Simulation and optimization of supported Ziegler–Natta catalyst preparation based on AI approach coupled with genetic algorithm

Abstract

In the present work we are aimed to use powerful and state of the art methodology to optimize catalyst synthesis procedure. In this way, the preparation process of supported Ziegler–Natta catalysts was simulated and optimized through artificial intelligence (AI) methodology coupled with genetic algorithm (GA). The yield of preparation process was investigated through assessing the catalyst activity. The effects of several variables including TiCl4 injection temperature, TiCl4/toluene ratio, and TiCl4 injection time on the activity of prepared catalyst were investigated. In model development, leave-one-out technique was used for training the network. The developed neural network model can be utilized to enhance the efficiency of the catalyst preparation process.