Design of ethylene oxide production process based on adaptive design of experiments and Bayesian optimization

Abstract

In process design, the values of design variables X for equipment and operating conditions should be optimized for entire processes, including all unit operations, such as reactors and distillation columns, to consider effects between unit operations. However, as the number of X increases, many more simulations are required to search for the optimal X values. Furthermore, multiple objective variables Y, such as yields, make the optimization problem difficult. We propose a process design method based on adaptive design of experiments and Bayesian optimization. Optimization of X values that satisfy target values of multiple Y variables are searched, and simulations for the optimized X values are then repeated. Therefore, X will be optimized by a small number of simulations. We verify the effectiveness of this method by simulating an ethylene oxide production plant.

A hybrid science‐guided machine learning approach for modeling chemical processes: A review

“…A survey of the literature has shown applications of hybrid modeling in bioprocesses, [21][22][23][24][25][26][27] chemical and oil and gas process industries, [28][29][30][31][32] and polymer processes 33,34 for more accurate and scientifically consistent predictions. This survey has also shown many topical focuses of applications in bioprocessing and chemical engineering, including process control, [35][36][37][38] design of experiments, 39,40 process development and scale-up, 41,42 process design, 43 and optimization. 13,44,45 In a recent study, Zhou et al 46 present a hybrid approach for integrating material and process design that holds much promise in process and product design.…”

“…Because of focus in batch process control is on the end‐of‐batch product quality targets, the use of time‐dependent or recurrent neural networks can usually offer good long‐range predictions. Therefore, the resulting hybrid residual model performs well in many batch process control and optimization applications 41,43,69–71 …”