1
Return

Design of a nonclipped H∞ controller for high-speed train magnetorheological suspension system

delete2026-05-08
delete0
PRE
AI
Y
Yaowen Zhang
C
Chen, Chujun *
F
Fei, Lingwei
DOI:10.1088/1361-6501/ae5cbedelete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
The prevalent controller design for high-speed train (HST) magnetorheological (MR) suspension systems relies on the clipped framework (CF), which potentially leads to high-frequency performance degradation, accelerated failure, exacerbated electromagnetic interference noises, and reduced hunting stability. To address these drawbacks, this paper proposes a novel nonclipped (NC) H-infinity controller. First, a seven-degree-of-freedom HST model with MR dampers is established. Next, a piecewise model is derived for approximating the nonlinear MR damper constraint via a reference force and a symmetric constraint. A NC framework (NCF) is subsequently proposed to eliminate truncation effects and ensure actuation smoothness. By integrating the piecewise model with the NCF, a piecewise controller is synthesized, wherein lateral ride comfort is ensured by H-infinity performance and a modified sector condition handles the nonaffine property introduced by the NCF. Ultimately, the controller design is formulated as a linear matrix inequality optimization problem, from which the control gains are derived. Comprehensive simulations demonstrate that the proposed controller substantially improves the actuation smoothness and hunting stability over CF-based controllers, without compromising ride comfort. The paper thus offers a more effective and reliable solution for HST-MR suspension systems.
Keywords:
H(infinity)control
semi-active control
high-speed train
magnetorheological suspension
nonlinear constraint
vehicle system dynamics and control

Journal

Measurement Science and Technology cover
Measurement Science and Technology
IF:
3.4
Papers:
2.6K
Citations:
2.3W

Organization

S
southwest jiaotong university
Scholars:
7.6K
Papers: 2.7K
Citations: 0
Cited Papers

Cited Papers

Citing Papers

Citing Papers