Optimizing TEG Dehydration Process under Metamodel Uncertainty

Natural gas processing requires the removal of acidic gases and dehydration using absorption, mainly conducted in tri-ethylene glycol (TEG). The process is accompanied by emission volatile organic compounds, including BTEX. In our previous work, multi-objective optimization was undertaken to determine optimal operating conditions terms parameters that can mitigate BTEX data-driven metamodeling metaheuristic optimization. Data obtained from a simulation ProMax® simulator were used develop metamodel with machine learning techniques reduce computational time iterations robust simulation. metamodels created limited samples some underlying phenomena must therefore be excluded. This introduces so-called uncertainty. Thus, performance resulting optimized variables may compromised lack adequately accounting for uncertainty introduced metamodel. present bias addressed parameter An algorithmic framework developed optimization, given these uncertainties. this framework, uncertainties are quantified real model data generate distribution functions. We then use novel Better Optimization Nonlinear Uncertain Systems (BONUS) algorithm solve problem. mitigation as objective Our allows determination condition TEG under BONUS determines compared those method, up 405.25 ton/yr.