After three successful editions, the HYMOP consortium is organising a fourth and final edition. The seminar focuses on the challenges of optimising the operation and maintenance of a fleet of machines by means of data-driven analytics and hybrid modelling strategies. (The seminar will be held in English.)
Many companies that operate a fleet of machines and continuously monitor its operation, do not fully exploit the enormous amount of collected data. Usually this data is merely used by maintenance and service personnel for reactive maintenance. Some lead user companies have recently shifted to proactive maintenance (using advanced exploitation of data), but even in these cases the analysis is realised at individual machine level. This leaves a huge potential of untapped exploitation opportunities at fleet level. By comparing the behaviour of a single machine with respect to the one of its fleet, more detailed and more accurate knowledge can be extracted.
This event is organised in the context of the SBO project HYMOP, in which VUB, Sirris, KU Leuven, imec and the University of Antwerp study hyper-modelling strategies on multi-stream time-series data for operational optimisation, in collaboration with an industrial advisory committee. During this seminar, the technologies that were realised during the project will be detailed and the possibilities for industrial valorisation will be explored.
You can also read our blog post with information related to our mastercourse session on the same topic on September 25: here.
13:30: Introduction & setting the context
13:50: Industrial keynote: Benefits & challenges of fleet-based analytics in the railway industry by Mohamed Bentefrit, Televic Rail NV
- Mechatronics in railway
- Condition based and safety fleet monitoring
- Anomaly detection and classification
14:25: Overview of HYMOP research results on industrial use cases - part 1
1. Farm-wide monitoring of wind turbines (University of Antwerp and VUB)
- How can we combine advanced signal processing with machine learning approaches in a hybrid approach?
- What can we learn from status logs simultaneously recorded over an entire farm?
- Which patterns occur that are interesting on the farm-level?
- How can we fuse outcomes of these different methods together?
2. Real-world anomaly detection from different data sources and with the user in the loop (KU Leuven and University of Antwerp)
- How can we combine different data sources using interpretable patterns?
- How can we ameliorate existing anomaly detection models through expert feedback?
- How can we leverage multiple datasets to increase performance of anomaly detection models?
3. Fleet-based method for estimating the end of life in never failed before asset types (Sirris)
- How to estimate the remaining useful lifetime of assets that are operating in a wide variety of operating conditions?
- How can the fleet be leveraged to predict the remaining lifetime of new asset types?
- How to guarantee qualitative end-of-life predictions over time?
15:20: Coffee break with demonstrations from HYMOP project partners
- A tool for time series pattern mining and anomaly detection” (KU Leuven, University of Antwerp)
- FleetACE – A portal for fleet-based data-driven analysis (Sirris)
- Dynamic dashboards for predictive maintenance: visualizing what the user wants when and where he needs it (imec)
15:50: Overview of HYMOP research results on industrial use cases - part 2
4. Hyper-modelling stategies for anomaly detection at fleet level (imec)
- Which are the challenges and opportunities with fleet-based models?
- How can we use hypermodelling for fleet-based anomaly detection?
- How do create aggregate detectors on multiple hierarchical levels?
5. Characterising the operational behaviour of an industrial asset (Sirris)
- How can the fleet be exploited to improve data quality?
- What is concept drift and why should I care?
16:25: Industrial keynote: Smart Powertrain Diagnostics - from Hindsight to Foresight, by Dr. Bernhard Peischl, AVL List GmbH
- Road to Predictive Maintenance
- Failure Prediction Approach
- Applications & Benefits of Failure Prediction
17:00: Concluding remarks
17:15: Networking reception
- Members of the HYMOP Industrial Advisory Committee & HYMOP partners: free
- Non-members: € 100 (excl. VAT)
If you are a Flemish SME you can also make use of the KMO-portefeuille. The KMO-portefeuille should be requested at latest 14 days after the seminar has taken place. Please note that if you do not apply in time your request will be declined. (Sirris accreditation no: DV.O105154).
Sirris, MRC Gent, Technologiepark-Zwijnaarde 48, 9052 Zwijnaarde
Our general terms and conditions
Any cancellation has to be made by e-mail (firstname.lastname@example.org). Cancellations made before the 3 business days preceding the event are free of charge. After this deadline, 50% of the participation fee will be charged (incl. VAT). In case of cancellation the day itself, the full amount of the registration will be due. In case of 'no-show' the full amount of the registration fee will be due too. Replacement by a colleague is always possible if notified in advance by e-mail to email@example.com.
The HYMOP project is being carried out with the support of VLAIO.
With the support of