRS system.