진학프로석·박사 채용 접수 플랫폼

We invite applications for a PhD project focused on advancing the design and functionality of deployable structures through the integration of 4D printing, architected materials, and scientific machine learning (sciML). A part of large-scale initiative between UK (DSTL) and France (DGA-AID) on the 4D printing technology, this project aims to develop next-generation Origami and Kirigami-inspired systems, where geometry, material distribution, and embedded functionalities are co-designed to achieve programmable, adaptive shape transformations in response to controlled stimuli.

The research will explore how filament and fibre architecture can be leveraged as key design variables in 4D-printed systems, enabling controlled deployment and reconfiguration. Particular attention will be given to advanced printing features such as graded or voxel-based material distributions, surface coatings, and embedded elements (e.g., electrodes, heating pathways and infrared fibre routing). These innovations aim to unlock new levels of multifunctionality and performance in smart structures, with applications in aerospace, soft robotics, biomedical devices and defence systems.

A central component of the project is the development of generative design methodologies integrating materials, geometry and actuation mechanisms. Multi-physics modelling approaches will be employed to simulate active material behaviour, structural deformation and coupled physical phenomena. These models will be combined with reduced-order techniques to enable efficient simulation and optimisation of shape transformations.

A key objective is to address inverse design problems under real-world constraints, including printability and variability in physical parameters. The project will leverage recent advances in sciML in computational mechanics and physics, including differentiable finite element methods (FEM) and deep learning approaches. Differentiable FEM enables direct access to sensitivity information within non-linear and coupled systems, while neural networks can be used to learn complex behaviours and accelerate optimisation processes. Hybrid approaches combining deep learning with reduced-order modelling will be explored to develop efficient and robust generative design frameworks.

The PhD candidate will be enrolled in an active research area, with an application to innovative structures and systems. He will contribute to the development of novel sciML-based design tools within a collaborative research environment. The work will be part of a larger project involving close interaction with experimental teams responsible for prototyping and testing. Therefore, robustness, scalability, and practical applicability of the developed methods will be key success factors.

Host institution and Supervision

• • Location: UTBM, Sevenans Campus, France
• • Supervisors: Prof. Frédéric Demoly and Dr. Thibaut Hirschler

Objectives

1. 1. Develop and validate multi-physics models describing actuation and mechanical response of 4D-printed structures.
2. 2. Develop and implement advanced numerical methods based on sciML and deep-learning tools for inverse design problems. 
3. 3. Establish a generative design framework for achieving 4D-printed Origami/Kirigami-inspired systems with optimal shape, material distributions and functional features.

Funding category: Autre financement public

Agence de l'Innovation de Défense (AID)

PHD title: Doctorat en Sciences de l'Ingénieure (Engineering PhD)

PHD Country: France