A novel multivariable optimization strategy for methanol to propylene fixed bed reactors using a hybrid algorithm

The methanol-to-propylene (MTP) process holds significant promise as a viable route for propylene production. To enhance the efficiency of this process, we conducted a comprehensive study employing a two-step approach that integrates micro- and macro-level perspectives. Firstly, we established a gas–solid two-phase reactor model tailored for an industrial six-stage fixed-bed reactor. Through a rigorous sensitivity analysis, we investigated the impact of catalyst particle size on bed pressure drops and product distribution, ultimately identifying an optimal particle size range between 2.5–3.5 mm. Subsequently, we developed a hybrid optimization algorithm that seamlessly integrates a stochastic algorithm with a deterministic algorithm. A key component of our methodology was the utilization of the orthogonal collocation method on finite elements (OCFE), which effectively converted differential equations (DEs) into algebraic equations (AEs), simplifying the optimization process. The results show that methanol throughput increases by 4.75 %, and propylene selectivity increases from 65.34 % to 73.98 %.