A probabilistic model for fast-to-evaluate 2D crack path prediction in heterogeneous materials

Kathleen Pele (1, 2) , Jean Baccou (3, 2) , Loïc Daridon (2, 4) , Jacques Liandrat (1, 5) , Thibaut Le Gouic (1, 5) , Yann Monerie (2, 4) , Frédéric Péralès (3, 2)

1 ECM - École Centrale de Marseille
2 MIST - Laboratoire de micromécanique et intégrité des structures
3 IRSN - Institut de Radioprotection et de Sûreté Nucléaire
4 MéTICE - Mécanique Théorique, Interface, Changements d’Echelles
5 I2M - Institut de Mathématiques de Marseille

This paper is devoted to the construction of a new fast-to-evaluate model for the prediction of 2D crack paths in concrete-like microstructures. The model generates piecewise linear cracks paths with segmentation points selected using a Markov chain model. The Markov chain kernel involves local indicators of mechanical interest and its parameters are learnt from numerical full-field 2D simulations of craking using a cohesive-volumetric finite element solver called XPER. The resulting model exhibits a drastic improvement of CPU time in comparison to simulations from XPER.

Mots clés: Markov chain; concrete; cracking prediction; machine learning

Domaines: Probabilités [math.PR]; Apprentissage [cs.LG]; Génie mécanique [physics.class-ph]

Kathleen Pele, Jean Baccou, Loïc Daridon, Jacques Liandrat, Thibaut Le Gouic, et al.. A probabilistic model for fast-to-evaluate 2D crack path prediction in heterogeneous materials. 2021. ⟨hal-03295926v2⟩