3 resultados para Approximate Bayesian computation
em Universidade Federal do Pará
Resumo:
Este artigo é dedicado ao estudo da controlabilidade finito-aproximada para a equação não linear de transferência de calor em domínios com fronteira móvel. A demonstração do resultado principal baseia-se no princípio de continuação única de Carolina Fabre 1996 e em argumentos de ponto fixo do tipo Leray-Schauder.
Resumo:
We prove the approximate controllability of the semilinear heat equation in RN, when the nonlinear term is globally Lipschitz and depends both on the state u and its spatial gradient Ñu. The approximate controllability is viewed as the limit of a sequence of optimal control problems. In order to avoid the difficulties related to the lack of compactness of the Sobolev embeddings, we work with the similarity variables and use weighted Sobolev spaces.
Resumo:
In this paper, we propose a hybrid methodology based on Graph-Coloring and Genetic Algorithm (GA) to solve the Wavelength Assignment (WA) problem in optical networks, impaired by physical layer effects. Our proposal was developed for a static scenario where the physical topology and traffic matrix are known a priori. First, we used fixed shortest-path routing to attend demand requests over the physical topology and the graph-coloring algorithm to minimize the number of necessary wavelengths. Then, we applied the genetic algorithm to solve WA. The GA finds the wavelength activation order on the wavelengths grid with the aim of reducing the Cross-Phase Modulation (XPM) effect; the variance due to the XPM was used as a function of fitness to evaluate the feasibility of the selected WA solution. Its performance is compared with the First-Fit algorithm in two different scenarios, and has shown a reduction in blocking probability up to 37.14% when considered both XPM and residual dispersion effects and up to 71.42% when only considered XPM effect. Moreover, it was possible to reduce by 57.14% the number of wavelengths.