Internship: Fast Physics

We offer you an Ocean of Possibilities. Join our family.

About Us Damen aims to become the world's most sustainable and digitally connected shipyard. The Research, Development & Innovation (RD&I) department develops and implements the technology and know-how to achieve these ambitions. We actively assist the business in creating an innovative product portfolio and provide forward-thinking guidance to improve the quality and performance of Damen's products and services.  You will be joining the Data Science team within Damen RD&I, located in Gorinchem. Our department focuses on applying cutting-edge data and AI solutions to Damen’s shipbuilding and maritime operations.  The team includes domain experts in physics-informed machine learning, simulation acceleration, predictive maintenance, computer vision, and operational analytics. This internship is part of a strategic project aimed at accelerating complex simulations for ship performance using machine learning and graph-based AI.

The role As an intern, you will work on the Fast Physics project, which aims to drastically reduce the runtime of high-fidelity computational fluid dynamics (CFD) simulations of ship hulls. These simulations are essential in predicting how a vessel behaves in water, but they can take hours to compute.  Instead of running time-consuming physics-based simulations, we use geometric deep learning, a type of machine learning that can learn from vessel designs and quickly estimate results like water resistance or flow around the hull.  The outcome is a working prototype that can support early-stage design exploration and simulation optimization.

You will contribute to enhancing the performance of an existing system that predicts physical quantities, such as ship resistance and flow fields, based on geometry and operating conditions. Your primary focus will be on a dedicated topic involving the training, validation, and extension of the framework to support multiple ship types and/or varying levels of simulation fidelity. The assignment can be a thesis/graduate internship and could start from September onwards.

Key accountabilities You will be responsible for the following aspects:

• Support the improvement of ML-based frameworks, focusing on geometric deep learning and graph neural networks.
• Preprocess CFD simulation data and ship hull geometries.
• Run experiments in Python using PyTorch.
• Work closely  in our team together with Data Scientists, domain knowledge naval architects, and external partners such as MARIN.
• Document results and present findings to the team regularly.

Skills & Experience We are looking for a student who:

• Is currently pursuing an Bachelor or Master in Mechanical Engineering, Applied Mathematics, Computer Science, Data Science or a related technical field.
• Has experience with Python, and ideally deep learning frameworks such as PyTorch or TensorFlow.
• Has familiarity with 3D geometry formats or CFD simulation and numerical data.
• Has a strong interest in physics-based modeling and applying AI to engineering problems.
• Communicates fluently in English.

What We Offer

• Mentoring at academic level will be available throughout the internship.
• Internship/graduation fee and travel allowance will be paid for the duration of the assignment.
• Opportunity to contribute to a high-impact innovation project in collaboration with leading maritime companies, institutes and universities.
• Research publication is likely possible with a possible extension of the internship period.
• Possibility to visit partner hubs or research centers (e.g., MARIN in Wageningen) depending on project needs and availability

Other Are you ready to sail into your new adventure at Damen? Don’t hesitate, send us your motivation letter and resume here.

Due to housing issues we cannot accept international students that do not have accommodation in the Netherlands yet. Recruiter Liselotte van Veenendaal

Email [email protected]

Please apply through the Apply Button. Due to GDPR reasons we cannot accept applications by email.