Dutch Prognostics Lab

The reliability and availability of technical systems is becoming important in increasingly more sectors. Businesses want to reduce maintenance costs while improving the performance of their systems. The large number of sensors in systems and the improved options for storing measurement data create new opportunities to identify deviations and trends in data using artificial intelligence. This can help with early detection and prediction of failures, enabling predictive maintenance. In practice, however, this proves to be challenging: data is often incomplete or of low quality. As a result, many existing methods are based on simulated or laboratory data. The Dutch Prognostic Lab wants to bridge this gap between theory and practice.

Objective 

The objective of the research is to develop a standard for collecting and documenting data from diagnostic and run-to-failure tests. By establishing a standardised data template, the usability and quality of the test data is guaranteed. The ultimate objective of the Dutch Prognostics Lab is to create a network of existing test setups at various organisations. An example is the Bearing Test System (BETSY) of The Hague University of Applied Sciences, developed by the Smart Sensor Systems research group in the NWA ORC project PrimaVera. This network enables the sharing of high-quality data, allowing organisations to develop and validate their methods based on realistic data. This creates a widely applicable and shared knowledge base.

Research  

Because data collection during real production processes is often not possible, controlled laboratory setups are required. However, designing, building and using these experimental setups is expensive and often focused on one specific application. That is why the Dutch Prognostics Lab focuses on standardising measurement and documentation protocols in order to enable different organisations to share and utilise their data. After establishing this standard, existing test setups from different partners are connected to each other. This results in a distributed testing facility with coordinated testing and central data exchange. researchers of the Smart Sensor Systems research group contribute to the development of standardisation methods and test these on data from their bearing test system, BETSY.  At a later stage, international collaboration and data sharing are explored. To this end, guidelines for data access and an appropriate reward structure are being drawn up. 

Expected results

A shared, standardised infrastructure for collecting realistic, high-quality test data for diagnostics and prognostics. The collaboration enables organisations to benefit from each other's setups and datasets, leading to faster development and validation of predictive maintenance methods. Ultimately, this contributes to more reliable systems and lower maintenance costs in the sector.

Collaboration partners

• University of Twente
• Amsterdam University of Applied Sciences
• NLR  
• Ministry of Defence
• Rijnland Water Authority

Duration of the project 

2024-2025

Funding

This project is funded by  the Ministry of Defence (MIND)

Team 

The Hague University of Applied Sciences – research group Smart Sensor Systems:

• dr. Sam Aerts, associate professor  
• Xavier van Rijnsoever, researcher
• Derek Land, senior researcher
• dr. Amey Vasulkar, AI and data science expert
• dr. John Bolte, professor Smart Sensor Systems en director of the Centre of Expertise Digital Operations & Finance 

Contact 

Do you want to know more about the Dutch Prognostics Lab? Please contact:  Sam Aerts.