(148f) Full Cycle Optimization of Simulated Moving Bed Processes
Simulated Moving Bed (SMB) was developed as a realization of continuous countercurrent operation of chromatographic separation. An SMB unit consists of several columns of the same length connected in series, where feed and desorbent are supplied and extract and raffinate are withdrawn continuously. This operation is repeated with shifting the supply/withdrawal points at a regular interval. It repeats the same operation using the identical columns, making this process symmetric. In this study, we explore asymmetric operating schemes through a full-cycle optimization model, where the operating condition of an entire cycle is considered as a Nonlinear Programming (NLP) problem. In addition to the standard SMB configuration, an SMB superstructure is also considered to find the optimal operating scheme. Further, we explore asymmetric designs, where the lengths of columns are not uniform. The SMB model described by Partial Differential Algebraic Equations (PDAEs) is fully discretized both in temporal and spatial domains that leads to a large scale NLP problem, and the optimization problem is implemented within the AMPL modeling environment. The problem is solved using IPOPT 3.1, an interior-point solver. We present optimization case studies to show that the problem is solved efficiently, and introducing a full-cycle formulation has the potential to increase throughput. Finally multi-objective optimization studies are carried out to investigate the trade-off of throughput maximization and desorbent minimization.