High-speed Rail Inspection ("Smart Trains")

The most severe rail flaws: Detail Fracture, Transverse Fissure, Vertical Split Head

UCSD "smart train" concept: passive rail inspection exploiting train wheel excitation

The UCSD team at the 2016 field test at the Transportation Technology Center, Pueblo, CO

The UCSD prototype at the 2018 TTCI field tests

Detection of joints and welds at 60 mph and 80 mph !!

Receiver Operating Characteristic (ROC) curve for detection of welds at 40 mph from 2018 and 2019 field tests at TTCI

Funding:

Collaborators:

Purpose:

To enable rail inspections at regular train speeds (e.g. 60 mph and beyond), so as to allow trains to perform their own inspections during regular operations ("smart train" concept). This accomplishment will allow (a) better rail maintenance practices because of the minimization or elimination of traffic disruptions caused by specialized inspection vehicles (usually running at less than 30 mph), and (b) improved probability of defect detection due to the great opportunity for redundancy afforded by multiple train passes over the same track.  

Synopsis:

Safety statistics data from the US Federal Railroad Administration for the ten years 2009-2019 indicate that the three leading causes of train accidents within the category “rail, joint bar and rail anchoring” are: the Detail Fracture (1st leading cause of accidents), the Transverse/Compound Fissure (2nd leading cause of accidents), and the Vertical Split Head (3rd leading cause of accidents). The project is targeting these three defects, in addition to the surface head checks.

UC San Diego is developing a rail inspection system based on (a) non-contact ultrasonic probing and (b) passive extraction of the rail Green's function by exploiting the acoustic excitation from the rolling wheels of the travelling train. The first prototype of this "passive" rail inspection system was tested in 2016 at the Transportation Technology Center at speeds up to 80 mph with good results. Additional development tests of this revolutionary technology are ongoing. 

Selected Publications:

Liang, A., Sternini, S., Capriotti, M., and Francesco Lanza di Scalea, “High Speed Ultrasonic Rail Inspection by Passive Non-contact Technique,” Materials Evaluations, Special Issue on NDT of Railroads, Dr. Anish Poudel, ed., 77(7), pp. 941-950, 2019. 

Xuan, P., Lanza di Scalea, F., Capriotti, M., Liang, A. and Sternini, S.,  “High-speed passive-only rail track integrity evaluation using deep learning-based anomaly detection,” paper no. 10971-25, SPIE Vol. 10971, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIII, SPIE Smart Structures and Nondestructive Evaluation Symposium, Denver, CO, March 3-7, 2019 

 Lanza di Scalea, F., Liang, A., Sternini, S., Capriotti, M., Datta, D., Zhu, X., and Wilson, R., “Ultrasonic Identification of Rail Tracks from Natural Wheel Excitations: Potential for High-Speed and High-Redundancy Rail Inspection,” CD-ROM Proceedings of the 12th International Workshop on Structural Health Monitoring (IWSHM 2019), pp. 1-9, Stanford University, September 10-12, 2019.

Lanza di Scalea, F., Zhu, X., Capriotti, M., Liang, A., Mariani, S., and Sternini, S., “Passive Extraction of Dynamic Transfer Function from Arbitrary Ambient Excitations: Application to High-speed Rail Inspection from Wheel-generated Waves,” ASME Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems, Inaugural Issue 1(1), pp. 0110051-01100512, 2018. 

Lanza di Scalea, F., Sternini, S., and Liang, A., “Robust Passive Reconstruction of Dynamic Transfer Function in Dual-Output Systems,” Journal of the Acoustic Society of America, 143(2), pp. 1019-1028, 2018. 

Lanza di Scalea, F., Liang, A., Sternini, S., Zhu, X., Capriotti, M. and Wilson, R., “Potential For High-Speed Rail Inspection By Passive-Only Ultrasonic Monitoring,” Proceedings of the ASME Joint Rail Conference (ASME-JRC), paper no. JRC2018-6140, pp. 1-2, Pittsburgh, PA, April 18-20, 2018. 

 Lanza di Scalea ,F.,  Liang, A., Sternini, S., Capriotti, M.,  Datta, D.,  Zhu, X., and Wilson, R., “Ultrasonic Identification of Rail Tracks from Natural Wheel Excitations: Potential for High-Speed and High-Redundancy Rail Inspection,” CD-ROM Proceedings of the 12th International Workshop on Structural Health Monitoring (IWSHM 2019), pp. 1-9, Stanford University, September 10-12, 2019.

Liang, A., Sternini, S., Capriotti, M., Zhu, X., Lanza di Scalea, F., And Wilson, R., “Passive Extraction Of Green’s Function Of Solids And Application To High-Speed Rail Inspection,” paper no. 10970-29, Proceedings SPIE Vol. 10970, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems (Conference 10970),  part of SPIE Smart Structures and Nondestructive Evaluation Symposium, Denver, CO, 3-7 March, 2019.

Xuan, P., Lanza di Scalea, F., Capriotti, M., Liang, A. and Sternini, S.,  “High-speed passive-only rail track integrity evaluation using deep learning-based anomaly detection,” paper no. 10971-25, Proceedings SPIE Vol. 10971, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIII,  part of SPIE Smart Structures and Nondestructive Evaluation Symposium, Denver, CO, 3-7 March, 2019