Scholarlink

<p>Integrating train-level operation control with the coordination of multiple power units in high-speed trains (HSTs) is challenging due to inter-unit coupler interactions and limited onboard computational resources. This paper proposes an integrated multi-power-unit control framework for HSTs that directly optimizes unit-level traction/braking commands—bypassing the conventional hierarchical scheme that first computes a train-level command and then allocates it to individual units—to improve speed tracking accuracy and enhance coordination among multiple power units. A multi-particle dynamic model is developed to capture traction/braking dynamics, nonlinear basic resistance, and inter-unit coupler forces. Based on this model, a multi-objective cost function is formulated to jointly penalize speed tracking errors, energy consumption, and coupler forces. To enable real-time implementation, a move-blocking strategy is embedded into a model predictive control (MPC) framework, substantially improving computational efficiency. Finally, case studies based on operational data from the Beijing-Shanghai high-speed railway are conducted. Compared with the hierarchical two-stage baseline (train-level command computation followed by unit-level allocation), the proposed integrated framework reduces the speed-tracking error by 18% and the coupler force by 52%, demonstrating improved control accuracy and stronger multi-unit coordination. Compared with the conventional MPC controller, the proposed move-blocking MPC reduces the per-iteration solution time of the finite-horizon optimal control problem from 7&#xa0;ms to 3.7 ms.</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

An efficient move-blocking MPC approach for coordinated control of multiple power units of high-speed trains

  • 期刊论文

  • 发表日期：

    2026年5月18日

查看全文
  • Zixuan Zhang,
  • Yuan Cao,
  • Zhongbei Tian,
  • Shuai Su

摘要

Integrating train-level operation control with the coordination of multiple power units in high-speed trains (HSTs) is challenging due to inter-unit coupler interactions and limited onboard computational resources. This paper proposes an integrated multi-power-unit control framework for HSTs that directly optimizes unit-level traction/braking commands—bypassing the conventional hierarchical scheme that first computes a train-level command and then allocates it to individual units—to improve speed tracking accuracy and enhance coordination among multiple power units. A multi-particle dynamic model is developed to capture traction/braking dynamics, nonlinear basic resistance, and inter-unit coupler forces. Based on this model, a multi-objective cost function is formulated to jointly penalize speed tracking errors, energy consumption, and coupler forces. To enable real-time implementation, a move-blocking strategy is embedded into a model predictive control (MPC) framework, substantially improving computational efficiency. Finally, case studies based on operational data from the Beijing-Shanghai high-speed railway are conducted. Compared with the hierarchical two-stage baseline (train-level command computation followed by unit-level allocation), the proposed integrated framework reduces the speed-tracking error by 18% and the coupler force by 52%, demonstrating improved control accuracy and stronger multi-unit coordination. Compared with the conventional MPC controller, the proposed move-blocking MPC reduces the per-iteration solution time of the finite-horizon optimal control problem from 7 ms to 3.7 ms.