Redução de ordem de modelos matemáticos lineares usando algoritmos genéticos

Nesta dissertação apresenta-se o problema de redução de ordem de modelos dinâmicos lineares, sob o ponto de vista de otimização via Algoritmos Genéticos. Uma função custo, obtida a partir da norma dos coeficientes do numerador da função de transferência do erro entre o modelo original e o reduzido, e minimizada por meio de um algoritmo genético, com consequente calculo dos parâmetros do modelo reduzido. O procedimento e aplicado em alguns exemplos que demonstram a validade da abordagem.

Aplicação de modelos matemáticos bi e tri-paramétricos na predição de isotermas de adsorção de umidade do guaraná (Paullinia cupana) em pó

Doze modelos matemáticos (seis bi-paramétricos e seis tri-paramétricos), amplamente utilizados na predição de isotermas, foram estudados, com o objetivo de predizer isotermas de adsorção de umidade do guaraná (Paullinia cupana) em pó, obtido em "sprey dryer". As isotermas de adsorção de umidade do produto à 15° C, 25° C e 35° C, foram ajustadas por análise de regressão não-linear, usando o aplicativo STATGRAPHICS 5.1. Entre os modelos testados os que apresentaram melhores resultados nos ajustes foram as equações tri-paramétricas de GAB, BET modificada e Anderson, e as equações bi-paramétricas de Handerson e Oswin.

Análise e simulação de modelos matemáticos para a estimativa da geração de gases em bioprocessos de decomposição anaeróbia de resíduos sólidos em aterros sanitários

Brazil faces a complex problem in respect to municipal solid waste, having been in recent years an increase of its generation without the country there be adequate for proper disposal thereof. In many states , the percentage of waste destined improperly , ie , in dumps , landfills, send- outs , among others , is greater than that disposed in landfills , which would be the most correct way to be made. It can be argued that this discrepancy is due to the high cost of implementation and operation of the landfill, and the same need large areas with physical characteristics that suit their operations . When there is a provision in properly constructed landfills , municipal solid waste grounded generate gases with high potential energy through biochemical reactions during the anaerobic decomposition of organic material stored . Such gases can be used for power generation within the landfill or other economic means . To estimate the gas generation will be sufficient for such economic compensation , there are mathematical models that make estimating the amount of gas produced . These models calculate the energy capacity and generation , using parameters obtained based on the characteristics of solid waste , climate of the region where they are grounded and grounding time . Such models have been raised and studied so that it was possible to perform simulations that demonstrate the behavior of biogas generation related to the external conditions of the landfill that interfere with biological reactions within. The results show differences between the values obtained , it shows that the preparation of the models found and used in the simulations were allocated amounts for different parameters that determine this difference in the estimate . Therefore, to rule, the models have difficulty understanding this because there is no clarity in the formulation of the equations , and the definition of variables and parameters would require a detailed study to...

A utilização de modelos matemáticos para a otimização de processos

This dissertation has as main theme the discuss about how the use of mathematical models for process optimization. The current scenario of strong competition to conquer the consumer market necessitates the development of improvements to better performance of the process as a whole, is to reduce costs, increase efficiency or effectiveness. Thus, the use of methodologies to assist in this process is becoming increasingly viable. Methodologies developed in the past are being studied and improved. An example is the Desirability, the object of the present study, which was developed in the 80's and has been improved over time. To understand and study this methodology was applied to the desirability function in three instances, where it was used Design of Experiments (DOE), taken from scientific papers, using the Solver tool (Excel ®) and desirability (Minitab ®). Thus, in addition to studying the methodology, it was possible to compare the performance of tools used for optimization in different situations. From the results of this study, it was possible to validate the superiority of one of the models studied compared fairly

Modelos Matemáticos e Computacionais no Manejo Integrado de Pragas

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Modelos Matemáticos e Métodos de Solução para Problemas de Dimensionamento de Lotes

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Modelos matemáticos e computacionais para descrever a transmissão de dois sorotipos de vírus de dengue

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Modelagem Matemática e Análise de Modelos Matemáticos na Educação Matemática

In this paper we present two studies, the first one completed and the second one in development, which are based in teaching approaches that propose the qualitative study of mathematical models as a strategy for the teaching and learning of mathematical concepts. These teaching approaches focus on subjects from Higher Education such as Introduction to Ordinary Differential Equations and Topics of Differential and Integral Calculus. We denominate this common aspect of the teaching approaches as Model Analysis and in a preliminary level we relate it with Mathematical Modeling. Furthermore, we discuss some questions related with the choice of the theme and the role of Digital Technologies when Model Analysis is applied.

Modelos matemáticos para o problema integrado de dimensionamento de lotes e corte de estoque unidimensional

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)