Porous materials for thermal management under extreme conditions

A brief analysis is presented of how heat transfer takes place in porous materials of various types. The emphasis is on materials able to withstand extremes of temperature, gas pressure, irradiation, etc., i.e. metals and ceramics, rather than polymers. A primary aim is commonly to maximize either the thermal resistance (i.e. provide insulation) or the rate of thermal equilibration between the material and a fluid passing through it (i.e. to facilitate heat exchange). The main structural characteristics concern porosity (void content), anisotropy, pore connectivity and scale. The effect of scale is complex, since the permeability decreases as the structure is refined, but the interfacial area for fluid-solid heat exchange is, thereby, raised. The durability of the pore structure may also be an issue, with a possible disadvantage of finer scale structures being poor microstructural stability under service conditions. Finally, good mechanical properties may be required, since the development of thermal gradients, high fluid fluxes, etc. can generate substantial levels of stress. There are, thus, some complex interplays between service conditions, pore architecture/scale, fluid permeation characteristics, convective heat flow, thermal conduction and radiative heat transfer. Such interplays are illustrated with reference to three examples: (i) a thermal barrier coating in a gas turbine engine; (ii) a Space Shuttle tile; and (iii) a Stirling engine heat exchanger. Highly porous, permeable materials are often made by bonding fibres together into a network structure and much of the analysis presented here is oriented towards such materials. © 2005 The Royal Society.

Uniform stability of a particle approximation of the optimal filter derivative

Sequential Monte Carlo methods, also known as particle methods, are a widely used set of computational tools for inference in non-linear non-Gaussian state-space models. In many applications it may be necessary to compute the sensitivity, or derivative, of the optimal filter with respect to the static parameters of the state-space model; for instance, in order to obtain maximum likelihood model parameters of interest, or to compute the optimal controller in an optimal control problem. In Poyiadjis et al. [2011] an original particle algorithm to compute the filter derivative was proposed and it was shown using numerical examples that the particle estimate was numerically stable in the sense that it did not deteriorate over time. In this paper we substantiate this claim with a detailed theoretical study. Lp bounds and a central limit theorem for this particle approximation of the filter derivative are presented. It is further shown that under mixing conditions these Lp bounds and the asymptotic variance characterized by the central limit theorem are uniformly bounded with respect to the time index. We demon- strate the performance predicted by theory with several numerical examples. We also use the particle approximation of the filter derivative to perform online maximum likelihood parameter estimation for a stochastic volatility model.