Adaptive machine learning-based surrogate modeling to accelerate PDE-constrained optimization in enhanced oil recovery

Abstract In this contribution, we develop an efficient surrogate modeling framework for simulation-based optimization of enhanced oil recovery, where particularly focus on polymer flooding. The computational approach is based adaptive training procedure a neural network that directly approximates input-output map the underlying PDE-constrained problem. process thereby focuses construction accurate model solely related to path outer iterative loop. True evaluations objective function are used finally obtain certified results. Numerical experiments given evaluate accuracy and efficiency heterogeneous five-spot benchmark