Deep Learning–Driven Design under Dynamic Loads

This hybrid course provides MSc and early-stage PhD students with the knowledge and tools to design lightweight structures capable of withstanding highly dynamic loads. The activity combines physical understanding of dynamic behavior and Fluid–Structure Interaction (FSI) with Deep Learning (DL) techniques for surrogate modeling and structural design. Students will learn how to (i) quantify the influence of FSI on shock-loaded structures and (ii) generate lattice-based metamaterials optimized for energy absorption.

Teaching methods and tools The course adopts a flipped and challenge-based learning approach designed to require no physical mobility. Indeed, the course is delivered following a hybrid model: sessions are held in-person for local students at the host institution and streamed online for students at partner universities.

Particularly, students first complete online micro-lectures and short live sessions to build a common background in Finite Element simulations, deep learning, and dynamic material behavior. Subsequently, the course is divided into two main blocks:

• A five-day session hosted by Politecnico di Milano (in-person for Polimi students, online for others) including focused international lectures, hands-on workshops, and a team challenge involving data-driven structural design and rapid prototyping of lattice metamaterials using 3D printing.
• A two-day follow-up seminar series hosted by NTNU (in-person for NTNU students, online for others) featuring international keynotes on AI and metamaterials, AI and blast loading, and PhD short talks from both institutions, fostering cross-university collaboration.

Innovative aspects The initiative promotes innovation through:

• Integration of physics-based and AI-driven modeling approaches;
• Application of DL to both FSI surrogate modeling and lattice structures design;
• Hands-on making sessions for physical validation;
• International co-teaching and peer learning between students from different universities.

Upon successful completion of the course, students will have a foundational understanding enabling them to:

• Analyze the dynamic response of lightweight structures under highly transient and shock-type loads, including Fluid–Structure Interaction effects;
• Apply finite element methods to model dynamic structural and FSI problems;
• Understand the principles of deep learning for surrogate modeling of complex dynamic systems;
• Develop data-driven models to approximate structural response under dynamic loading;
• Understand AI-assisted approaches for the design of lattice-based metamaterials for energy absorption.

Basic knowledge of structural engineering and introductory programming skills in MATLAB or similar languages are required.

Before the in-person sessions, online learning materials will be made available and short live sessions will be held online to provide fundamental concepts on numerical simulations, machine learning techniques, and dynamic loading conditions, in order to align the background level of all participants.

• Online part (common to all participants): April 7, 9, 14, 16 — from 2:00 PM to 4:00 PM
• Core course at Polimi: April 20–24 — from 9:00 AM to 6:00 PM
• Core course at NTNU: June 23–24 — from 9:00 AM to 4:00 PM

In this structure, students attend on-site only when lessons are held at their home university, joining remotely for all other sessions.

The two-level mutual trust-based quality assurance scheme has been adopted:

• at the university level: Politecnico di Milano has applied its internal quality assurance procedures and structures to the proposal of Deep Learning–Driven Design under Dynamic Loads it submitted to ENHANCE and to its implementation - the related learning activities,
• at the Alliance level: the body composed of Education Officers has made decisions regarding the inclusion of Deep Learning–Driven Design under Dynamic Loads proposed by Politecnico di Milano to the Innovative Learning Campus part of the joint ENHANCE educational offer, based on the compliance with the formal requirements and ENHANCE goals.

The evaluation is based on:

• Participation and engagement (20%);
• Team challenge: technical presentation and final report (50%);
• Follow-up activities: short scientific reflection and contribution to discussion (30%).

Successful completion requires active participation (in-person or online) in both the Milan and Trondheim phases. The course awards 4 ECTS and promotes technical competence, creativity, and collaboration in AI-based structural design. The course will be considered successfully completed even in case of absences, provided they do not exceed 30% of the total class hours.

Applications are open from February 19 to March 12 at 12:00 PM following the procedures listed below. Please, take the time to read them carefully in order to choose the right path.

ENHANCE students (a part from Polimi):

Applications will be submitted via the following form.

Please, be aware that for submitting your application you should have to be logged in with a Google account.

OR

Polimi students (only):

Applications will be submitted via the following link: Passion in Action.

Please, be aware that only the applications submitted through the Passion in Action website will be considered valid.

Applications will be evaluated based on the motivation letter, giving priority to PhD students.

Available seats:

• 8 Polimi ;
• 6 NTNU ;
• 6 places are reserved for students from the ENHANCE Alliance. In case of unfilled places or specific needs, these spots will be made available to Polimi and NTNU students.