Methodology to Solve the Multi-Objective Optimization of Acrylic Acid Production Using Neural Networks as Meta-Models

Modelling and Multi-Objective Optimization of the Sulphur Dioxide Oxidation Process

Sulphuric acid (H2SO4) is one of the most produced chemicals in the world. The critical step of the sulphuric acid production is the oxidation of sulphur dioxide (SO2) to sulphur trioxide (SO3) which takes place in a multi catalytic bed reactor. In this study, a representative kinetic rate equation was rigorously selected to develop a mathematical model to perform the multi-objective optimization (MOO) of the reactor. The objectives of the MOO were the SO2 conversion, SO3 productivity, and catalyst weight, whereas the decisions variables were the inlet temperature and the length of each catalytic bed. MOO studies were performed for various design scenarios involving a variable number of catalytic beds and different reactor configurations. The MOO process was mainly comprised of two steps: (1) the determination of Pareto domain via the determination a large number of non-dominated solutions, and (2) the ranking of the Pareto-optimal solutions based on preferences of a decision maker. Results show that a reactor comprised of four catalytic beds with an intermediate absorption column provides higher SO2 conversion, marginally superior to four catalytic beds without an intermediate SO3 absorption column. Both scenarios are close to the ideal optimum, where the reactor temperature would be adjusted to always be at the maximum reaction rate. Results clearly highlight the compromise existing between conversion, productivity and catalyst weight.