Voltar
Artigo Acesso aberto Produção Nacional Revisado por pares
BIBTEX RIS

Sequential particle filter estimation of a time-dependent heat transfer coefficient in a multidimensional nonlinear inverse heat conduction problem

2020; Elsevier BV; Volume: 89; Linguagem: Inglês

10.1016/j.apm.2020.07.020

ISSN

1872-8480

Autores

W.B. da Silva, Júlio César Sampaio Dutra, C. E. P. Kopperschimidt, D. Lesnic, Robert G. Aykroyd,

Tópico(s)

Structural Health Monitoring Techniques

Resumo

In the applied mathematical modelling of heat transfer systems, the heat transfer coefficient (HTC) is one of the most important parameters. This paper proposes a combination of the Method of Fundamental Solutions (MFS) with particle filter Sequential Importance Resampling (PF-SIR) to estimate the time-dependent HTC in two-dimensional transient inverse heat conduction problems from non-standard boundary integral measurements. These measurements ensure the unique solvability of the boundary coefficient identification problem. Numerical results show high performance on several test cases with both linear and nonlinear Robin boundary conditions, supporting the synergy between the MFS and simulation-based particle filter sequential analysis methods.

Referência(s)