A Manifold Learning Approach for Integrated Computational Materials Engineering

Elena Lopez (1) , David Gonzalez (2) , Jose Aguado (1) , Emmanuelle Abisset-Chavanne (1) , Elias Cueto (2) , Christophe Binetruy (1) , Francisco Chinesta (1)

1 GeM - Institut de Recherche en Génie Civil et Mécanique
2 I3A - Aragón Institute of Engineering Research [Zaragoza]

Image-based simulation is becoming an appealing technique to homogenize properties of real microstructures of heterogeneous materials. However fast computation techniques are needed to take decisions in a limited timescale. Techniques based on standard computational homogenization are seriously compromised by the real-time constraint. The combination of model reduction techniques and high performance computing contribute to alleviate such a constraint but the amount of computation remains excessive in many cases. In this paper we consider an alternative route that makes use of techniques traditionally considered for machine learning purposes in order to extract the manifold in which data and fields can be interpolated accurately and in real-time and with minimum amount of online computation. Locally Linear Embedding is considered in this work for the real-time thermal homogenization of heterogeneous microstructures.

Mots clés: Machine learning; Locally Lineal Embedding; manifold learning; Real time thermal simulation; Computational homogenization; Model Order Reduction; Composite materials

Domaines: Mécanique des structures [physics.class-ph]

Elena Lopez, David Gonzalez, Jose Aguado, Emmanuelle Abisset-Chavanne, Elias Cueto, et al.. A Manifold Learning Approach for Integrated Computational Materials Engineering. Archives of Computational Methods in Engineering, 2018, 25 (1), pp.59-68. ⟨10.1007/s11831-016-9172-5⟩. ⟨hal-02265692⟩