Optimisation-based simulation of a pressure swing adsorption process

Abstract

In this paper, an optimisation-based approach is developed for the determination of the cyclic steady-sate (CSS) of a pressure swing adsorption process (PSA). It consists in treating the simulation problem as a single dynamic optimisation problem where the performance index is the CSS condition, the decision variables are the state variables at the start of the cycle and the constraints are given by the process model equations with associated initial conditions. The resulting optimisation problem is solved using a gradient-based non linear programming (NLP) method, e.g. SQP method, where the gradients are computed by means of four different methods: finite differences, numerical and analytical sensitivities and adjoint system methods.

New formulation of optimization‐based simulation of a pressure swing adsorption process: Hybrid dynamic optimization

“…However, the computational time in Liu et al10 is huge compared to the adjoint system method developed here. Moreover, when the gradients are computed by means of sensitivity method and used either in the SQP solver or in a Newton‐based approach, it was shown in Latifi et al5 that the adjoint system method is faster and more accurate.…”

“…These methods are well known in the literature and are extensively studied in order to improve their robustness and convergence rate. However, alternative and interesting approaches have recently emerged and are based on the use of optimization methods for the determination of CSS 1–5. Their basic principle consists in the formulation of the simulation problem as a dynamic optimization problem where the performance index is defined as the sum of square differences between the state variables (gas‐phase concentrations, solid‐phase concentrations and temperature) obtained at the beginning and at the end of a cycle.…”

“…In Latifi et al5 the spatial discretization approach used is based on orthogonal collocation method, and the adjoint system method is used for gradients computation. They showed that the adjoint method has the fastest convergence rate compared to finite differences method and both numerical and analytical sensitivity methods.…”

Adsorption and Separation of the H2O/H2SO4 and H2O/C2H5OH Mixtures: A Simulated and Experimental Study

“…Obtaining and separating the H 2 SO 4 from other elements is important, since it is used in different industrial fields: pharmaceutical industry, fertilizers, sugar cane, etc. [14][15][16][17][18][19].…”

“…The parameters for the model are based on the data from an operating plant as well as data from the literature. Another work related to the PSA process was done by Latifi et al [17], who used an optimization-based approach for the determination of the Cyclic Steady State (CSS) of a PAS process. The optimization problem is solved using a gradient-based nonlinear programming (NLP) method and the gradients are computed by means of four different methods: finite differences, numerical sensitivities, analytical sensitivities, and adjoint system methods.…”