Principal Investigator

Hae Young Noh

Status

Completed

Start Date

Sept. 1, 2018

End Date

Aug. 31, 2019

Project Type

Research Applied

Grant Program

FAST Act - Mobility National (2016 - 2022)

Grant Cycle

2018 Traffic21

Visibility

Public

Abstract

This project aims to indirectly monitor railway bridges for structural damage diagnosis using instrumented operational trains. Our group will conduct an on-site experiment on the Panhandle bridge that carries two rail lines of the Port Authority "T" line across the Monongahela River in Pittsburgh. The ultimate objective of this project is to develop a system that would provide continuous monitoring of bridges and rail tracks by collecting vibration data from sensors on in-service trains.    

Description

Though the private freight rail industry makes $9.7 billion capital investment in maintaining the network, which is comprised of almost 140,000 miles of track and over 100,000 bridges in 2015, the Federal Railroad Administration still reported 1,096 derailments including 428 track caused accidents in the nation in 2017. In addition, the 2017 National Bridge Inventory of the Federal Highway Administration found that 47,619 out of 615,002 bridges in the U.S. were in poor condition across the nation. These make apparent the need for monitoring rail networks and bridge conditions. Recently, indirect structural health monitoring of bridges has become popular since it is a low-cost and low-maintenance approach in which sensors on the vehicle are used to detect infrastructure changes and damage.

Supported by the University Transportation Center at CMU, our group has been monitoring Pittsburgh’s light rail network from sensors placed on passenger trains. Cooperating with Port Authority of Allegheny County, we instrumented one train in fall 2013, and a second train in summer 2015. We have been continuously collecting the trains’ position using GPS and their dynamic responses using accelerometers. Over time, we learned how the trains respond to each section of track and used a combination of data-driven and physics-based approaches to detect changes to the track condition relative to its historical baseline [1-3].

In 2017, we focused on indirectly monitoring railway bridges for two reasons. First, structural damage, especially internal damage on bridges, can be detected by using accelerometers. Secondly, after managing to detect sudden changes in the tracks, we want to detect and quantify gradual changes using acceleration signals. We conducted lab-scale experiments and provided evidence of the applicability of the indirect damage diagnosis approach through better accuracy from the indirect sensors than the direct bridge sensors in certain cases [4, 5]. However, further study is required to validate the robustness of this damage diagnosis framework with a more realistic system. For instance, on-site tests are needed to verify the feasibility of this work in practice, including the influence of different types of damage scenarios, vehicle velocity, ongoing traffic, and environmental factors.

Currently, our group is preparing an on-site experiment on the Panhandle bridge that carries two rail lines of the Port Authority "T" line across the Monongahela River in Pittsburgh. We will use one or two train cars as weight and position it along one of the tracks on the bridge. The other track will be used for running the instrumented train back and forth with the sensors. One of the main goals of this experiment would be to verify the feasibility and reliability of damage diagnosis algorithms for indirect structural health monitoring of bridges.

The ultimate objective of this project is to develop a system that would provide continuous monitoring of bridges and rail tracks by collecting vibration data from sensors on in-service trains. Ideally, such a system would be able to not only detect, localize and quantify the damage of bridges soon after they begin to occur but also achieve transferable damage diagnosis for different bridges. Also, in the future, we want to expand the application of our approach to indirectly monitor and inspect regular road and bridges using cars or trucks.

We've been talking to potential industry partners for collaboration and funding (Wabtec, Norfolk,  Sixense, etc.) and we plan to seek grants from NSF, PennDOT, FHWA, FRA, etc. The proposed budget will be used to support one Ph.D. student to deploy sensors, collect data, develop and apply damage diagnosis algorithms.

