Smart Rail Crack Detection System

I&T Wish Smart Rail Crack Detection System
(REF : W-0530)
Matched I&T Solution
Trial Project
Summary and Challenges
  • This innovation project presents a trial that leverages a combination of technologies to monitor and assess the structure integrity of railway tracks, both internally and externally, specifically focusing on detecting rail cracks. The passenger train travelling at maximum 80kph will be equipped with different types of sensor and computing equipment, enabling it to scan railway track, during regular traffic hours. The collected data will then undergo AI analysis to automatically compare and evaluate the health conditions of the rails, swiftly identifying any signs of flaw, cracks or broken sections. 
  • Project Challenges:
    • Technology Integration: Installation and integrating different sensing and computing equipment onto a passenger train seamlessly may pose technical challenges.
    • Data Collection and Processing: Collecting accurate and comprehensive data from the scanning process at a passenger train travelling at speed upto 80 kph across the designated railway track segment can be complex. Ensuring the smooth transmission and processing of this data for AI analysis is crucial.
    • Real-Time Analysis: Performing AI analysis in real-time during normal traffic hours requires high computational power and efficient processing capabilities. Ensuring timely results and responses to potential rail cracks is essential for effective maintenance and safety.
    • Localisation: Accurately identifying and localising the imaging of rail crack, defects or flaw is essential for the maintenance team to efficiently plan and execute repair work. A mobile network signal may not be feasible inside a tunnel.
    • Environmental Factors: Factors such as dark environment inside tunnel or during night time, weather conditions, vibrations, and varying track conditions can impact the accuracy of the monitoring system.
  • Remarks: The solution provider should provide the I&T solution proposal to EMSD on or before the closing date. It should be included the preliminary design proposal with the block diagram, workflow, proven project reference and budgetary ballpark cost estimate. EMSD may invite the proposer to conduct a presentation for introducing the proposal.
Expected Outcome
  • The equipment to be installed on a passenger train can timely detect and predict abnormalities of rail crack or broken rail and prompt remedial action to be taken, thus preventing railway incidents. 
  • The success of the project depends on the AI algorithm's ability and the front-end sensing equipment detection capability.
Expected Trial Duration 12-month
Contact Information
I&T Wish Proposer:Electrical and Mechanical Services Department (EMSD)
Contact Person:Mr Kenneth NG
Position:Engineer/Railways 9/1
Tel: 3757 6288
Email: scng@emsd.gov.hk
Upload Date 2024-04-08
Closing Date 2024-05-08