Postdoc - Optimisation under uncertainty for nonlinear structural dynamics

INRIA; France; On-site; EUR 32,000 - 42,000; Full time

Job summary

INRIA Saclay Centre de Recherche recherche un post-doctorant pour développer des méthodes d’optimisation robustes appliquées aux diagrammes de bifurcation et à l’incertitude dans les systèmes dynamiques non linéaires. Le candidat travaillera dans l’équipe Platon auprès d’experts en quantification de l’incertitude et optimisation.

Le poste est temporaire et s’inscrit dans des projets innovants en modélisation numérique, avec supervision par des chercheurs de renom et possibilités de collaboration

Qualifications

• Docteur·e en génie mécanique et/ou mathématiques appliquées.
• Maîtrise des outils de calcul scientifique et des méthodes numériques.
• Connaissance des méthodes d’optimisation et d’incertitude.

Responsibilities

• Développer des méthodes d’optimisation robustes pour des diagrammes de bifurcation.
• Explorer les propagations d’incertitude dans des systèmes dynamiques non linéaires.
• Mettre en œuvre et tester des stratégies méta-modèles pour réduire le coût numérique.

Skills

Uncertainty quantification

Robust optimisation

Non-linear dynamics

Scientific computing

Education

PhD en génie mécanique

PhD en mathématiques appliquées

Job description

Organisation/Company INRIA Department Centre de Recherche INRIA Saclay - Île-de-France Research Field Engineering » Mechanical engineering Mathematics » Applied mathematics Researcher Profile Recognised Researcher (R2) Positions Postdoc Positions Application Deadline 31 Dec 2026 - 12:00 (Europe/Paris) Country France Type of Contract Temporary Job Status Full-time Offer Starting Date 1 Jan 2027 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

Offer Description

Project description and objectives

The optimisation and the design of the dynamic behaviour of mechanical structures play a key role in many industries to meet stringent environmental and performance requirements. The consideration of the non-linearities in such structures is essential. These non-linearities are at the origin of numerous and complex behaviours such as the softening or stiffening of the resonance peak, the existence of multiple dynamic solutions and the appearance of bifurcations in the dynamic behaviour. Bifurcations represent a stability limit in the parameter space characterised by a qualitative and quantitative change in the dynamics of the system (e.g. number and type of responses).

For example, in the development of HALE (High Altitude Long Endurance) or HAPS (High Altitude Pseudo-Satelite) drones, such as the HELIOS drone developed by NASA, the control of aeroelastic instability phenomena is a major challenge. Among these, aeroelastic flutter—resulting from the coupling between structural dynamics and aerodynamic forces—can lead to severe structural failures if not predicted with sufficient accuracy. Classic methods for predicting the critical flutter speed typically rely on deterministic models [5], whereas in practice, numerous sources of uncertainty exist, particularly related to aerodynamic properties, structural mechanical characteristics, or operational conditions. Explicitly accounting for these uncertainties is therefore crucial to identify reliable and robust designs [6].

Recent works from the team have focused on the deterministic optimisation of mechanical structures to reach desired bifurcation behaviours [1,2]. However, numerous uncertainties are present, either from the aerodynamic properties or from mechanical properties. The impact of those uncertainties is critical as the system stability can be impacted [3,4]. Their consideration from the structural optimisation is crucial to ensure the robustness and reliability of the mechanical design.

The objective of the postdoc is to develop robust optimisation methods for bifurcation diagrams. The aim is to combine technics for the analysis of bifurcation of optimization and of uncertainty quantification. Large parametric variations will be considered in the optimisation, leading to large structural variations and so a large range of dynamic behaviours. The bifurcation analysis as well as uncertainty propagation steps are numerically expensive and surrogate-based strategies will be investigated in order to reduce the numerical cost. Three main objectives have been identified for the postdoc:

• The development of the uncertainty propagation methods for the caracterisation of bifurcation behaviour of stochastic nonlinear dynamic systems,
• The development of robust optimisation methods for bifurcation diagrams,
• The development of methods able to deal with real-world scenarios, and more particularly on the test case of a HALE drone for flutter mitigation.

The person recruited will have to numerically implement, test and compare the different identified approaches developed during the postdoc.

Supervision

The postdoc will be supervised by E. Denimal Goy and P.M. Congedo , experts in uncertainty quantification methods for engineering applications. He/She will be also supervised by B. Chouvion from CREA/Ecole de l’Air et de l'Espace, where he has developed a physical solver for flutter calculation and characterization for mechanical structures with geometric nonlinearities and aerodynamic coupling.

The work will be conducted in the Platon team , a joint research group between Ecole Polytechnique and CNRS, hosted by the Center for Applied Mathematics (CMAP) of École Polytechnique. The Platon project-team focuses on developing innovative methods and algorithms for uncertainty management in numerical models, including advanced calibration strategies from data (observations, measurements, other model predictions) and uncertainty reduction.

[1] A. Mélot, E. Denimal, L. Renson, Multi-parametric optimization of bifurcation structures, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2024, 480:2023050520230505

[2] A. Mélot, E. Denimal Goy, L. Renson, Control of isolated response curves through optimization of codimension-1 singularities, Computers & Structures, 2024, 299: 107394.

[3] E. Denimal, J-J. Sinou, Efficient parametric study of a stochastic airfoil system based on hybrid surrogate modelling with advanced automatic kriging construction, European Journal of Mechanics-A/Solids, 2023, 99: 104926

[4] E. Denimal, J-J. Sinou, S. Nacivet, Influence of structural modifications of automotive brake systems for squeal events with kriging meta-modelling method, Journal of Sound and Vibration, 2019, 463: 114938

[5] R. Alcorta, B. Chouvion, G. Michon, O. Montagnier, On the use of frictional dampers for flutter mitigation of a highly flexible wing, International Journal of Non-Linear Mechanics, 2023, 156 :104515.

[6] N. Razaaly, N., B. Chouvion, Quantile-Based Reliability-Constrained Optimization of a Nonlinear Absorber for Passive Aeroelastic Control under Aleatory Uncertainty, Structural and Multidisciplinary Optimization, 2026.

Where to apply

E-mail [e-mail]

Requirements

Research Field Engineering » Mechanical engineering Education Level PhD or equivalent

Research Field Mathematics » Applied mathematics Education Level PhD or equivalent

Skills/Qualifications

Candidates must hold a PhD in mechanical engineering, applied mathematics or a related discipline with background in at least one of these fields: non-linear dynamics, uncertainty quantification, robust optimisation or related fields. In particular, candidates must be proficient scientific computing

Applicants should submit a detailed academic CV with history of scientific production, evaluation documents of their PhD if available and a cover letter detailing the knowledge, skills and experience you think make you the right candidate for the job. For further details, please contact E. Denimal Goy (enora.denimal-goy [at] inria.fr) and B. Chouvion (benjamin.chouvion [at] ecole-air.fr).

Work Location(s)

Number of offers available 1 Company/Institute Inria Saclay Country France City Palaiseau Postal Code 91120 Street Bâtiment Alan Turing - 1 rue Honoré d'Estienne d'Orves - Campus de l'École Polytechnique Geofield