Frequency Based Approach For The Detection of Railway Tracks Faults in Pakistan
Pakistan has quite a shoddy record when it comes to railway accidents. Within the past decade, the country has witnessed several catastrophic train accidents, and the number has increased over the past few years. Following are some significant accidents witnessed in the history of Pakistan:
2025-06-28 16:32:41 - Adil Khan
Frequency Based Approach For The Detection of Railway Tracks Faults in Pakistan
Project Area of Specialization RoboticsProject SummaryPakistan has quite a shoddy record when it comes to railway accidents. Within the past decade, the country has witnessed several catastrophic train accidents, and the number has increased over the past few years. Following are some significant accidents witnessed in the history of Pakistan:
- January 4, 1990: The Bahauddin Zakaria Express was supposed to pass through Sangi village, but incorrectly-set Switches sent it into a siding, and it struck an empty 67-car freight train at a speed of at least 56 km/h. Around 20 people died.
- July 2, 2015: At least 20 people died when three carriages of a special train fell into a canal, and another passenger train derailed near Gujranwala.
- September 27, 2018: A Peshawar-bound train had derailed in Sehwan, Sindh, due to which 11 bogies overturned.
According to BBC News, there are three leading causes of railway accidents in Pakistan, among which "Lack of Track Maintenance" tops the list. Statistics shows, half of the accidents occur near the railway junctions, and most of them are caused due to train derailments, which occurs due to damaged railway tracks. Therefore, to ensure passenger's safety and avoid disruption in services, the entire rail needs to be monitored periodically. Unfortunately, the current solution to detect track faults in Pakistan is "Manual Rail Inspection". The railway management team makes a visual survey of the track with a handheld trolley or a small bogie. This monitoring method is labor-intensive, time taking and prone to errors as it is hard to get creditable inspection results from the human eye.
Our project aims to solve this problem by implementing an automated system in a survey vehicle to detect gaps or damages in tracks. The device will install on the bogie's axle box. The model will initially store the expected bogie frequency. The inspection vehicle periodically travels over the rail network, and with the aid of accelerometer sensors, it continuously measures the frequency. The location where the expected frequency of recording vehicle gets disturbed is recorded using a GPS module. Although this model will only be a prototype for now, yet it would be sufficient to show how we can proceed to make the Railway Monitoring Systems in Pakistan completely automatic only with the assistance of Electronic devices. This automated system is expected to increase the robustness of safety systems in railway transport by providing effective results compared to Manual Monitoring.
Project ObjectivesOur project aims to establish a proper Railway Track Monitoring system in Pakistan with the aid of electronic devices, so that faults in railway tracks can be easily and efficiently detected and maintenance can be done at the right time when needed, This as result will help us reduce the number of disastrous accidents.
Following are the short term goals or objectives of our project:
- To design a digital 3D model that can easily fit on a bogie's axle box according to our desired dimensions.
- To print that model and start building a working device using raspberry pi and accelerometer sensors in order to record the change in acceleration of the inspection vehicle.
- Utilize the measurement methods of Axle Box Acceleration (ABA) for easy and cost-effective railway track monitoring.
- To employ the algorithms of Fast Fourier Transform (FFT) for converting data from the time domain to the frequency domain.
- To track down the reallocation of the damaged track where the frequency is disturbed, using GPS technology.
- To alert the nearby railway station about the damaged location of track so that maintenance of that area can start as soon as possible.
- Get effective and faster results as compared to Manual Rail Inspection.
- Make Railway Track Detection in Pakistan easy and economical.
