Multi-objective optimization and validation of a simulating moving bed: abstract

Simulating moving bed (SMB) processes are widely used in sugar, petrochemical, and pharmaceutical industries. However, determining the design and operating parameters of SMB processes remains a challenge. The optimization problem is complicated by the relative large number of decision variables (such as flow rates, particle size, column length and configuration) and various conflicting objectives (such as maximization of the adsorbent productivity and the product purity and recovery). A multi-objective optimization is adopted opposed to a single-objective optimization. Using a single-objective optimization, several objectives are combined into a single scalar objective function using arbitrary weight factors. Drawbacks are the sensitivity to the values of the weighting factors used (which are difficult to assign on an a-priori basis) and the risk of losing some optimal solutions. The solution of a multi-objective optimization problem is given by an optimum Pareto set, which is a set of operating conditions such that no operating condition can be found that would lead to objective values which are all better that those in the Pareto set. The feasibility of the approach is demonstrated for sugar and acid removal from fruit juices. Fruit juices have a healthy reputation. However this image is at risk because of their high sugar and acid content, resulting in high calorie intake and tooth erosion respectively. Selective removal of sugar and acid will result in product that meets the requirements imposed by consumer organizations. After an introduction to the case, the results for the selection of adsorbents for sugar and acid removal are presented. Two SMB's were designed based upon the selected adsorbentia. The SMB for sugar removal was designed and optimized using a multi-objective optimization. The SMB for acid removal was designed as a cyclic process with successive adsorption, rinsing, regeneration and rinsing. The overall process design is presented and the results of the optimization effort are evaluated.