Process synthesis under uncertainty: A penalty function approach

Abstract

Abstract With the growing environmental concern, it is necessary to improve process simulation and develop design tools to account for environmental factors in the synthesis of large‐scale chemical processes. A major obstacle in tackling this problem is uncertainties in some of the technical and economic parameters, which lead to uncertainties in design, plant performance, and cost estimates. Further, a conceptual process design involves the identification of an optimal flowsheet structure from many alternatives stituting the “superstructure.” Synthesis and optimization of large‐scale processes involving uncertainties often require considerable computational effort. A novel algorithm presented here is based on simulated annealing for the process synthesis of large‐scale flowsheets having several configurations and considers uncertainties in the process design systematically. This new “stochastic annealing algorithm,” provides an efficient approach to stochastic synthesis problems by incorporating a penalty term in the objective function and balances the trade‐off between accuracy and efficiency based on the annealing temperature. It has been used to study a benchmark synthesis problem in the HDA process. Savings of up to 80% in CPU time has been achieved without significant loss of solution precision with stochastic annealing, compared to simulated annealing with a fixed sample size. It can be applied to analyze efficiently any complex process flowsheet and provide valuable insights into process feasibility based on optimal design, plant performance, and uncertainty issues.