Authorized maintenance with the use of trackside surveillance systems
Maximize the use of existing trackside monitoring systems for issues related to Condition Based Maintenance (CBM).
Detecting the degradation of rolling stock at early stages is important to avoid stops, stalled wagons and delays that are costly and very inconvenient, as they dramatically reduce infrastructure capacity and vehicle availability. Therefore, it is of utmost importance to assess vehicle degradation using appropriate condition monitoring techniques. The condition monitoring (CM) applied in railway systems is divided into two categories: trackside monitoring systems and on-board monitoring systems. In general, on-board systems are suitable for fast evolving defects or for defects that are not detectable with trackside equipment. For slowly evolving defects, trackside monitoring systems have the advantage of not requiring large-scale installation of measuring equipment on many vehicles. In general, trackside monitoring systems have traditionally focused on detecting critical events that may lead to further failures, damage to the infrastructure or disruption of traffic.
To increase the usefulness of current trackside monitoring systems, new strategies should be developed to extract information related to the wheel sets, not only for safety-related issues but also to improve condition monitoring from the perspective of vehicle owner maintenance. As a result, methods to extract relevant information have been developed in the framework of the FR8RAIL project. A deeper analysis of Hot-Box/Hot-Wheels and Wheel Impact Detectors (WILD) has been performed on a specific railway line from Aitik to Skellefteå in northern Sweden. The study focuses on several key points that can improve the condition monitoring of wheelsets:
- Ensure the quality of transmitted data between different detectors, as discrepancies can occur at different locations (variation between HB types, position along the line, load impact, seasonality, dynamic characteristics...). These deviations can lead to false alarms and affect preset functions used for prognostics.
- Development of a concept (Figure 1) based on extracting temperature signatures of trains or wagons from Hot-Box/Hot-Wheel measurements. This concept can help to detect anomalies related to a change of condition (lubrication, dirt, misalignment or unknown causes) without reaching the current safety limits.
- Select the appropriate function according to wagon type for Wheel Impact Detectors. Train dynamics, load and speed may influence the readings of these specific functions.
- Investigate - from the available data - specific wheels or wagons that have slower degradation processes, which can be followed using the dynamic load as the main feature (Figure 2). The degradation model can depend on the type of failure (stepwise, linear, exponential...). A labeling of failure type and maintenance for specific wheels could be interesting to develop degradation models using trackside monitoring systems.
- By combining the quality control of different detectors, the transformation of features as a function of specific predictors and tracking well-chosen features from RFID tagged wagons as a function of time or mileage, the transition to a proactive approach may be feasible, especially for Wheel Impact Load detectors. To ensure that case studies showing a slow degradation process are related to wheel set degradation, the infrastructure manager should be in close relation with the vehicles' owners to tag the current data set with respect to fault types and maintenance activities. This interaction would lead to more accurate degradation models, which could result in a better estimation of the remaining useful life.
Sponsor: FR8RAIL - Swedish Transport Administration
Researchers: Matti Rantatalo (PL), Florian Thiery
Updated: