商用车侧翻的灰色-马尔科夫链预测方法研究  

Research on Grey-Markov Chain Prediction Method of Commercial Vehicle Rollover

作　　者：辜志强[1] 王朝阳 Gu Zhiqiang;Wang Chaoyang(Wuhan University of Technology,Wuhan 430070)

机构地区：[1]武汉理工大学,武汉430070

出　　处：《汽车技术》2022年第3期42-48,共7页Automobile Technology

基　　金：国家自然科学基金青年基金项目(51305313)。

摘　　要：为精确评价商用车侧翻危险程度和补偿气压制动系统的气压建立时延,提出一种商用车侧翻的灰色-马尔科夫链预测方法。首先,建立灰色预测模型并在不同预测时间下进行仿真测试;然后,针对灰色模型预测精度不足的问题,采用马尔科夫链理论进一步优化,并同时探究基于残差划分的状态数量对预测结果的影响,仿真结果表明,在状态数量为11个的情况下,预测最大误差已降至5.9%左右;最后,将电子控制单元(ECU)接入IPG/TruckMaker硬件在环试验台进行验证,结果表明,ECU在车辆侧翻前0.555 s输出预警信号,说明该预测方法有效。In order to accurately evaluate the risk of commercial vehicle rollover and compensate for the air pressure establishment time delay of the pneumatic braking system,the paper proposes a gray-Markov chain prediction method for commercial vehicle rollover.Firstly,a gray prediction model is established to simulate at different prediction times;then,for the problem of insufficient prediction accuracy of the gray model,Markov chain theory is used for further optimization,and at the same time,the influence of the number of states on the prediction results is explored based on the residual division.The simulation results show that,when the number of states is 11,the maximum prediction error has been reduced to about5.9%;finally,the ECU is connected to the IPG/TruckMaker Hardware-In-the-Loop test bed for verification.The results show that the ECU outputs an early warning signal 0.555 s before the vehicle rolls over,indicating that the prediction method is effective.

关 键 词：气压制动时延 灰色模型 马尔科夫链理论 硬件在环 

分 类 号：U461.91[机械工程—车辆工程]