Constitutive artificial neural networks: a general anisotropic constitutive modeling framework utilizing machine learning

PAMM, vol. 21

Abstract

Abstract In this contribution, a novel machine learning architecture for data‐driven modeling of the mechanical constitutive behavior of materials, constitutive artificial neural networks (CANNs) [1], will be introduced. CANNs incorporate basic material modeling fundamentals from continuum mechanics while relying on artificial neural networks for material‐specific relations. Their architecture allows them to process stress‐strain curves and arbitrary additional information (e.g., about the microstructure or manufacturing parameters). With only a low‐to‐moderate amount of training data and training time, they can predict the constitutive behavior of complex nonlinear and anisotropic materials. The ability to utilize additional material‐specific information enables CANNs to predict the mechanical behavior of previously unseen materials if the CANN was sufficiently trained with many similar materials.

Authors 6

1. Markus Hillgärtner corresponding Aachen Department of Continuum Mechanics

   RWTH Aachen University

   Affiliation as printed

   Department of Continuum Mechanics RWTH Aachen University Germany

   Department of Continuum Mechanics RWTH Aachen University Germany

2. Kevin Linka

   Hamburg University of Technology

   Affiliation as printed

   Department of Continuum and Materials Mechanics Hamburg University of Technology Germany

   Department of Continuum and Materials Mechanics Hamburg University of Technology Germany

3. Kian P. Abdolazizi Aachen Department of Continuum Mechanics

   RWTH Aachen University · Hamburg University of Technology

   Affiliation as printed

   Department of Continuum Mechanics RWTH Aachen University Germany

   Department of Continuum and Materials Mechanics Hamburg University of Technology Germany

   Department of Continuum Mechanics RWTH Aachen University Germany

   Department of Continuum and Materials Mechanics Hamburg University of Technology Germany

4. Roland Can Aydin

   Helmholtz-Zentrum Hereon

   Affiliation as printed

   Institute of Materials Research Materials Mechanics Helmholtz-Zentrum Geesthacht Germany

   Institute of Materials Research Materials Mechanics Helmholtz-Zentrum Geesthacht Germany

5. Mikhail Itskov Aachen Department of Continuum Mechanics

   RWTH Aachen University

   Affiliation as printed

   Department of Continuum Mechanics RWTH Aachen University Germany

   Department of Continuum Mechanics RWTH Aachen University Germany

6. Christian Johannes Cyron

   Hamburg University of Technology · Helmholtz-Zentrum Hereon

   Affiliation as printed

   Department of Continuum and Materials Mechanics Hamburg University of Technology Germany

   Institute of Materials Research Materials Mechanics Helmholtz-Zentrum Geesthacht Germany

   Department of Continuum and Materials Mechanics Hamburg University of Technology Germany

   Institute of Materials Research Materials Mechanics Helmholtz-Zentrum Geesthacht Germany