基于神经网络的参数自调整车辆牵引力控制算法  

Self-adjusting Traction Control Algorithm Based on Neural Network for Automobiles

作　　者：刁增祥[1] 余建星[1] 李静[2] 石桂花[2] 

机构地区：[1]天津大学建筑工程学院,天津300722 [2]吉林大学汽车工程学院,长春130022

出　　处：《吉林大学学报（工学版）》2005年第3期249-253,共5页Journal of Jilin University:Engineering and Technology Edition

基　　金：中国人民解放军总装备部"十五"预先研究项目(2001-CL-0102)

摘　　要：应用神经网络方法设计了牵引力控制系统的油门位置和制动压力增量PI控制器参数在线自调整算法,介绍了调试控制算法的硬件在环试验台,采用硬件在环试验的研究方法完成了参数自调整牵引力控制算法的调试。试验结果表明:参数自调整牵引力控制算法能将驱动轮控制在目标滑转状态,并有良好的鲁棒性。The PI controller parameters on-line self-adjusting algorithms based on the neural network for the engine accelerator position and the brake pressure in the wheel brake cylinder of a traction control system were compiled. A hardware-in-the-loop simulator to debug the control algorithms was introduced, and the debug of the proposed traction control system was performed. The results show that the proposed algorithms are characterized by good robustness and capable to limit the driving wheel slip rate within the target value.

关 键 词：车辆工程 神经网络 牵引力控制 硬件在环 算法 

分 类 号：U463.54[机械工程—车辆工程]