Laetitia Perez

Skin burns after laser exposure: histological analysis and predictive simulation.

Thermal effects of laser irradiation on skin are investigated in this paper. The main purpose is to determine the damage level induced by a laser exposure. Potential burns induced by two lasers (wavelength 808nm and 1940nm) are studied and animal experimentations are performed. Several exposure durations and laser powers are tested. Based on previous works, a mathematical model dedicated to temperature prediction is proposed and finite-element method is implemented. This numerical predictive tool based on the bioheat equation takes into account heat losses due to the convection on skin surface, blood circulatory and also evaporation. Thermal behavior of each skin layer is also described considering distinct thermal and optical properties. Since the mathematical model is able to estimate damage levels, histological analyses were also carried through. It is confirmed that the mathematical model is an efficient predictive tool for estimation of damage caused by lasers and that thermal effects sharply depend on laser wavelength.

On-line identification of temperature-dependent thermal conductivity

In this communication, the investigated study deals with on-line parametric identification in a one-dimensional thermal system. The main objective is the determination of the temperature-dependent thermal conductivity considering noisy temperature measurements. In such a way, the identification problem is written as a minimization one. Since inverse heat conduction problems are ill-posed, a regularization method has to be numerically implemented. Thus, the conjugate gradient method (a well-known iterative regularization method) has been adapted for on-line purposes.

Parametric identification using photothermal data inversion: application to a nonlinear thermal system

Military structures, devices and vehicles used on battlefields are likely to be exposed to heavy thermal aggressions, such as fires or explosions. Protection of personnel and material engaged in combat against such aggressions is a top priority, and needs the use of adapted and optimized systems. Intumescent paints have the ability to swell up when they are heated, building a thick multi-layered coating which provides efficient thermal insulation to the underlying material. In order to evaluate such coatings efficiency in a military framework and to identify its physical properties, experimental tests were carried out. A mathematical model describing intumescent paints behavior under several types of thermal aggressions was developed for system state evaluation in the case of long lasting fires. The model structure has been validated for thermal fluxes induced by several fire configurations and by brief, violent explosions. However, in order to supply the model with reliable input parameters, those must be identified. Thus, a whole identification process was carried out (modeling, sensitivity analysis, experiment, cost function minimization). The identification method is based on the frequency analysis of the heat waves generated by a modulated thermal excitation on the materials surface.

Thermal diffusivity identification based on an iterative regularization method

This article deals with the identification of a space and time dependent material thermal diffusivity. Such parameter is involved in heat transfers described by partial differential equations. An iterative regularization method based on a conjugate gradient algorithm is implemented. Such approach is attractive in order to efficiently deal with measurement noises and model errors. Numerical results are illustrated according to several simulations.

Prototype expérimental pour l'identification de sources chauffantes se déplaçant sur une surface

Resume – Le travail propose dans cette etude s’articule au sein d’un projet visant a deplacer un ensemble de capteurs mobiles de maniere optimale afin d’identifier la trajectoire et la puissance d’une (ou plusieurs) sources evoluant sur une plaque metallique fine. Cette experimentation actuellement en cours de conception et de realisation a pour objet de tester la validite de differentes strategies de deplacements de capteurs. L’objectif principal est d’identifier en ligne le couple puissance & trajectoire de chaque source via une methode de regularisation iterative basee sur la methode du gradient conjugue. Cette communication vise a presenter la conception et la realisation du dispositif experimental. Les sources chauffantes sont embarquees sur des robots mobiles (Khepera III) qui evoluent sur une plaque d’aluminium fine de 4m² et d’epaisseur 2mm. La puissance des sources de chaleur est regulee de maniere a decrire differents profils d’evolution. Sur cette meme plaque, plusieurs robots mobiles sont equipes de pyrometres afin de mesurer la temperature de la plaque de maniere quasi ponctuelle. Les temperatures mesurees ainsi que la position des points de mesures sont transmises a un calculateur central via une technologie sans fil (WIFI). Les problemes inverses pour l’identification sont ensuite resolus quasi en ligne et simultanement, les prochaines trajectoires des pyrometres sont estimees et envoyees aux robots observateurs qui se deplacent alors vers leurs prochaines positions. La position absolue des robots etant necessaire au bon fonctionnement du systeme (robustesse de la procedure d’identification), nous utilisons en complement de la mesure odometrique un systeme de localisation global par camera. Cette derniere est positionnee au-dessus de la plaque ; les differentes mesures de position obtenues par vision sont synchronisees avec celles envoyees par les robots observateurs pour une eventuelle correction. Le dispositif presente dans cette communication est schematise Fig. 1.