What is it about?
Large steel infrastructures, like bridges, tunnels and sheet piles for water ingress protection are essential for modern society. Many of these existing structures are subject to degradation, resulting in structural safety issues. This has a negative impact on our comfort and well-being due to unforeseen (partial) closing of structures and large associated investments. Many developed countries are struggling with this issue and it is urgent in The Netherlands because of the large number of heavy vehicles, methods of construction in previous times and brackish water conditions. Consequently, the Dutch Ministry of Infrastructure and Water Management is now budgeting large investments for renovation and replacement requiring 3 to 5 billion euros annually and causing large disruption due to downtime. In addition to these economic implications, steel production requires a vast amount of raw material and energy. This conflicts with the Dutch aim of achieving a sustainable economy and meeting the Paris climate agreements. Central question is, how can we increase availability, reduce costs and contribute to enhanced circularity for these structures?
Our ambition is to develop a Digital Twin platform with data handling techniques and Structural Prediction Models that allow for the integration of Structural Health Monitoring. This will allow the analysis of the current structural state, thereby reducing the degradation uncertainty and predicting the future structural state, including the remaining life span.
Three major objectives are set for the SUBLIME project.
- Integrate Structural Prediction Models and Structural Health Monitoring in one Digital Twin framework. This will enable assessing the structural condition of an entire structure. It will be helpful for asset management and will enable a vast extension of the safe life span compared to business as usual.
- Develop an integrated evaluation of the maintenance approach, where not only the technical possibilities but also the socio-economic and environmental consequences are quantified. This will lead to a significant improvement in making maintenance decisions as compared to business as usual.
- Reduce the environmental footprint of large structures and associated CO2 emissions by prolonging their service life.
In our approach we will combine data and physics in a surrogate model. Machine learning will be applied to train the SPM by the SHM data. We will reciprocally extend the DT with meso and macro environmental and socio-institutional analyses of the maintenance options, giving the necessary input for an optimized governance of sustainable and reliable steel structures.
I’m a full professor and chair of Aluminium Structures at Eindhoven University of Technology (TU/e). My expertise and prime research interests are in the area of advanced steel and aluminium structures, fatigue and fracture, structural design, joining technology and design, computational mechanics, and damage and fracture mechanics. I am involved as fatigue expert in the assessment and design of existing and new bridges. My main research interest is in combining measurements and theoretical models to understand and predict fatigue, with the aim of extending the life of structures and optimize their design.