Optimization of Random Pore Model Parameters of Metallurgical Coke Nonisothermal Gasification via Genetic Algorithm and Least Squares

Abstract

This paper kinetically studied the metallurgical coke nonisothermal gasification by thermogravimetry. The random pore model (RPM) was made use of to describe gasification kinetic behavior. The RPM parameters were optimized using two methods, viz. the method using genetic algorithm alone and the method combining genetic algorithm and least squares. By comparison with the former method, the latter yielded more accurate parameters. Theoretical curves obtained from the method combining genetic algorithm and least squares matched experimental ones well.