INVESTIGADORES
PUIG julieta
congresos y reuniones científicas
Título:
Model-based methodology coupled with a Laserbased experiment for non-destructive thermal parameter estimation of an Epoxy nanocomposite
Autor/es:
OTERO, FERNANDO A.; ALTUNA F.; HOPPE, CE; ARENAS, GUSTAVO F.; ANTONACCI J.; CHIURO, CARLOS; PUIG, J; PONTIS, ANDRÉ; FRONTINI, GLORIA; ELICABE, GUILLERMO
Reunión:
Congreso; 2020 IEEE Congreso Bienal de Argentina (ARGENCON); 2020
Resumen:
This article addresses the estimation of thethermal conductivity k and the specific heat capacity Cp of asynthetic nanocomposite, by coupling a laser-based experimentand a reduced model-based inverse problem solved via a FiniteElement Method (FEM) approach. The associated multiphysicsproblem can be described as an optical-radiative-thermalcoupled process where a laser source is irradiating the samplecorresponding to an Epoxy-Based Vitrimer (EV) with dispersedgold nanoparticles (NPs). A 1D reduced model for an infiniteslab (having both boundaries subjected to natural convectionand under certain conditions on both experiment and materialproperties) has been used to solve the forward radiative transferproblem involved, in order to achieve an initial approximationfor the equivalent thermal source Q. This approximation is thenused in the heat transfer equation, whose solution is computedusing the finite element method (FEM). The correspondingstatistical solution for the thermal conductivity k and the heatcapacity Cp is calculated by the inversion of the T-thermocoupletemperature time series measured at the center of the irradiatedspecimen, using a stochastic version of the LevenbergMarquardt algorithm by embedding the algorithm inside aMonte Carlo routine. Reduced models have been analyzed for1D and 2D reduced geometries as well as approximations for Qin the 3D case. Results show a confidence interval for theachieved estimates of the thermal parameters in a goodagreement to the Differential Scanning Calorimetry (DSC)referential values. As a conclusion, even when the performedlaser remote heating experiment has been developed as a selfhealing process, it also appears to be a very promising NonDestructive Testing (NDT) methodology for retrieving thethermal parameters along with the proposed computationalapproach in some thermosetting polymers, such as the EVstudied here.