Our project is based on detecting damages and faults in railway tracks with the aid of the IMU system. We will be using Axle Box Acceleration (ABA) measurement method, which is by far the most efficient approach for monitoring the railway tracks. We will start by designing a 3D model with a suitable dimension to be installed on a train bogie/Survey vehicle. The model will be designed using AutoCAD or SketchUp 3D. The designed model will then be sent for printing to Proto21 (a 3D printing services company). Once it is printed, it will serve as a base for our main components, i.e. Raspberry pi, GPS module and inertial sensors. Inertial sensors measure an object's acceleration or angular velocity along three mutually perpendicular axes (X, Y, and Z). We will use the accelerometer inertial sensors in our project. The accelerometer sensors will continuously monitor the acceleration of the bogie during the track survey. That acceleration will then be changed from the time domain to frequency using the FFT algorithms installed in the raspberry pi. The raspberry pi is connected to the GPS module which records the real-time location, as aforementioned. The complete model can be installed on the bogie’s axle box, which inspects the rail network periodically. The expected frequency of bogie will be already stored on the system so that system may know when the bogie’s normal frequency is disturbed. The location where the expected normal frequency of bogie is disturbed or changed is recorded by the GPS and is redirected to a nearby railway station. Our model is a minor attempt to determine how much efficient results the fully Automated Railway Maintenance System will yield.
Benefits of using our automated Railway Monitoring System are:
- This prototype model can lead to the idea of converting the Manual Rail Inspection System in Pakistan into an Automated Rail Inspection Approach, only with the help of electronics devices.
- The railway maintenance labor-cost will be decreased.
- Our Model will minimize the track inspection time and assist the survey technicians in monitoring the tracks smartly.
- Less chances of errors as the inspection is carried out by machines and not humans.
- The main advantage of using Axle Box Acceleration (ABA) measurements over other railway condition-monitoring systems are its lower cost, ease of maintenance, and implementation in trains running at commercial speeds.
- We can further improve the system to set up Automatic alerts, if the train diverges from its route or start behaving erratically – such as in case of an engine failure.
- By implementing this type of monitoring systems in Pakistani Railway tracks, we can prevent disastrous accidents and ensure train security.
The final product is in the form of an electronic/hardware model or device which can be installed on the axle box of an inspection vehicle or bogie to provide the basis for an automated railway track monitoring system for the detection of joint or switch gaps and track damages. The final version of our device will be composed of raspberry pi, multiple three-axis acceleration sensors and Global Positioning System (GPS) module, interconnected with the help of multiple jumper wires. The device will use algorithms of Fast Fourier Transform (FFT) to convert the acceleration inputs recorded in time-domain to the frequency-domain. The frequency-domain output will then be compared with the regular expected frequency of the inspection bogie/train (which is already stored in system). If it matches, then the bogie will continue its survey. In case, at some point, the recorded frequency does not match with the original frequency. Then that location will be redirected to the nearby railway junction so that the officials can take measures to redress the damaged or gapped track.
Final Deliverable of the Project Hardware SystemCore Industry ITOther Industries IT , Transportation Core Technology RoboticsOther Technologies OthersSustainable Development Goals Affordable and Clean Energy, Decent Work and Economic Growth, Industry, Innovation and Infrastructure, Sustainable Cities and Communities, Responsible Consumption and ProductionRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 70000 | |||
| Raspberry pi 3 | Equipment | 2 | 8000 | 16000 |
| 64-GB SD (Memory card) for Raspberry pi | Equipment | 2 | 1600 | 3200 |
| Accelerometer 335 | Equipment | 6 | 550 | 3300 |
| GPS module | Equipment | 2 | 2000 | 4000 |
| Proto 21 3D model | Equipment | 1 | 25000 | 25000 |
| HDMI cable for Raspberry pi | Equipment | 2 | 900 | 1800 |
| Soldering Iron | Equipment | 1 | 400 | 400 |
| Vero Board | Equipment | 2 | 300 | 600 |
| Bread board | Equipment | 2 | 350 | 700 |
| Glue Gun | Equipment | 1 | 500 | 500 |
| Glue Gun Sticks | Equipment | 20 | 30 | 600 |
| Computer LCD | Equipment | 1 | 3000 | 3000 |
| Keyboard and Mouse | Equipment | 1 | 500 | 500 |
| Jumper Wires | Equipment | 30 | 30 | 900 |
| Soldering Wire | Equipment | 10 | 50 | 500 |
| 12V Dry Battery | Equipment | 1 | 2000 | 2000 |
| Poster printing | Miscellaneous | 1 | 1000 | 1000 |
| Thesis Book printing | Miscellaneous | 1 | 3000 | 3000 |
| Project survey cost | Miscellaneous | 3 | 1000 | 3000 |