Simulation, sensitivity analysis and optimization of a continuous catalytic reforming process

A simple kinetic scheme involving 17 lumps and 17 reactions was incorporated in a process model. This model then was used to predict temperature and reformate composition profiles of a commercial reforming process. The agreement between predicted and commercially observed results is very good. The key process variables affecting the reformer yields, such as liquid hourly space velocity (LHSV), the reformer inlet temperature (WAIT), molar hydrogen-to-oil ratio and reaction pressure, were analyzed in detail to guide the selections of naphtha feedstock and operating conditions. Lagrange-Marquardt method was used to solve the nonlinear programming problem with process constraints to optimize process and maximize plant profit. The optimized process variables can bring about 1 percent increase in aromatics yield in weight. The optimization result is approved to not be sensitive to the uncertainty of model parameters.