Periodic Reporting for period 2 - LOCATE (Locomotive bOgie Condition mAinTEnance)
Reporting period: 2020-11-01 to 2022-04-30
It is the goal of LOCATE to contribute to an optimization of the availability of rolling stock, the quality of service, maintenance costs and return of investment.
The main objective of the LOCATE project is to replace as necessary as possible the preventive conditional or scheduled maintenance of mechanical parts of the bogie by predictive maintenance.
LOCATE contributes to overcome some of the historical challenges of maintenance subject, by:
1. Developing optimized condition-based maintenance strategies using dynamic tools locating and evaluating the impact on the such overall approach of CBM in maintenance planning and railway operations;
2. Developing intelligent tools integrated for the operation of the overall maintenance policy: supporting localization of faulty components, maintenance scheduling and integrating maintenance operations tasks into daily services, while ensuring appropriate inventory control of stock and spare parts and assigning maintenance crew/ technicians according to their skills/competences;
3. Testing and validating in practice an open architecture able to carry asset management data to the operator beyond the locomotive bogie, locating events and defects on freight wagons, track condition, etc.
4. Developing a minimal digital twin for the bogie system, based on vehicle dynamic simulations and post-processing, while considering the local requirements of the system. In this way, providing a comprehensive methodology to derive minimal digital twins of complex mechatronic railway systems;
5. Applying a cost-effective and reliability-based sensor design to locate defects and monitor structural integrity of critical and high cost components of the bogie following an in-depth analysis of freight-specific use cases;
6. Improving overall competitiveness of freight rail transport, increasing the freight reliability and availability, and providing a shift from inspection activities and associated costs to cost-effective remote defect localization and monitoring solutions.
The LOCATE project showed a good adherence between the digital twin and the models deployed and demonstrator measurements, providing and end-to-end framework, contributing directly to:
• Reduction of bogie maintenance costs by the decreased number of inspections and respective release of human resources.
• Predictive maintenance framework that will potentiate investment optimization and respective capacity increase.
• Improvement of system reliability and contribute to less disruptions of service using the predictive maintenance approach.
The LOCATE project was able to specify, through discussion with project consortium members, assessment of failure records and application of FMECA, the components to be monitored which have the greatest (positive) impact for Condition Based Maintenance (CBM). It was also possible, based on a preliminary installation of sensors and initial data collection exercise on a locomotive to evaluate and select the technologies to implement in the LOCATE monitoring system. It was also identified the Operational Constraints in a Maintenance Shop to consider on the development of the CBM Framework.
The first measurement campaign was key to support the development and consolidation of the LOCATE, including the sensors set and preliminary validation of the digital twin models.
A demonstrator was installed, and the end-to-end framework validated, including the sensors set, acquisition system, onboard unit and data processing, cloud services for data management and reporting and front-end user interface.
The data analysis showed a good adherence between the measured responses and the digital twin outputs with very promising results.
The exploitation of the main results can be mainly divided into:
-Sensor Set and Monitoring Strategy, which includes the types of sensors and sensors disposition, localization, and installation to provide the means for a bogie monitoring strategy for the different components.
- FMECA Analysis, Analysis of the failure modes and assessment of the impact of the different failure modes.
- Framework of Maintenance Scheduling and Predictive Maintenance Program implementation, Methods of decision making, and scheduling analysis and the tailoring needed for implementing Predictive Maintenance Programs.
- Computational Models and Digital Twin, Design of Computational Experiments based on data collections and Vehicle dynamic simulations for the digital twin modelling.
- Monitoring and Threshold rules Specification, Threshold rules to be used together with the digital model for the component status monitoring.
- Predictive Maintenance Framework, End-to-end framework of the bogie monitoring system in a comprehensive integration and deployment description.
This was performed with involvement of all partners and members of the Advisory Board and FR8Rail III Project, qualifying the project results applicability to a wider range of ECMs.
These product concepts are being exploited, specially by the SMEs partners, by including in their CBM products and service development roadmaps.
- Validated detailed Bogie components Digital Twins and respective surrogate models for resource-constrained computational environments and near-real time information;
- Validated set of sensors and post-processed data analysis with key conclusions with respect to the added value of the sensor data;
- Onboard unit Hardware and Software architecture for bogie condition monitoring;
- Optimization tool for the condition based maintenance operations.
The key project results enable the introduction of condition based maintenance in the bogie maintenance operations, providing the capability to extend the components life-time, decrease the amount of unecessary periodic maintenance inspections and optimize the overall maintenance operations. The involvement of an end-user like FGC was of paramount importance to guide and provide hints on the most significant problems and added value extraction.
The project impact can be mainly highlighted as:
- Improvement of system reliability and contribution to less disruptions of service and anticipating the problems to come.
- Predictive maintenance can, with less investment, contribute to a network in better operating conditions.
- The LOCATE technology enables maintenance policies decision supported on asset condition data and reduce manual decisions and controls.
- Better management of the costs (people intervening, spares) and assets life cycle management.
- LOCATE contributes directly to increased availabilities as outcome of the predictive maintenance approach and unnecessary and corrective maintenance activities optimization.