Modelos de ordem reduzida e simulação em colunas de destilação

AUTOR(ES)
DATA DE PUBLICAÇÃO

2000

RESUMO

One of the largest difficulties found in the analysis of staged separation processes is the great dimension of their models. Thus, the simulation of these processes involves a high number of algebraic-differential equations, causing a high computational effort. Several researchers have been proposing order reduction strategies, that is reduce the dimension of the models. This work has as objectives the development of reduced order models for systems of staged separation processes through collocation strategies, the selection of the best collocation strategy, as well as the selection of the algorithm more adapted for the solution of the equations of the model seeking that combination of model structure and of algorithm that allows to obtain an appropriate simulation and minimum computational cost. The application of the method of orthogonal collocation is extended to the solution of differences equations, starting from the development of a methodology that unifies the use of orthogonal collocation for the solution of differential equations as well as for the one of differences equations. The results obtained in the simulations demonstrate the viability of the application of discrete orthogonal collocation in the simulation of distillation columns in stationary state. The discrete mixed orthogonal collocation led to a better approach than the modified mixed orthogonal collocation. Besides, a considerable reduction in the computational time in comparison to the complete model was obtained

ASSUNTO(S)

modelos matematicos destilação simulação (computadores) polinomios

ACESSO AO ARTIGO

Documentos Relacionados