Looking into more efficient infrastructure, one of European rail’s ambition is to develop a system able to live-monitor the status of every rail asset in the network.
Knowing the status of bridges, tunnels and rails would not only help repair broken assets faster but engineers expect it could predict a failure before it happens, thanks to a prognosis algorithm making rail infrastructure upkeep more effective and sustainable.
The Shift2Rail Research & Innovation programme, worth €920 million euro, is managed by the Joint Undertaking, an institutional public-private partnership established by the European Union.
Shift2Rail’s dedicated rail infrastructure R&I programme aims to develop predictive maintenance schemes, but also smarter energy management strategies and disruptive technologies to make rail upkeep faster and further automated.
Shift2Rail’s size is especially important when it comes to infrastructure R&I, which often requires large investments.
“Historically, high cost barriers have prevented any significant change to the design of track and civil engineering such as embankments and bridges”, says Andy Doherty, CTO at Network Rail, Britain’s national rail network management entity.
The programme looks into the basic engineering, seeking new technology, new understandings of the physics of railways, and test-validated solutions ready for market uptake, Mr Doherty explains.
Harnessing the power of data
Today, technicians review rail assets periodically and repair them when broken. Thanks to live information from sensors placed on different elements, engineers expect to know in real time if an asset needs servicing leading to reduced downtime of infrastructure and rolling stock, which entails less delays for passengers and reduced use of energy.
“Automatic and fully integrated monitoring systems make available huge volumes of real-time heterogeneous data from different sources, paving the road for the application of big data techniques. The big challenge is to transform data into knowledge that allows the development of data-driven risk- aware decision support systems”, says Nadia Mazzino, Vice-President for Digital Railways and Innovative Technologies at Ansaldo STS, a rail manufacturer and Shift2Rail-Member.
In this context, the definition of an ‘integration layer’ allows the collection and exchange of the available data in a canonical data format. “It aims at supporting fusion, integration, and adaptation of different data sources, thus facilitating interoperability and development of new data-driven tools”, says Ms Mazzino.
Research in this field goes even a step further. Live information on the network, processed using Artificial Intelligence (AI) techniques, could support fast and well- informed decisions.
“Decisional processes, which often still rely on the skills of specialised human operators, will achieve a high level of automation through the introduction of AI, based on prescriptive analytics, computational models, and optimisation techniques”, explains Ms Mazzino.
The application of AI in the rail sector means shifting from descriptive analytics, which makes it possible to automatically collect and visualize the available data, to diagnostic and predictive analytics, which shows the current and future status of rail infrastructure, allowing for prescriptive analytics, developing automated tools to support decision-making.
AI would enable rail providers allow to offer additional services to customers. On the other hand, infrastructure managers and service operators could identify critical defects earlier and adapt quickly to unexpected events, improving service for passengers and making railway infrastructure more reliable, resilient and available.
Improving the rail energy flows
To improve network’s efficiency, Shift2Rail- funded researchers are also developing methodologies to provide network managers with real-time information on the use of energy of rail assets such as switches, tracks and catenaries.
One of Shift2Rail’s projects developed a proof of concept for ‘smart energy metering’ on the light rail network of Reims, France.
Researchers developed a solution for measuring the energy flow both on-board and on the electrical grid. Feasibility of this model, which makes use of AI, was tested on-board of tramways and on ground in a traction substation.
The tramways and the substation were equipped with current and voltage sensors, GPS, accelerometer, CO2 and temperature sensors, which compiled and sent their data to a central server.
Research aims to produce an exact map of how the energy flows within the entire railway system. This would enable engineers to better understand the systems’ energy usage and devise strategies to improve it, making railway systems more sustainable.
“Some interesting findings revealed that ‘low-hanging fruits’ in terms of improvements could be easily found and implemented if a system approach is undertaken”, says Marius Iordache, Simulations & Performances and Energy Efficiency Manager at Alstom, a member of Shift2Rail.
Disruptive ideas, welcome
Looking into new radical ideas is also an essential feature of Shift2Rail’s innovation programme. Some researchers are looking at nature for inspiration. It might seem odd, but some widespread solutions like Velcro are inspired by animal or plant behaviour.
Henk Samson, Senior Programme Manager for Business Development and Innovation at Strukton Rail, a Shift2Rail member, thinks ‘bio- mimicry’ (emulating nature with robotics and automation), might revolutionise railway and drops some ideas.
For example, inspired by spiders’ movement, autonomous rail vehicles could get in and out of the tracks just when needed, avoiding unnecessary occupation of rails. Another proposal: sets of hundreds of light and small robots could enter the track to perform maintenance tasks where power and strength are needed an idea based on swarm robotics, inspired by beehives and ant colonies.
“These ideas might be futuristic but there are some solutions that can be implemented in a shorter term”, says Mr Samson. “For example, adding some robot features and extra add-ons to excavators used on tracks to be used as robot platforms could make a start”.
Working for greener rails
Sustainability is at the core of the Shift2Rail programme as one of its goals is to halve carbon emissions produced by the rail sector.
“We believe that data-driven infrastructure management and smart energy metering are some of the most promising areas for making railways more sustainable”, says Carlo M. Borghini, Shift2Rail’s Executive Director.
“We are developing innovative solutions to improve railways’ sustainability even further. We work not only on infrastructure, but also to devise lighter train parts, reduce vibration and noise emissions, and create an IT eco- system to make railways more attractive to passengers; thus reducing transport emissions overall”, explains Mr Borghini.
“A shift to an even greener rail would be a huge contributor to the zero-emission future we aspire to in Europe”. The first phase of the Shift2Rail initiative, which launched its first projects in 2016 under the Horizon 2020 Programme, with some Lighthouse Projects already started in 2014, is due to launch its last call for new projects in 2020.
“Until now, we have set the foundations for a real shift to rail, developing cutting edge solutions in only two years of work at full capacity. We have solid ground to substantially continue investing in R&I beyond 2020 for more efficient, greener, digital, automated and integrated rail in Europe”.