Search

It is proposed a new (hybrid) approach for the mathematical description of the black liquor burning process in an industrial recovery boiler furnace. The system is divided into four different regions and the concentration of each chemical substance in each region is calculated through direct minimization of the corresponding Gibbs free energy. The particulate formation is described separately through a neural network trained with industrial data. The resulting hybrid model reproduces satisfactorily black liquor burning data obtained from industrial sources, and also data reported in the open technical literature. [Copyright &y& Elsevier]

An improved genetic algorithm approach, based on a new ranking strategy, has been proposed to conduct multi-objective optimization of chemical engineering problems. New operators have been introduced to enhance the algorithm performance and reduce the computational effort. A Pareto-set filter operator has been implemented to avoid missing Pareto optimal points during the evolutionary process. A niche operator has been adopted to prevent genetic drift, and an elitism operator, to insure the propagation of the best result of each objective function. A fitness function based on each rank population size and rank level has been used to determine the reproduction ratio. Constraints are handled through a fuzzy penalty function method. The algorithm has been applied to a batch free-radical styrene polymerization process in order to maximize the monomer conversion rate and minimize the concentration of initiator residue in the product. The algorithm proved to be robust, handling satisfactorily multi-modal and multidimensional problems. [Copyright &y& Elsevier]

A model for the steady-state operation of styrene polymerization in a tubular reactor considering radial variation of the relevant variables was solved by the method of lines using either finite volume discretization or global spline orthogonal collocation. The finite volume solution has shown to be slightly more efficient and more robust than the orthogonal collocation solution. The former solution was used to optimize the steady-state operation of a three-section tubular reactor by selecting the wall temperature of each section. The optimization target is to obtain maximal conversion with minimum polymer polydispersity. These conflicting objectives were joined in a parametric global objective function. As it is multimodal, its global minimum for each value of the parameter was obtained by the particle swarm optimization method. The optimal operational conditions allow the production of polymers with lower polydispersity at the same conversion level obtained when uniform temperature exists at the reactor wall. [Copyright &y& Elsevier]

An equilibrium model is proposed for the reactive system ZnSO4/di-2-ethylhexyl phosphoric acid (D2EHPA)/diluent being considered by the European Federation of Chemical Engineering as a test system, valid over a wide concentration range of extraction and stripping conditions. The model is supported by the statistical analysis of estimated parameters. Two simple consecutive reactions are assumed to give two zinc complexes with relative proportions changing as conditions change from extraction to stripping. The model was solved by Newton''s method, and all parameters were estimated by data fitting using the direct search method of Hooke and Jeaves. The model predicts aqueous concentrations with an average relative error lower than 11% by assuming activity coefficients of 1.0. The incorporation of the activity coefficients may improve the model accuracy but cannot validate other simpler extraction mechanisms. [Copyright &y& Elsevier]