elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Statistical process control model for switch failure detection and maintenance effectiveness assessment

Narezo Guzman, Daniela and Groos, Jörn Christoffer (2018) Statistical process control model for switch failure detection and maintenance effectiveness assessment. 1st European Railway Asset Management Symposium, 27.-28. March 2018, Nottingham, United Kingdom. (Unpublished)

Full text not available from this repository.

Abstract

Railway switches are crucial not only for normal operation of the railroad system as they guide trains to a track or platform, but also when disruptions occur since they allow trains to take alternative routes. Switch components and functions require frequent inspection, maintenance and renewal, making of switches a costly asset. The switch moving parts are subject to high deterioration and prone to malfunctioning, posing a safety hazard if no immediate action is taken. Nowadays online condition monitoring, standardization of inspection and maintenance actions, as well as data-based models are some of the tools supporting decision making for preventive planning, cost reduction and process effectiveness. This contribution presents a data-based model (derived from features extracted from measured point engine current during switch blade movement) for switch status nowcast and forecast applying statistical process control (SPC) methods. The SPC model is capable of identifying abnormal switch behavior; through examples it will be demonstrated how emerging failures in an early stage of development can be detected without the need of a labelled training data set of historical failures. The SPC model offers advantages over commonly used monitoring systems, as it does not rely on manually set switch-specific thresholds and references to detect the switch blades movements used to trigger alarms in these systems. Switch maintenance takes place regularly, sometimes significantly affecting the switch functional normal behaviour. Thus maintenance restricts somewhat the applicability of the SPC model. This contribution includes the discussion of methods applied for integrating the maintenance actions into the model. In turn, the SPC model output is used to assess the effectiveness of maintenance and the completeness of the reported actions performed on the switch. This work is partly funded by the EU H2020 and Shift2Rail Joint Undertaking projects In2Rail and In2Smart. The measurement data of the railway switches is provided by Strukton Rail.

Item URL in elib:https://elib.dlr.de/116390/
Document Type:Conference or Workshop Item (Speech)
Title:Statistical process control model for switch failure detection and maintenance effectiveness assessment
Authors:
AuthorsInstitution or Email of AuthorsAuthors ORCID iD
Narezo Guzman, DanielaDaniela.NarezoGuzman (at) dlr.dehttps://orcid.org/0000-0001-9748-1354
Groos, Jörn ChristofferJoern.Groos (at) dlr.dehttps://orcid.org/0000-0003-3871-0756
Date:27 March 2018
Refereed publication:No
Open Access:No
Gold Open Access:No
In SCOPUS:No
In ISI Web of Science:No
Status:Unpublished
Keywords:Railway, Switch and Crossings, Statistical Process Control, Asset Management
Event Title:1st European Railway Asset Management Symposium
Event Location:Nottingham, United Kingdom
Event Type:international Conference
Event Dates:27.-28. March 2018
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Transport
HGF - Program Themes:Traffic Management (old)
DLR - Research area:Transport
DLR - Program:V VM - Verkehrsmanagement
DLR - Research theme (Project):V - Next Generation Railway Systems III (old)
Location: Berlin-Adlershof
Institutes and Institutions:Institute of Transportation Systems
Deposited By: Narezo Guzman, Daniela
Deposited On:07 Aug 2018 09:09
Last Modified:07 Aug 2018 09:09

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Copyright © 2008-2017 German Aerospace Center (DLR). All rights reserved.