Simulation And Optimization Of Nonideal Polymerization Reactors Based On Molecular Weight Distribution Using Computational Fluid Dynamics Method

Molecular weight distribution(MWD)is essential for describing the microstructural quality of polymers.The end-use properties of polymers are largely dependent on their MWDs,which affect processing through the polymer melting point and the flow properties of melted polymers.Thus,calculating and optimizing polymer MWD are important in industrial polymerization processes.Most of the studies on MWD are based on an ideal reactor assumption,which is actually inappropriate for a large-scale industrial reactor with poor mixing.Computational fluid dynamics(CFD)methods is a useful tool to deal with the nonideal reactors.In recent years,researchers have used CFD methods to obtain more accurate simulation results.However,studies with CFD have not been extended to polymer MWD due to the computational difficulties.In this study,a new strategy is proposed to simulate the spatial MWD for nonideal reactors using the method of moments with interface to the CFD platform.Subsequently,given a target MWD curve,process optimization is proposed to achieve the optimal operating conditions.The detailed research work is summarized as follows:1.This study calculates the MWD for nonideal reactors using the method of moments and CFD for free radical polymerization.Applying the method of moments allows the replacement of the polymerization scheme with a conventional reaction scheme that has a manageable size.The MWD is calculated by introducing Flory distribution theory.An interface is designed for the commercial software Fluent to extend the CFD simulation to MWD calculation.The tubular and autoclave reactors of the low-density polyethylene process are demonstrated for the simulations.2.Direct search algorithm is proposed to maximize monomer conversion and meet the MWD error constraint.The CFD simulation is embedded in the optimization like a black box.Given a target MWD curve,process optimization is attained by combining CFD and MWD calculations using the interface between C language and Fluent.The Nelder-Mead simplex method,a typical and efficient gradient-free optimization algorithm,is used to search for the optimal solution.The optimizations of the tubular and autoclave reactors are demonstrated to show the practicability of the proposed method.3.As the Nelder-Mead simplex method is slow,a new method combined with the trust-region algorithm is proposed to improve the performance and efficiency.The trust-region method has advantage to find a feasible solution quickly.However,its convergence speed slows down after entering the feasible region and it is easy to fall into a local optitmal solution.Then the combined optimization method is proposed by taking the advantages of the two algorithms.A feasible solution is found by trust-region method first.Then the simplex algorithm is used to continuing seeking the optimal solution.With the tubular polymerization process as an example,the method is proved to be effective in improving the optimization result and accelerating the convergence speed.This study proposed a method for the simulation and optimization of MWD in nonideal reactors.An interface between MWD optimization and CFD simulation is designed and proved to be effective.Finally,a summary of the research and the perspectives of the future work are presented and discussed.