EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Min–Max Model Predictive Control Method With Dynamic Feedback for Virtually Coupled Urban Rail Trains

Yu Zhu, Yadong Zhang, Jilie Zhang, Jin Guo

IET Intelligent Transport Systems · 2026-01 · 20 (1)

DOI: 10.1049/itr2.70380 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IET Intelligent Transport Systems, online erschienen am 15.11.2026. Verfasst von Yu Zhu und 3 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Compared with conventional train control technologies, virtual coupling enables trains to operate with shorter inter‐train distances, but its performance is more sensitive to disturbances and modelling uncertainties. To address this issue, this paper develops a distributed disturbance‐feedback min–max model predictive control (DF2M‐MPC) method for virtually coupled urban rail trains. First, a longitudinal train dynamics model subject to bounded disturbances is established within a distributed predecessor‐following architecture, and a min–max optimal control problem (OCP) incorporating tracking objectives, operating constraints, and inter‐train safety requirements is formulated. Building on established closed‐loop min–max MPC and disturbance‐feedback parameterisation theory, a disturbance‐affine dynamic‐feedback policy is introduced into each local robust OCP, in which the nominal control sequence and dynamic feedback matrix are jointly optimised. The resulting robust OCP is then reformulated as a tractable semidefinite programming (SDP) problem using robust linear matrix inequality (LMI) techniques. Simulations based on operational data from a metro line in Beijing are conducted to evaluate computational performance, disturbance rejection, formation coordination and operational safety. In the four‐train full‐process scenario, DF2M‐MPC reduces the full‐process velocity‐tracking mean absolute error (MAE) by 10.2%–11.7% relative to Tube MPC and by 19.5%–20.6% relative to Nominal MPC, while reducing the spacing MAE by 7.4%–14.2% and 16.0%–20.9%, respectively. DF2M‐MPC also provides larger safety margins and lower safety‐boundary violation levels. These results indicate improved disturbance‐rejection, formation‐coordination and safety‐related performance under the considered simulation conditions.

Förderung

  • National Natural Science Foundation of China
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Zhu, Y., Zhang, Y., Zhang, J., & Guo, J. (2026). A Min–Max Model Predictive Control Method With Dynamic Feedback for Virtually Coupled Urban Rail Trains. IET Intelligent Transport Systems, 20(1). https://doi.org/10.1049/itr2.70380

Vancouver

Zhu Y, Zhang Y, Zhang J, Guo J. A Min–Max Model Predictive Control Method With Dynamic Feedback for Virtually Coupled Urban Rail Trains. IET Intelligent Transport Systems. 2026;20(1). doi:10.1049/itr2.70380

BibTeX
@article{Zhu2026,
  author = {Zhu, Yu and Zhang, Yadong and Zhang, Jilie and Guo, Jin},
  title = {A Min–Max Model Predictive Control Method With Dynamic Feedback for Virtually Coupled Urban Rail Trains},
  journal = {IET Intelligent Transport Systems},
  year = {2026},
  volume = {20},
  number = {1},
  doi = {10.1049/itr2.70380}
}
Kostenlos starten →