摘要
针对拥堵工况下车辆自动变速器频繁换挡的问题,选取车辆平均车速、平均节气门开度和采样时间内制动踏板作动次数为评价因子,建立T-S模糊神经网络进行拥堵工况辨识,提出基于拥堵工况辨识的车辆自动变速器分层修正控制策略;将车辆自动变速控制分为上层辨识决策层与下层换挡执行层,上层采用T-S模糊神经网络进行拥堵工况辨识与换挡修正决策;下层接收上层修正控制指令执行换挡修正。仿真与实车试验结果表明:采用TS模糊神经网络可准确识别拥堵工况,基于拥堵工况辨识的车辆自动变速分层修正控制策略可有效避免拥堵工况时频繁换挡,减少换挡执行部件和制动系统的磨损。
In view of the undesirable frequent shifting of vehicle automatic transmission under congestion conditions,a T-S fuzzy neural network is set up to identity congestion conditions and a hierarchical correction control strategy for automatic transmission based on congestion condition identification is proposed with the average vehicle speed,the average throttle opening,and the average times of brake pedal actuation in sampling period selected as evaluation factors. The vehicle automatic transmission control is divided into two layers: the upper layer for identification and decision-making while the lower layer for shifting execution. The upper layer adopts T-S fuzzy neural network to identify congestion conditions and make decisions of shift correction,while the lower layer executes corrected shift according to control instructions from the upper layer. The results of simulation and real vehicle test show that using T-S fuzzy neural network can accurately identify congestion conditions,and the hierarchical correction control strategy based on congestion conditions identification can effectively avoid frequent shifting in congestion conditions,and hence reduce the wear of shift actuation components and brake system.
作者
夏光
邹斌
唐希雯
陈无畏
Xia Guang1, Zou Bin1, Tang Xiwen2, Chen Wuwei3(1. Automotive Engineering Technology Research Institute, Hefei University of Technology, Hefei 230009;2. School of Radar Confrontation, National University of Defense Technology, Hefei 230037;3. School of Automotive and Traffic Engineering, Hefei University of Technology, Hefei 23000)
出处
《汽车工程》
EI
CSCD
北大核心
2018年第4期465-474,共10页
Automotive Engineering
基金
国家自然科学基金(51205101)
科技部重点研发计划项目(2016YFD0700604和2016YFD0700605)资助
关键词
自动变速器
拥堵工况辨识
换挡修正
分层控制
T-S模糊神经网络
AT
congestion condition identification
shifting correction
hierarchical control
T-S fuzzy neural network