Reaction Modeling and Optimization Using Neural Networks and Genetic Algorithms:  Case Study Involving TS-1-Catalyzed Hydroxylation of Benzene

Abstract

This paper proposes a hybrid process modeling and optimization formalism integrating artificial neural networks (ANNs) and genetic algorithms (GAs). The resultant ANN-GA strategy has the advantage that it allows process modeling and optimization exclusively on the basis of process input-output data. In the hybrid strategy, first an ANN-based process model is developed from the input-output process data. Next, the input space of the model representing process input variables is optimized using GAs, with a view to simultaneously maximize multiple process output variables. The GAs are stochastic optimization methods possessing certain unique advantages over the commonly used gradient-based deterministic algorithms. The efficacy of the hybrid formalism has been evaluated for modeling and optimizing the zeolite (TS-1)-catalyzed benzene hydroxylation to phenol reaction whereby several sets of optimized operating conditions have been obtained. A few optimized solutions have also been subjected to the experimental verification, and the results obtained thereby matched the GA-maximized values of the three reaction output variables with a good accuracy.

Genetic Programming Assisted Stochastic Optimization Strategies for Optimization of Glucose to Gluconic Acid Fermentation

“…The training and test set RMSEs corresponding to the optimal model were 0.0377 and 0.0489, respectively. Once an optimal MLP-based model was obtained, in the next phase its input space comprising the three operating variables was optimized using the GA and SPSA methods (Nandi et al, 2001(Nandi et al, , 2002. The GA/SPSA procedure for this optimization remained same as used in optimizing the input space of the GP-based model.…”

Different Tools on Multi-Objective Optimization of a Hybrid Artificial Neural Network – Genetic Algorithm for Plasma Chemical Reactor Modelling

“…The GA is one of the strategic randomized search techniques, which are well known for its robustness in finding the optimal or near-optimal solution since it does not depend on gradient information in its walk of life to find the best solution. Various kinds of algorithm were reported by previous researchers (Tarca et al, 2002;Nandi et al, 2002Nandi et al, , 2004Kundu et al, 2009;Bhatti et al, 2011).…”

“…The detail hybrid algorithm for simultaneous modelling and multiobjective optimization has been developed in previous publication which focused on plasma reactor application (Istadi & Amin, 2005, 2006, 2007. They reported that the hybrid ANN-GA technique is a powerful method for process modelling and multi-objectives optimization (Nandi et al, 2002(Nandi et al, , 2004Ahmad et al, 2004;Stephanopoulos & Han, 1996;Huang et al, 2003;Radhakrishnan & Suppiah, 2004;Fissore et al, 2004;Nandi et al, 2002Nandi et al, , 2004Ahmad et al, 2004;Kundu et al, 20009;Marzbanrad & Ebrahimi, 2011;Bhatti et al, 2011). The method is better than other technique such as response surface methodology (RSM) (Istadi & Amin, 2006, 2007, particularly for complex process model.…”

Plastic Waste Conversion to Liquid Fuels Over Modified-Residual Catalytic Cracking Catalysts: Modeling and Optimization Using Hybrid Artificial Neural Network – Genetic Algorithm

“…The combination of ANN and Genetic Algorithm (GA) has been used by previous researchers for modeling and optimization of integrated process (Nandi et al, 2002;Nandi et al, 2004;Ahmad et al, 2004). The detail hybrid algorithm for simultaneous modeling and optimization using ANN and GA has been developed (Istadi and Amin, 2007;Istadi and Amin, 2006) for complex plasma reactor application.…”