Scholarlink

Purpose <p>This study aims to address the fundamental challenge of vehicle/guideway coupled vibration in high-speed maglev systems, which limits their stability and performance. The objective is to develop an integrated framework that effectively suppresses vibrations across a wide operational speed range.</p> Methods <p>A comprehensive approach combining reduced-order modeling and intelligent control is proposed. First, a computationally efficient coupled dynamics model is developed, featuring a 10-degree-of-freedom vehicle model interacting with a guideway represented by its dominant modes extracted from a finite-element analysis. The core of the methodology is a novel adaptive PID controller, where the control gains are automatically tuned online by a Radial Basis Function (RBF) neural network to maintain optimal performance under varying conditions.</p> Results <p>Numerical simulations confirm the superiority of the proposed method. The approach not only accurately characterizes the system’s dynamic interactions but also delivers exceptional vibration suppression across a wide speed spectrum (50-600 km/h). Notably, the RBFNN-enhanced PID controller reduces the overshoot and settling time by 40% and 35% respectively compared to conventional PID for standing still levitation control, and maintains stability with a reduction of up to 70% at critical high speeds (400-600 km/h) where conventional PID fails.</p> Conclusion <p>This study demonstrates the high efficacy of synergistically combining mechanics-based model reduction with data-driven control. The proposed framework provides a robust solution for vibration suppression in advanced maglev systems, successfully bridging the gap between the interpretability of traditional control and the adaptability of intelligent learning for high-performance operation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Dynamics and Adaptive Control of Vehicle/Guideway Vibrations in High-Speed Maglev using a Reduced Order Model and RBF Neural Network

  • 期刊论文

  • 发表日期：

    2026年4月6日

查看全文
  • Haiqin Li,
  • Shanqiang Fu,
  • Ang Li,
  • Juntao Du,
  • Xin Liang,
  • Dalian Yu,
  • Qian Ding

摘要

Purpose

This study aims to address the fundamental challenge of vehicle/guideway coupled vibration in high-speed maglev systems, which limits their stability and performance. The objective is to develop an integrated framework that effectively suppresses vibrations across a wide operational speed range.

Methods

A comprehensive approach combining reduced-order modeling and intelligent control is proposed. First, a computationally efficient coupled dynamics model is developed, featuring a 10-degree-of-freedom vehicle model interacting with a guideway represented by its dominant modes extracted from a finite-element analysis. The core of the methodology is a novel adaptive PID controller, where the control gains are automatically tuned online by a Radial Basis Function (RBF) neural network to maintain optimal performance under varying conditions.

Results

Numerical simulations confirm the superiority of the proposed method. The approach not only accurately characterizes the system’s dynamic interactions but also delivers exceptional vibration suppression across a wide speed spectrum (50-600 km/h). Notably, the RBFNN-enhanced PID controller reduces the overshoot and settling time by 40% and 35% respectively compared to conventional PID for standing still levitation control, and maintains stability with a reduction of up to 70% at critical high speeds (400-600 km/h) where conventional PID fails.

Conclusion

This study demonstrates the high efficacy of synergistically combining mechanics-based model reduction with data-driven control. The proposed framework provides a robust solution for vibration suppression in advanced maglev systems, successfully bridging the gap between the interpretability of traditional control and the adaptability of intelligent learning for high-performance operation.