Application of Genetic Algorithm to Estimating Kinetic Model Parameters of Propylene Ammoxidation

Jian GU,Jing WEI,Feng-qiu CHEN,Qing-lian DAI,De-Wei L(U)
DOI: https://doi.org/10.3321/j.issn:1000-8144.1999.04.004
1999-01-01
Abstract:The application of genetic algorithm in estimating kinetic model parameters of propylene ammoxidation was described. In accordance with the peculiarity of the model, the scientific notation coding method and the delayed mutation strategy were used to improve the search efficiency of GA. The study shows that GA is sensitive to the variation in the weight of component concentration. Additionally, the traditional optimization methods (Powell method, simplex method)and GA were compared according to their error of fitting. The parameters estimated by GA are satisfactory and superior to those by traditional methods.
What problem does this paper attempt to address?