[1] Lederman, G., Chen, S., Garrett, J., Kova?evi?, J., Noh, H. Y., & Bielak, J. (2017). “Track-monitoring from the dynamic response of an operational train.” Mechanical Systems and Signal Processing, 87, 1-16.
[2] Lederman, G., Chen, S., Garrett, J. H., Kova?evi?, J., Noh, H. Y., & Bielak, J. (2017). “Track monitoring from the dynamic response of a passing train: a sparse approach.” Mechanical Systems and Signal Processing, 90, 141-153.
[3] Lederman, G., Chen, S., Garrett, J. H., Kova?evi?, J., Noh, H. Y., & Bielak, J. (2017). “A Data Fusion Approach for Track Monitoring From Multiple In-Service Trains.” Mechanical Systems and Signal Processing, 95, 363-379.
[4] Liu, J., Chen, S., Bergés M., Bielak, J., Garrett, J. H., Kova?evi?, J., & Noh, H. Y. “Damage diagnosis algorithms for indirect structural health monitoring of bridges via dimensionality reduction.” Mechanical Systems and Signal Processing, Under review.
[5] Liu, J., Bergés M., Bielak, J., Garrett, J. H., Kova?evi?, J., & Noh, H. Y. “A damage localization and quantification algorithm for indirect structural health monitoring of bridges using multi-task learning.”, In AIP Conference Proceedings, Under review.

Timeline

1. Sept. 2018 – Nov. 2018: Set and conduct experiments on the Panhandle bridge;
2. Dec. 2018 – Jan. 2019: Analyze collected data; develop and test damage diagnosis algorithms;
3. Feb. 2019 – May. 2019: Summarize findings, remark conclusions and plan future works.

Strategic Description / RD&T

Deployment Plan

Supported by the University Transportation Center at CMU, our group has been monitoring Pittsburgh’s light rail network from sensors placed on passenger trains. Cooperating with Port Authority of Allegheny County, we instrumented one train in fall 2013, and a second train in summer 2015. We have been continuously collecting the trains’ position using GPS and their dynamic responses using accelerometers. Over time, we learned how the trains respond to each section of track and used a combination of data-driven and physics-based approaches to detect changes to the track condition relative to its historical baseline [1-3].

In Fall 2018, our group is conducting on-site experiments on the Panhandle bridge that carries two rail lines of the Port Authority "T" line across the Monongahela River in Pittsburgh. We use one or two train cars as weight and position it along one of the tracks on the bridge. The other track is used for running the instrumented train back and forth with the sensors. One of the main goals of this experiment would be to verify the feasibility and reliability of damage diagnosis algorithms for indirect structural health monitoring of bridges.

Expected Outcomes/Impacts

1. Instrumented light rail trains
2. An algorithm to analyze train response data to extract bridge states (metric: accuracy, false alarm, missed detection)

Expected Outputs

TRID

Individuals Involved

Email	Name	Affiliation	Role	Position
marioberges@cmu.edu	Berges, Mario	CEE	Co-PI	Faculty - Untenured, Tenure Track
noh@cmu.edu	Noh, Hae Young	CEE	PI	Faculty - Untenured, Tenure Track

Budget

Amount of UTC Funds Awarded

$48420.00

Total Project Budget (from all funding sources)

$48420.00

Documents

Type	Name	Uploaded
Data Management Plan	Data_Management_Plan_rXaJqw2.pdf	Nov. 5, 2018, 10:25 a.m.
Publication	Evaluation of vehicle vibration-based indirect structural health monitoring on an in-service railway truss bridge	March 26, 2019, 9:19 a.m.
Progress Report	273_Progress_Report_2019-03-30	March 26, 2019, 9:19 a.m.
Publication	Dynamic responses, GPS positions and environmental conditions of two light rail vehicles in Pittsburgh	Sept. 29, 2019, 9:27 a.m.
Presentation	An expectation-maximization-based framework for vehicle-vibration-based indirect structural health monitoring of bridges	Sept. 29, 2019, 9:27 a.m.
Progress Report	273_Progress_Report_2019-09-30	Sept. 29, 2019, 9:27 a.m.
Final Report	273_-_Final_Report.pdf	Nov. 5, 2019, 5:32 a.m.

Match Sources

No match sources!

Partners

No partners!