EFFICIENT and optimised rolling stock maintenance and refurbishment is essential if rail is to increase service reliability, improve the customer experience and reduce the whole life cost of rolling stock. Progress has been made both in Italy and Britain to better track and respond to rolling stock faults to improve reliability.
Trenitalia has 50 maintenance sites which employ 5000 workers. Trenitalia is the owner and entity in charge of maintenance (ECM) for the rolling stock it operates, and it is this set-up that allows it to optimise maintenance plans and refurbishment.
Trenitalia’s vision for several years has been to switch to data-driven maintenance. The Dynamic Maintenance Management System (DMMS) project combines on-board information from trains with data from lineside systems. This data is integrated through monitoring systems, diagnostic rules, algorithms and health indicators to enable dynamic maintenance.
The system is now used to manage 25 fleets. Each vehicle continuously downloads data, streaming between 3000 and 5000 signals showing the status of the rolling stock and its components. More than 2200 active diagnostic rules on the system monitor the fleet and automatically generate maintenance notifications or operation alerts, while 1000 monitoring dashboards have been developed, and train maintenance plans have been modified to include dynamic maintenance procedures.
The next steps will be the construction of a digital twin for combined tracking of operations and components, and the addition of feeds from more depot equipment. This includes the Autonomous Robotic Inspection of Rolling Stock (Argo) system developed and patented by Trenitalia. The robot makes accessing equipment under the train simpler and safer, enabling quick inspections to be carried out even when the train is away from a maintenance site, for example in stations. Argo uses a robotic arm equipped with additional thermal, ultrasonic and acoustic sensors, which allows images and data to be collected from under the train. Combined with the DMMS, this allows technical problems that are crucial for the availability and safety of trains to be predicted.
At the World Congress on Railway Research (WCRR) in Birmingham on June 6-10, Trenitalia will host a congress masterclass covering the progress and successes achieved so far, as well as outlining future plans for the enhancement of DMMS.
Rolling stock must evolve during its long life, so heavy maintenance and refurbishment provides a great opportunity to upgrade trains to provide new services to passengers, introduce new safety equipment, or increase train sustainability.
More than 2200 active diagnostic rules on the system monitor the fleet and automatically generate maintenance notifications or operation alerts.
Trenitalia has also designed, developed and installed a “Wi-Fi Fast” system to enable passengers travelling on its trains at 300km/h to experience the same reliability and bandwidth as at home. The system has now been installed on all 58 ETR 1000 Frecciarossa and 17 ETR 700 high-speed trains with a rollout underway on 59 Frecciarossa ETR 500 trains.
The innovative internet connection system simultaneously uses the 4G-MiMo cellular coverage offered by all cellular providers in the area the train is passing through. This solution, which is already 5G ready, allows the aggregation of the bandwidth capacity of the various providers. This guarantees that passengers on Trenitalia trains connected to onboard Wi-Fi can surf the internet, watch movies, listen to music or have an international video conference. This system has already led to a significant improvement in the perception of travel quality for passengers. It also allows data for the DMMS or from onboard CCTV to be transferred via a stable and high-quality connection to the ground.
This service was recently activated outside of Italy and is now available on Trenitalia’s Milan - Paris service, which is operated with ETR 1000 trains. The technology will also be made available as a “plus” service on future Trenitalia trains operating in Spain, Britain and Greece.
To make its services more environmentally sustainable, Trenitalia is retrofitting part of its fleet of 750 class E464 locomotives with batteries to speed up shunting on non-electrified routes, enable trains to be rescued in case of problems with the catenary, and to allow short sections of battery operation in areas with particular constraints. An example is loading the train ferry linking the Italian mainland to Sicily, where using a battery locomotive will save an hour on the journey time.
The E464 locomotive is well suited to being retrofitted, as it is a very light single cab locomotive with free space at the back. The removal of ballast has made it possible to install batteries totalling 120kWh which, along with the converter and cooling systems, only added 500kg. The locomotive provides 400kW of power when in battery mode.
Research is also underway to enhance the reliability of Britain’s passenger fleets, where reliability improved between 2007 and 2013 but has since slowed. This is perhaps in part due to “settling in” periods as new trains have been introduced. However, there are a range of avenues that are being pursued by rolling stock owners, operators and researchers to continue to drive improvements in train reliability and improve maintenance efficiency.
In recent years the Institute of Railway Research (IRR) at Britain’s University of Huddersfield has carried out research to improve the efficiency and effectiveness of train monitoring and maintenance. These projects have led to the development of new Remote Condition Monitoring (RCM) systems; modelled component degradation; established methods to make better use of existing RCM data; and linked RCM outputs to the maintenance decision process.
IRR has partnered with the Rail Safety and Standards Board (RSSB) on several research projects to inform changes to industry standards and develop new industry guidance aimed at extending the life of rolling stock components and improving safety driven monitoring and maintenance practices, for both passenger and freight vehicles. A very successful example of the former is the work done on economic tyre turning, while examples of the latter include enabling the use of wheel impact load data to identify and characterise freight wagon defects, and the recent work completed on wheel flange wear limits.
IRR researchers will be presenting three papers at WCRR on this topic: a framework for locomotive bogie condition-based maintenance (Locate); optimisation of wheelset maintenance strategies using a combination of physical and data-driven models; and using wheel impact load detector data for the identification of vehicle defects in freight wagons.
In September 2019, the IRR launched the development of the Smart Rolling Stock Maintenance Research Facility (SRSMRF), funded by the European Regional Development Fund (ERDF), creating a focus for the existing research into rolling stock maintenance and expanding IRR’s capabilities in this area. The SRSMRF has three workstreams: robotics and automation for rolling stock maintenance; condition monitoring and prognostics; and virtual depot software for workflow optimisation.
The SRSMRF robotics and automation workstream is developing the use of industrial robots and collaborative robots for inspection and regular monitoring, refuelling, topping up fluids or sand and emptying CET tanks, overhaul of components “off-train,” cleaning underframes or cab fronts, and the removal and replacement of components. A new lab has been built which is configured to represent a short section of a maintenance depot with an inspection pit. Two industrial robots on linear tracks are installed either side of the pit with a third collaborative robot on a linear track in the pit.
The SRSMRF condition monitoring and prognostics workstream will develop new condition monitoring hardware where necessary; analyse existing RCM data to develop more detailed and fully validated degradation models; and enhance tools to link condition monitoring data to maintenance planning.
The SRSMRF virtual depot software is being developed to optimise workflow planning in maintenance depots. The tool will represent a model of a train fleet, including asset condition, depot resources and layout, and maintenance rules. The tool will be used to investigate how depot workflow could be optimised and how prognostic maintenance could be implemented in practice, whilst making best use of depot resources and allowing for depot constraints. Delegates attending WCRR 2022 will be able to take a 3D virtual tour of the SRSMRF lab when visiting the University of Huddersfield exhibition stand.
WCRR 2022 will be held in Birmingham, Britain, on June 6-10. To register to attend, visit www.wcrr2022.co.uk