TNO
About this position
There are major challenges in the technical industry waiting to be solved in the coming years. Think about the structural safety of structures; climate adaptation; renovation and replacement of infrastructure; the energy transition; material re-use; offshore constructions; and digitalisation. For example, the effect of increasing traffic intensity and vehicle loads on current, sometimes outdated infrastructure, new types of fuels for shipping and the effects on the integrity of the structure, the effect of the energy transition on the way we deal with sustainability in the design or re-use of structural elements or materials, or the design of increasingly larger offshore wind turbines. These are major social issues that TNO does not shy away from, but rather tackles them by combining many different areas of expertise in in-depth, independent research. This way TNO is at the basis of the implementation of technological innovations.
What will be your role?
The aging civil infrastructure in the Netherlands poses severe challenges regarding the replacement and/or renovation of structures that reach their end-of-service life. A typical example constitutes the Dutch bridges which encapsulate multiple complications regarding safety-related consequences as well as highly expensive and time-consuming computational models to accurately arrive at maintenance strategies. Machine learning methods offer an efficient alternative to the traditional detailed physics-based models; However, they rely heavily on the amount of data available.
To this end, TNO is exploring two approaches to circumvent the large data dependency. On the one hand investigating the benefits of incorporating physics in such models to reduce the dependency on large amounts of data. On the other hand, TNO is building exact Predictive Twin models, such as a real bridge, which can facilitate generating large amounts of new real-life sensor data. By combining the two methods, sufficient data can be generated for the reduced data dependent models. With the aid of such a model, more accurate and realistic predictions can be made about the remaining service life of a structure and a projection of its current deterioration in a much shorter time frame.
A plethora of interesting avenues exists for solving such a problem formulation, such as investigating the most efficient way to include physics in machine learning models (Physics Informed Neural Networks, Variational Autoencoders, etc); Addressing and quantifying the ever-present uncertainties in such high-complexity physical systems in an efficient and accurate manner (coupling existing state-of-the-art machine learning algorithms with Bayesian principles); Validating these machine learning models using predictive twins, quantifying how well they describe reality and estimating the reliability of their predictions.
What we expect from you
We offer a graduation position to students that:
Of course, we may expect additional knowledge and skills depending on the research topic you are interested in.
What you’ll get in return
You want an internship opportunity on the precursor of your career; an internship gives you an opportunity to take a good look at your prospective future employer. TNO goes a step further. It’s not just looking that interests us; you and your knowledge are essential to our innovation. That’s why we attach a great deal of value to your personal and professional development. You will, of course, be properly supervised during your work placement and be given the scope for you to get the best out of yourself. Furthermore, we provide:
