Adaptive scenario subset selection for worst-case optimization and its application to well placement optimization

In this study, we consider simulation-based worst-case optimization problems with continuous design variables and a finite scenario set. To reduce the number of simulations required increase restarts for better local optimum solutions, propose new approach referred to as adaptive subset selection (AS3). The proposed subsamples support construct objective function in given neighborhood, introduce such subset. Moreover, develop algorithm by combining AS3 covariance matrix adaptation evolution strategy (CMA-ES), denoted AS3-CMA-ES. At each algorithmic iteration, scenarios is selected, CMA-ES attempts optimize computed only through scenarios. reduces executing on subset, rather than all numerical experiments, verified that AS3-CMA-ES more efficient terms brute-force surrogate-assisted lq-CMA-ES when ratio total relatively small. addition, usefulness was evaluated well placement carbon dioxide capture storage (CCS). comparison lq-CMA-ES, able find solutions because frequent restarts.