Optimization design of crude oil distillation unit using bi-level surrogate model

Crude Oil Distillation Unit (CDU) is one of the most important separation installations in petroleum refinery industries. In this work, a Bi-level Surrogate column model Aided Constrained Optimization Design (Bi-SACOD) proposed for time-consuming objectives and constraints evolutionary optimization design CDUs. The main components Bi-SACOD include bi-level surrogate construction (Bi-SMC), management (Bi-MM), particle swarm (PSO) mixed-integer constrained (PSO-MICE) search. Bi-SMC implements feasible domain identification. Bi-MM combines models with rigorous CDU simulations to perform management, PSO-MICE optimum search works. results CDUs indicate that outperforms single-level approaches, are more consistent approach, whereas evaluation numbers significantly reduced.