Laser-based Non-destructive Spike Defect Inspection System

For centuries, spikes, with marginal changes in terms of geometry and material, have been consistently providing reliable restrictions to rails. With the increasing axle load and operational speed, spikes are subjected to more demanding loading conditions, especially in the territories where tracks have high curvature. Because the cracks are typically underneath the spike head, it is very difficult to distinguish the broken spikes without physical examination, which causes formidable challenges in track health evaluation and operational safety. The objective of this research effort is trying to develop a low-cost, non-destructive and contact-free intelligent inspection system to identify broken spikes at the real-time rate. The proposed system will integrate laser excitation, acoustic analysis, computer vision and pattern recognition, and artificial intelligence (AI). The success of this research may significantly improve the efficiency and accuracy in spike inspection and enhance railroad track safety.

Record URL:

Language

English

Project

Status: Active
Funding: $100000
Contract Numbers:
Project 44
Sponsor Organizations:
Safety Innovations Deserving Exploratory Analysis (IDEA)

Transportation Research Board
500 Fifth Street, NW
Washington, DC United States 20001
Project Managers:
Fitzpatrick, Velvet
Performing Organizations:
University of South Carolina

,
Principal Investigators:
Qian, Yu
Start Date: 20201113
Expected Completion Date: 0
Actual Completion Date: 0

Subject/Index Terms

TRT Terms: Acoustic properties; Artificial intelligence; Computer vision; Defects; Excitation; Inspection; Lasers; Nondestructive tests; Pattern recognition systems; Spikes
Subject Areas: Maintenance and Preservation; Railroads; Safety and Human Factors; Vehicles and Equipment;

Filing Info

Accession Number: 01757984
Record Type: Research project
Source Agency: Transportation Research Board
Contract Numbers: Project 44
Files: TRB, RiP
Created Date: Nov 16 2020 3:08PM