Neural-network acceleration of projection-based model-order-reduction for finite plasticity: Application to RVEs
This work addresses efficiency improvements for computational mechanics researchers and engineers dealing with elastoplastic simulations, though it appears incremental as it builds on existing projection-based methods with neural network enhancements.
The paper tackles the computational cost of projection-based model-order-reduction for finite plasticity in RVEs by developing a recurrent neural network acceleration, resulting in equation-free online simulations that eliminate iterative equation solving and stiffness matrix construction, with stress updates computed only once per increment.
Compared to conventional projection-based model-order-reduction, its neural-network acceleration has the advantage that the online simulations are equation-free, meaning that no system of equations needs to be solved iteratively. Consequently, no stiffness matrix needs to be constructed and the stress update needs to be computed only once per increment. In this contribution, a recurrent neural network is developed to accelerate a projection-based model-order-reduction of the elastoplastic mechanical behaviour of an RVE. In contrast to a neural network that merely emulates the relation between the macroscopic deformation (path) and the macroscopic stress, the neural network acceleration of projection-based model-order-reduction preserves all microstructural information, at the price of computing this information once per increment.