Based on Artificial Neural Network to Realize K-Parameter Analysis of Vehicle Air Spring System

2017 International Conference on Sustainable Development on Energy and Environment Protection

能源学;生态环境科学

Hung, San-Shan^1 ; Hsu, Chia-Ning^1 ; Hwang, Chang-Chou^1 ; Chen, Wen-Jan^1

No.100, Wenhua Rd., Xitun Dist., Taichung City

407, Taiwan^1

关键词: Air spring suspension;    Classical controllers;    Control techniques;    Feasible control strategy;    Neural network designs;    Nonlinear characteristics;    Pressure variations;    Vehicle suspension systems;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/86/1/012023/pdf
DOI  :  10.1088/1755-1315/86/1/012023

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

In recent years, because of the air-spring control technique is more mature, that air- spring suspension systems already can be used to replace the classical vehicle suspension system. Depend on internal pressure variation of the air-spring, thestiffnessand the damping factor can be adjusted. Because of air-spring has highly nonlinear characteristic, therefore it isn't easy to construct the classical controller to control the air-spring effectively. The paper based on Artificial Neural Network to propose a feasible control strategy. By using offline way for the neural network design and learning to the air-spring in different initial pressures and different loads, offline method through, predict air-spring stiffness parameter to establish a model. Finally, through adjusting air-spring internal pressure to change the K-parameter of the air-spring, realize the well dynamic control performance of air-spring suspension.