RAILWAY AGE, DECEMBER 2021: Trackside inspection equipment helps railways ensure that rolling stock operates safely and efficiently.
Machine vision and artificial intelligence (AI) are among the tools that allow railways and transit agencies to automatically and proactively monitor the condition of rolling stock, identifying mechanical problems before they occur. do not become serious failures. Transportation Technology Center, Inc .; Duos Technologies; Trimble® Beena Vision® Solutions; and Cogniac shared with Age of the railway how it’s made.
Machine vision camera technology captures images of moving rail cars from various angles. With an inspection system from Duos Technologies, for example, images are processed through AI servers on the ground that look for issues defined by the railroad, such as brake pistons engaged. “The images from the camera that contain the brake piston are processed by the algorithm,” explains Scott Carns, Duos sales manager. “The results are passed from our system to the end user. All of this happens for each car in a matter of milliseconds as each passes through the system at track speed. “
All Class I railroads have machine vision systems on their networks, ranging from brake shoe and wheel inspection systems to full inspection gates “that picture the whole train in 360 degrees”, explains Matt Witte, TTCI scientist. Among the benefits: “Higher and more consistent performance for repetitive inspection tasks versus inherent human limitations. We want to use our skilled workforce where it can provide the most benefit by repairing rather than inspecting rail cars. Inspection systems also offer greater throughput and have the potential to increase average train speeds by reducing the dwell time in the yard for manual inspections.
– “The completeness of the data flow from machine vision inspection helps track the condition of components,” says Matt Witte, TTCI. –
“The completeness of the data flow from machine vision inspection helps track the condition of components,” adds Witte. “The end result is a way to predict when preventative maintenance can be performed on a railcar just before the end of its useful life. ”
Other benefits include reducing security risks to personnel, providing compliance statistics, and maintaining visibility across the fleet of component condition, says Timothy Francis, Sales Director Americas for Trimble Beena Vision.
Of course, there are also technological challenges. “Developing algorithms to automate fault detection is not always easy in the railway environment,” explains Witte of TTCI. “Variations in the design of cars and components make the observation, detection and evaluation of acceptable baseline conditions a far greater challenge than the detection of product defects in a manufacturing environment where the machine vision inspection is common. And reliable inspection requires a clear view of the components to be inspected. Common variations in railcar hardware, such as the placement of handbrake components, can block the optical path between the target and the camera. When combined with adverse environmental conditions such as rain or snow, this obscured line of sight can affect the performance of the machine vision system.
Railways use track inspection technology like Duos’ – which incorporates a canopy to help mitigate environmental factors – to increase safety and improve speed, but the application ultimately comes down to their pain points. specific, explains Scott Carns. And the positioning of the system tends to be well outside the yard and more on the main track, he adds. “Our measurements have shown that a good rule of thumb is about one system per 1,000 kilometers of road.”
How do railways capture data? “With the volume of data generated by our systems, we are currently deploying data centers on the ground,” he says. “All data is captured, processed and stored at the edge. From there, we developed Application Programming Interfaces (APIs) that integrate directly into existing rail systems for maintenance reporting and bad order labeling. Currently, because this type of technology is at the cutting edge of technology, the general philosophy is to leverage systems as a new source of data into existing processes throughout their organization.
Duos’ latest technology is ObliqueVUE, a crosshead-mounted system that captures two-way images from eight cameras and “sees” at least 25 new inspection points, Carns notes. “We have also started implementing the latest 8K line scan and 5MP area scan cameras that allow us to image at speeds above 125 mph. “
What’s in the pipeline? For one customer, Duos is currently expanding the number of imaging perspectives from nine to 34, allowing them to “see” more than 100 inspection points on each car.
“Another very exciting opportunity is the creation and short-term implementation of a ‘fast lane’ for rail border crossings,” Carns said. “The intention here is to provide all inspection data to various federal agencies in partnership with the railways to enable them to change the current process on how trains pass through our land border crossings in the United States. . Ultimately, this will dramatically improve speed, allow for higher throughput, and help the various law enforcement agencies responsible for this effort. ”
Trimble Beena Vision Solutions
Trimble’s non-contact wayside measurement and inspection technologies assess the condition of rolling stock from component level to a full train inspection. Rail operators use the data generated to prioritize train maintenance and derailment prevention, explains Timothy Francis. “In addition, the correlation between the data generated by different detectors provides additional opportunities to analyze and understand the condition of any individual rolling stock or of the entire fleet. “
The adoption of condition monitoring continues to grow, says Francis, noting that the company has installed numerous systems in all Class I vehicles and at several global operators, such as Aurizon, BHP, Rio Tinto and FMG. in Australia ; SNCF in France; VR in Finland; Vale, VLI and MRS in Brazil; and Etihad in the United Arab Emirates.
Trimble’s latest offering is Trimble® TreadView®, a non-contact automatic optical inspection system that images and inspects the wheel surface (tread, flange and plate surface) at operating speeds of the wheel. main line and in all ambient lighting and weather conditions. “The high resolution images and high density 3D data of the wheel surface are used to determine any abnormalities in the outer surface of the wheel tread,” explains Francis. “The processed data and images from the Trimble TreadView system are integrated with the Trimble CMMS ™ (Condition Monitoring Management System) software to provide web access for data viewing, alarm management and data analysis. ”
Pulling actionable insights from trackside data is the key to improving railroad efficiency and reliability. “With improved and faster data preparation and contextualization, as well as the application of smart analytics, data can be modeled to show trends and patterns that would otherwise remain hidden,” says Francis. “Trimble is working within the rail industry to create meaningful predictive analytics that reduce risk and improve efficiency. “
“Cogniac’s solution enables rail companies to assess, in near real time, train wheels and tracks at speeds of up to 60 mph,” says the company, which is deploying its AI machine vision platform “to enable railways to process the images they capture on-site and in a minute, providing near-instant assessment of critical assets in motion.
Cogniac is currently working with a Class I and, on a monthly basis, monitors 22 million wheels and 32,500 miles of track. The company “processes hundreds of thousands of images taken by cameras at the front of more than 450 moving trains every month. Images are sent by Cogniac’s EdgeFlow and evaluated for cracks, cracks or missing bolts. A human expert is alerted if any images are flagged as faulty. The human supervisor can then make a decision on how to proceed with the identified fault.
– “Since January 2020, Cogniac’s vision system has stopped more than 100 trains where there was a potentially devastating problem that could lead to a derailment, sparing the railroad up to $ 350 million in damage. –
Cogniac also worked with this railroad to install trackside gantries, taking high-resolution images of wheels. Images are processed and those showing cracks or other issues are sent to a human inspector for secondary examination. “As of January 2020, Cogniac’s vision system has stopped over 100 trains where there was a potentially devastating problem that could lead to a derailment, saving up to $ 350 million in damage to the railroad,” said Cogniac.
“The next big hurdle for the industry is regulatory modernization,” said Witte of TTCI. Age of the railway. “Historically developed regulations are not necessarily structured to take full advantage of the safety, repeatability and efficiency benefits offered by automated inspection. Through AAR committees, railways strive to demonstrate and quantify the safety benefits of automated inspections. TTCI supports this effort by objectively evaluating the performance of inspection systems in a controlled environment where known defects can be safely included in the population of cars being tested.