2022 Annual Meeting
(173ab) On the Performance of Differential Evolution Optimization in Kinetic Parameter Determination of Propene Polymerization through Modeling & Simulation
Author
Most popular classes of EA in current usage are genetic algorithms (GAs) and differential evolution (DE). While GA is more suitable for discrete optimization and the problem is encoded in a series of bit strings that are manipulated by the algorithm, on the other hand DE are more natural and suitable for continuous optimization and uses real-coded variables. Decision variables and problem functions are used directly. These capabilities allow differential evolution to converge faster to solutions.
Metallocene catalysts, allow exceptional control of polymer molecular design, which yields products having improved properties. Metallocene catalyst apportions the production of tailor-made polymers with properties that can be specifically designed. Metallocene catalysts have acknowledged enormous consideration for the stereospecific polymerization of an array of monomers.
In this work, a kinetic model is developed to analyze the kinetic behavior of propene polymerization using metallocene catalysts. A detailed simulation methodology using Differential Evolution approach of optimization is discussed to estimate the kinetic parameters viz. propene concentration, catalyst concentration, reaction temperature, and cocatalyst to catalyst ratio.
This work aims to scrutinize the performance of DE variants, namely the standard DE, the composite DE (CODE), the adaptive DE for handling constrained optimization problems associated with estimation of kinetic parameters for propylene polymerization. A novel strategy of DE ânatural logarithmic DE (NLDE)â is proposed and compared with the existing strategies of DE.