Multi-field decomposed hyper-reduced order modeling of damage-plasticity simulations
This paper presents a multi-field decomposed approach for hyper-reduced order modeling to overcome the limitations of traditional model reduction techniques for gradient-extended damage-plasticity simulations. The discrete empirical interpolation method (DEIM) and the energy-conserving sampling and weighting method (ECSW) are extended to account for the multi-field nature of the problem. Both methods yield stable reduced order simulations, while significantly reducing the computational cost compared to full-order simulations. Two numerical examples are presented to demonstrate the performance and limitations of the proposed approaches. The decomposed ECSW method has overall higher accuracy and lower computational cost than the decomposed DEIM method.