Railways Embrace New Technology & Automation
Connector companies are bringing new technology advances, including AI, to railways in the U.S. and abroad, making them safer, more efficient, and more cost-effective.
Much of the current rail infrastructure in the U.S. is based on systems that have been in use for half a century or more, due to the industry’s notable hesitation to adopt new technology. Similarly in Europe, where trains are far more prevalent, train operators err on the conservative side when it comes to improvements. Nevertheless, several U.S. high-speed rail projects are in the works, including Amtrak’s Acela between Washington, D.C., and Boston and the Brightline between Los Angeles and Las Vegas. Artificial intelligence (AI) is being developed to improve rail systems in Europe and the U.S.
Paul Johnson and DJ Niemiec, railway industry managers at WAGO, explained that in the U.S. cost is at least partly to blame for this reluctance to change. For starters, while the interstate highway system was funded by the government, rail projects have mainly been privately funded. Since moving freight was more lucrative for private railroad owners than ridership, said Johnson, passenger railway has grown at a slower rate in this country compared to Europe, where railways are supported by the government as critical infrastructure. Due to the limitations of the existing infrastructure, getting rail into cities that don’t already have it is difficult and lines between cities, such as the one from L.A. to Las Vegas, are extremely expensive to build.
“It’s also not very cost effective to take out something that’s been in service for 50 years,” said Niemiec. Nevertheless, he and Johnson agreed that change is starting to happen.
“As the management teams get younger, they’re more willing to try different technology. We are seeing a shift,” said Johnson. “We had one of the young vice presidents of engineering say that he’s ready for a generational evolution and when you look at what WAGO is doing compared to the old way of doing things, that’s exactly what it is, a generational evolution.”
Change is happening more readily in the new systems being built. “We’re getting a lot of traction in helping the suppliers and the end users develop solutions in conjunction with new projects,” said Niemiec. “Also, through the federal infrastructure bill, there is now a lot of money available for transit organizations to buy new rail cars and to upgrade their facilities.”
New technology offers efficiency, safety, and better information gathering
Products such as WAGO’s modular terminal blocks make it faster and easier to terminate underground cable or interior wiring, which means projects can be built out much sooner. Another factor that aids both speed and safety is touchsafe equipment. “The application in use today is not touchsafe. If someone drops a tool or gets their hands down in there, they could be shocked, or the tool could arc out,” said Niemiec. Newer products remove that risk making it safe to work anywhere around that location.
Automation is another area that is gaining attention. “We’ve found that the railroads are finally catching up and making funding decisions based on analytics like the rest of the world, and the more data they have the better,” said Johnson. WAGO’s Health Monitoring System is a general data recorder that monitors the health of a piece of equipment and can be tracked remotely. “We can put our equipment inside a location, like a railway crossing, and it alerts users when there’s something wrong based on the information that’s being collected. That information is logged and the historical data can be used to identify trends. Operators can create a maintenance matrix that allows them to see that something isn’t working the way it was three months earlier and know to go look at it. That information is not readily available today,” said Niemiec. “The nice thing about our system is that, because it is modular, it can go into older equipment as well and provide that same information.”
“The system is located within the infrastructure of the railway, inside the bungalow right next to the tracks at a highway crossing or a control point where the switch is. It can be accessed from anywhere with the right security and passwords,” said Niemiec. “Tri-Rail’s expectation is that it will have one desk where all the crossings on their railroad can be monitored. A city like Miami or Fort Lauderdale may have three or four different train lines running across the same street within a mile or so. When a call comes in from the public to say that the gates are down, for example, the caller doesn’t always know which railroad is responsible, so it may be an hour before the right people get to where they need to be. This system gives Tri-Rail the ability to know for sure whether their crossing is involved and get somebody out to look at it. It’s a general shift in how things are being done.”
AI gathers an extensive database to predict future outcomes
Schaltbau, based in Munich, is developing high voltage, high current DC contactors for rolling stock as well as for industrial applications. So far, these contactors are used in a kind of master-slave configuration, whereby the switch is always the command receiver. As part of a new development, the contactors are being supplemented with measurement equipment, an evaluation unit and a communication port. This turns the classic DC contactor into a smart contactor.
Guido Bachmann, director of R&D, electronics and software at Schaltbau, explained that system-wide energy efficiency requires synchronizing energy generation and energy usage. Reduction of the consumption of fossil fuels for electrical power generation requires complex forecast models to make the best use of regenerative energy sources like wind turbines and solar energy arrays. In addition to this, in rail applications electrical braking is considered an energy source, too.
“There we have two possibilities,” Bachmann said. One is battery storage or a flywheel that can be located on the train, or somewhere in the infrastructure, to store the energy generated by the braking train. The other is to synchronize trains so that the braking trains that are arriving at the station deliver the energy to accelerate the trains that are leaving the station. Maximizing that synchronization is the challenge. “This can be done with a cloud-based computing system, but we need data,” he said. “Our idea now is to go away from the standard switch, a master-slave configuration, to a system with an open or closing contact and sensing. We learned in the past that our systems now create a huge amount of data, so we have a big data lake. Now we have started to train some AI algorithms to identify certain patterns in our overall system.” By taking measurements of, for example, voltage, current, and temperature without the need for additional measurement equipment, the AI algorithms have a source of information that allows them to learn from the system’s past behavior to predict the system’s behavior in the future.
In this context, AI helps reach energy-saving targets. Training AI algorithms requires reasonable data sets.
“We can then give information about the state of health of connected components, for example, of the battery. We can protect the switches themselves. We can identify whether some connected loads, maybe some drives, seem to be faulty,” Bachmann said. “We will collect the data. We will make the pre-evaluation of the data for a kind of condition-based maintenance for all the connected loads.” Rather than relying on “best-guess” timeframes or mileage numbers, maintenance would be conducted based on identification of patterns developed from extensive evaluation of data-driven performance histories. In the future, intelligent loads will do this condition-based maintenance on their own. For the loads that don’t have communication and intelligence capabilities, AI structures will be able to predict their future behavior. This information will be used to increase the availability of rolling stock equipment. “Availability is the most important KPI for the train. Something might break, but if we predict the breakdown, we can organize the repair in time. So that’s our main idea behind this,” said Bachmann.
Unlike know-how that is lost when career employees leave with all their knowledge and experience, AI retains all its knowledge and can continue to build on it. “If I wrote down everything we know, I would need someone to read and understand all this. One of the big advantages of artificial intelligence is that this intelligence can be copied into the new structure, into the new system,” Bachmann said. “When we have this expert know-how from contactor A and from contactor B, and we can combine this and transfer it to a new, similar contactor C, then we have, immediately, a lot of pattern evaluation with a very high level of confidence.”
To learn more about the companies mentioned in this article, visit WAGO and Schaltbau.
Like this article? Check out our other Rail, Artificial Intelligence and Terminal Blocks articles, our Transportation Market Page, and our 2024 Article Archives.
Subscribe to our weekly e-newsletters, follow us on LinkedIn, Twitter, and Facebook, and check out our eBook archives for more applicable, expert-informed connectivity content.
- What are Variable Reluctance Speed Sensors? - October 8, 2024
- EMI/EMP Connectors Product Roundup - October 8, 2024
- October 2024 Connector Industry News - October 8, 2024