Search

Numerical simulations of ice crystal icing within an experimental single-stage compressor test rig representative of realistic aeronautical engines are investigated. This paper aims to assess the icing models in CEDRE multi-physics CFD code recently developed and implemented. Gas flow, ice crystal trajectories, fragmentation dynamics and melting along the geometry are successively computed before the ice accretion in the test article is calculated using a surfacic solver that includes recently developed erosion and porosity models. Three experimental operating points are considered, with wet bulb temperatures increasing from negative to positive values and respectively exhibiting no accretion, significant accretion and moderate accretion. The comparison of simulation results with experiments for these cases shows a good agreement in terms of occurrence, location and surfacic extension of accretion phenomena. Nevertheless, a global shift in the inlet temperature is necessary to reproduce experimental results.

International audience; The current work deals with the numerical simulation of ice crystal accretion within an experimental single stage compressor test rig. Gas flow, ice crystal trajectories and impingement are successively computed before accretion is calculated using a surfacic solver. The modeling includes recently developed erosion and porosity models in the framework of the european MUSIC-haic project. These models are firstly applied to an application case representative of realistic aeronautical engines that provides quantified data for the set-up. Three experimental operating points are considered, with wet bulb temperatures increasing from negative to positive values and respectively exhibiting no, significant and moderate accretion. The comparison of simulation results with experiments for these three cases shows a good agreement in terms of occurrence, location and surfacic extension of accretion phenomena.

International audience; The current work deals with the numerical simulation of ice crystal trajectories within an experimental single stage compressor test rig. First, ice crystal trajectories and the associated fragmentation dynamics within the compressor stage are investigated for different rotor speeds. To this purpose, numerical simulations with two different fragmentation models are compared for cold conditions, i.e. with a sufficiently negative wet bulb temperature to totally inhibit melting. The comparison between the fragmentation models aims at assessing the capabilities of a recently proposed ice crystal fragmentation model with respect to a state of the art model. Both models show good agreement with experimental size distribution data as they reproduce the size reduction of the ice crystals with increasing rotational speed. In a second step, three experimental operating points with wet bulb temperatures raising from negative to positive values and respectively exhibiting no, significant and moderate melting, are simulated. Here, the comparison of numerical and experimental results highlights the strong sensitivity of the ice particles' melt ratio to inlet temperature variations, even by a few degrees.

This paper is devoted to investigate the temporal and spatial accuracy in transient conjugate heat transfer problems where the influence of unsteadi-ness in the fluid is supposed to be negligible. The numerical method is based on an iterative procedure with interpolated boundary conditions on the solid side. Emphasis is put on these interpolation procedures. First it is shown that a temporal interpolation is by no means the most appropriate way of estimating the flow boundary conditions prescribed on the coupled solid surfaces. In contrast, when these conditions are based on the heat flux/temperature linearity, significantly better results are obtained. Indeed, they compare very well against a reference solution provided by a non-iterative and non-interpolated coupled computation. Moreover, important improvements were made by adequately defining the integral contribution of the boundary condition using the Gauss points in the finite element solid code. Finally, it was also revealed that a relevant approximation of the heat transfer requires a sufficiently fine solid mesh at the interface.; Cet article est consacré à l'étude de la précision temporelle et spatiale dans les problèmes aérothermiques en régime transitoire où l'influence des effets instationnaires dans le fluide est supposée être négligeable. La méthode numérique est basée sur une procédure itérative avec des conditions aux limites interpolées sur l'interface solide. L'accent est mis sur ces procédures d'interpolation. Tout d'abord, il est montré qu'une interpolation temporelle n'est en aucun cas le moyen le plus approprié d'estimer les conditions limites de l'écoulement prescrites sur les surfaces solides couplées. En revanche, lorsque ces conditions sont basées sur la linéarité flux de chaleur/température, les résultats obtenus sont nettement meilleurs. En effet, la comparaison avec une solution de référence obtenue par un calcul couplé non itératif et non interpolé est satisfaisante. De plus, d'importantes améliorations ont été apportées en définissant adéquatement la contribution intégrale de la condition limite à l'aide des points de Gauss situés sur l'interfaces du solide. Enfin, il a également été révélé qu'une approximation pertinente du transfert de chaleur nécessite un maillage solide suffisamment fin à l'interface fluide-structure.

This paper is devoted to investigate the temporal and spatial accuracy in transient conjugate heat transfer problems where the influence of unsteadi-ness in the fluid is supposed to be negligible. The numerical method is based on an iterative procedure with interpolated boundary conditions on the solid side. Emphasis is put on these interpolation procedures. First it is shown that a temporal interpolation is by no means the most appropriate way of estimating the flow boundary conditions prescribed on the coupled solid surfaces. In contrast, when these conditions are based on the heat flux/temperature linearity, significantly better results are obtained. Indeed, they compare very well against a reference solution provided by a non-iterative and non-interpolated coupled computation. Moreover, important improvements were made by adequately defining the integral contribution of the boundary condition using the Gauss points in the finite element solid code. Finally, it was also revealed that a relevant approximation of the heat transfer requires a sufficiently fine solid mesh at the interface.; Cet article est consacré à l'étude de la précision temporelle et spatiale dans les problèmes aérothermiques en régime transitoire où l'influence des effets instationnaires dans le fluide est supposée être négligeable. La méthode numérique est basée sur une procédure itérative avec des conditions aux limites interpolées sur l'interface solide. L'accent est mis sur ces procédures d'interpolation. Tout d'abord, il est montré qu'une interpolation temporelle n'est en aucun cas le moyen le plus approprié d'estimer les conditions limites de l'écoulement prescrites sur les surfaces solides couplées. En revanche, lorsque ces conditions sont basées sur la linéarité flux de chaleur/température, les résultats obtenus sont nettement meilleurs. En effet, la comparaison avec une solution de référence obtenue par un calcul couplé non itératif et non interpolé est satisfaisante. De plus, d'importantes améliorations ont été apportées en définissant adéquatement la contribution intégrale de la condition limite à l'aide des points de Gauss situés sur l'interfaces du solide. Enfin, il a également été révélé qu'une approximation pertinente du transfert de chaleur nécessite un maillage solide suffisamment fin à l'interface fluide-structure.

International audience; In conjugate heat transfer (CHT) analysis, the fluid-solid heat transfer calculation is realized by expanding the fluid capability to include heat conduction in solid regions neighboring the fluid. Stability conditions of CHT analysis can be provided by a simplified model problem and from this model, two fundamental parameters are introduced in the coupled approach: a "numerical" Biot number, and an optimal coefficient. The first parameter defines the nature of the fluid-solid interaction. The optimal coefficient ensures unconditional stability. Results presented in this paper consider a wide range of steady thermal phenomena from weak to very strong fluid-solid interaction (insulating materials). At the end of the paper, an example of transient CHT is also shown. The methods proposed in this paper illustrate that a model problem can provide effective and practical solutions directly applicable to complex conjugate heat transfer problems.