A disturbance observer based lumped-mass catenary model for active pantograph design and validation

Research output: Contribution to journalArticlepeer-review

102 Downloads (Pure)

Abstract

To understand and thus improve the dynamic behaviour of the railway pantograph-catenary system, researchers have developed a number of approaches to the catenary modelling. The Finite Element (FE) approach provides a high degree of accuracy, but at the cost of significant computing time. In contrast, researchers in active pantographs generally use relatively simple (and fast running) lumped-mass catenary models. This paper aims to improve on the models available for the design and validation of active pantographs. Firstly, the veracity of the classic lumped-mass model is assessed and a new catenary model is proposed, by augmenting the lumped-mass model with a dynamics-matched mass and a disturbance which can be found using a disturbance observer (DO). Open-loop, passive pantograph, results show that the proposed model can represent the dynamic behaviour exhibited in the FE catenary very well. Secondly, two PID controllers are designed for an active pantograph, and tested with the FE, classic and DO-based lumped-mass catenary models. It is shown that if a controller is validated on the classic lumped-mass model, the credibility will be low. In contrast, the new DO-based model is found to deliver more robust results, providing efficient simulations for active pantograph design and validation.
Original languageEnglish
Number of pages18
JournalVehicle System Dynamics
Early online date13 Jun 2022
DOIs
Publication statusE-pub ahead of print - 13 Jun 2022

Bibliographical note

Publisher Copyright:
© 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

Keywords

  • Pantograph-catenary modelling
  • Dynamic Behaviour
  • disturbance observer
  • active pantograph

Fingerprint

Dive into the research topics of 'A disturbance observer based lumped-mass catenary model for active pantograph design and validation'. Together they form a unique fingerprint.

Cite this