Identificação de falhas em compósitos laminados usando Unscented Kalman Filter
Nenhuma Miniatura disponível
Data
2018-03-09
Autores
Florindo, Leone Bernardo
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal do Espírito Santo
Resumo
Composite materials have numerous applications in the most varied areas of engineering.In this scope, mastery of these materials becomes necessary. The composite materialacquires properties of each materials that constitute it. However, between its layers theremay be adhesion failures, compromising its quality. In this way, it is proposed to estimatethe thermal contact conductance at the interface of a laminated composite, which canbe directly associated with adhesion failures. Therefore, it was proposed to apply theUnscented Kalman Filter with the method of lines to solve an inverse heat transfer problem.It was established a physical problem characterized by being transient regime and two-dimensional, then the thermal contact conductance at the interface between the layers is afunction of time and space in one direction. The temperature measurements used as inputdata for the equations that model the problem were obtained through simulations. In thetests carried out, a variety of materials and composite dimensions were used. A sensitivityanalysis was made regarding noises in the measuared temperatures and a verification ofthe influence of the time in the physical problem relating to the computational effort. Theresults have shown that the methods applied were effective to indentify the laminatedcomposite failure
Descrição
Palavras-chave
Unscented Kalman Filter , Inverse problems , Bayesian filters , Thermal contact conductance , Laminated composite , Filtros bayesianos , Condutância térmica de contato , Compósitos laminados
Citação
FLORINDO, Leone Bernardo. Identificação de falhas em compósitos laminados usando Unscented Kalman Filter. 2018. 69 f Dissertação (Mestrado em Engenharia Mecânica) - Universidade Federal do Espírito Santo, Centro Tecnológico, Vitória, 2018.