Steady-state flowsheet simulations can be too complex to be amenable to direct optimization. An indirect route involves data collection from the simulation and fitting of less complex surrogates: metamodels, which are more readily optimized. The use of metamodels allows the optimization to proceed while requiring only a small number of solutions to be obtained from the simulation. This paper describes a general methodology for optimization of process simulations via metamodeling. Implementation issues surrounding the data collection and model generation phases of the methodology are identified. Examples of effective techniques are given.
Return to Kurt Palmer's homepage