Inverse dynamics of vehicle matching with mechanical elastic wheels based on genetic neural network

Affiliation of Author(s):能源与动力学院

Journal:Huazhong Ligong Daxue Xuebao

Abstract:To analyze the relationship between mechanical elastic wheel (MEW) cornering properties and vehicle stability, a nonlinear 3-DOF (degree of freedom) vehicle model was built up by Simulink.The MEW model parameters were identified by genetic algorithm, and cornering stiffness and lateral force peak value were utilized as the MEW cornering characteristics evaluation indexes.Vehicle stability evaluation parameters with different tire performance were obtained under steering wheel angle sin input.Based on the genetic algorithm optimized back propagation (BP) neural network, and the inverse dynamics model was built, by using the vehicle stability evaluation parameter as the input and the matched tire performance as the output.Research results show that the estimate error of MEW parameters by genetic algorithm is less than 4%, the maximum error of test sample by genetic algorithm neural network (GANN) is 6.2%, and the average error is 3.1%, so the forecasting ability of GANN model for wheel cornering characteristics is satisfied. © 2019, Editorial Board of Journal of Huazhong University of Science and Technology. All right reserved.

ISSN No.:1671-4512

Translation or Not:no

Date of Publication:2019-05-23

Co-author:Li, Haiqing,Yan, Xi,Xiu Qiaoyan,l,wanglinfeng

Correspondence Author:Li, Haiqing,Zhao Youqun

Date of Publication:2019-05-23