Complexe de Recherche Interprofessionnel en Aérothermochimie

This study reports 2-year survival outcomes among patients with advanced nonprogressive stage IIA to IV lung cancer randomized to symptom monitoring during chemotherapy via web-based patient-reported outcomes vs standard scheduled imaging after treatment to detect symptomatic recurrence.

This chapter contains sections titled: Introduction Two-component PIV Three-component PIV Bibliography

Background: The use of web-based monitoring for lung cancer patients is growing in interest because of promising recent results suggesting improvement in cancer and resource utilization outcomes. It remains an open question whether the overall survival (OS) in these patients could be improved by using a web-mediated follow-up rather than classical scheduled follow-up and imaging. Methods: Advanced-stage lung cancer patients without evidence of disease progression after or during initial treatment were randomly assigned in a multicenter phase III trial to compare a web-mediated follow-up algorithm (experimental arm), based on weekly self-scored patient symptoms, with routine follow-up with CT scans scheduled every three to six months according to the disease stage (control arm). In the experimental arm, an alert email was automatically sent to the oncologist when self-scored symptoms matched predefined criteria. The primary outcome was OS. Results: From June 2014 to January 2016, 133 patients were enrolled and 121 were retained in the intent-to-treat analysis; 12 deemed ineligible after random assignment were not subsequently followed. Most of the patients (95.1%) had stage III or IV disease. The median follow-up was nine months. The median OS was 19.0 months (95% confidence interval [CI] = 12.5 to noncalculable) in the experimental and 12.0 months (95% CI = 8.6 to 16.4) in the control arm (one-sided P = .001) (hazard ratio = 0.32, 95% CI = 0.15 to 0.67, one-sided P = .002). The performance status at first detected relapse was 0 to 1 for 75.9% of the patients in the experimental arm and for 32.5% of those in the control arm (two-sided P < .001). Optimal treatment was initiated in 72.4% of the patients in the experimental arm and in 32.5% of those in the control arm (two-sided P < .001). Conclusions: A web-mediated follow-up algorithm based on self-reported symptoms improved OS due to early relapse detection and better performance status at relapse.

Heat release effects on laminar flame propagation in partially premixed flows are studied. Data for analysis are obtained from direct numerical simulations of a laminar mixing layer with a uniformly approaching velocity field. The structure that evolves under such conditions is a triple flame, which consists of two premixed wings and a trailing diffusion flame. Heat release increases the flame speed over that of the corresponding planar premixed flame. In agreement with previous analytical work, reductions in the mixture fraction gradient also increase the flame speed. The effects of heat release and mixture fraction gradients on flame speed are not independent, however; heat release modifies the effective mixture fraction gradient in front of the flame. For very small mixture fraction gradients, scaling laws that determine the flame speed in terms of the density change are presented.

▪ Abstract Non-premixed turbulent combustion processes control most practical applications of combustion. Studying these mechanisms has been the objective of numerous theoretical and experimental works in the last century. In the past 10 years, direct numerical simulation (DNS) has emerged as a new methodology. It has become an essential tool to understand and model turbulent combustion. DNS numerically solves the set of equations describing turbulent flames by resolving all chemical and flow scales. Because formulated assumptions can be tested accurately, the resulting data provide unique information to build turbulent combustion models. This review first discusses the fundamental properties of laminar diffusion flames. It then presents various DNS results to illustrate the different problems that can be studied using this tool. These problems include validity of the modeling hypothesis, topology of flame surfaces, and ignition and extinction mechanisms. The review also discusses the different analysis techniques employed to extract information from DNS and explores the limits of these methods.

A strategy to build the next generation of fluid dynamics solvers able to fully benefit from high-performance computing is discussed. The procedure relies on a domain decomposition of unstructured meshes that is organized in two levels. The computing cells are first gathered at an elementary level in cell groups; at a second level, cell groups are dispatched over processors. Compared to the usual single-level domain decomposition, this double domain decomposition allows for easily optimizing the use of processor memory and therefore load balancing in both Eulerian and Lagrangian contexts. Specific communication procedures to handle faces, edges and nodes are associated to this double domain decomposition, which strongly reduce the computing cost; input–output times are optimized as well. In addition, any multi-level solution techniques, as deflated preconditioned conjugate gradient, are well-adapted to such mesh decomposition. This approach has been used to develop the YALES2 code, which also benefits from a non-degenerescent tessellation algorithm for tetrahedra to automatically generate high-resolution meshes on super-computers. To illustrate the capabilities of the YALES2 algorithmic, an aeronautical burner is fully simulated with a mesh of 2.6 billion cells, followed by a demonstration test over 21 billion cells.

International audience

In this article, a Eulerian model for the atomization of a liquid jet is proposed. Atomization is considered as turbulent mixing in a flow with variable density in the limit of large Reynolds and Weber numbers. An assumption similar to the Kolmogorov hypothesis has been invoked: large-scale features of the flow are supposed to be independent of viscosity and surface tension at high Reynolds and Weber numbers; small-scale features do depend on viscosity and surface tension. Dispersion of the liquid in the gas phase is computed by a classical equation for the turbulent diffusion flux of the liquid. The mean size of the liquid fragments is obtained with a new equation for the mean surface area of the liquid–gas interface per unit of volume. Discussions concerning this new equation are presented. Several comparisons with experiments are presented, сoncerning the liquid dispersion as well as the mean size of fragments produced.

cited By 124

The generalized Lorenz-Mie theory deals with the interaction between spheres and arbitrarily shaped illuminating beams. An efficient use of the theory requires efficient evaluation of the so-called beam-shape coefficients involved in the description of the illuminating beam. A less time-consuming method of evaluation relies on the localized approximation. However, it lacks flexibility when the description of the illuminating beam is modified. We present a new version of this method, called the integral localized approximation, that exhibits the desired property of flexibility.

cited By 115

An efficient numerical procedure for computing the scattering coefficients of a multilayered sphere is discussed. The stability of the numerical scheme allows us to extend the feasible range of computations, both in size parameter and in number of layers for a given size, by several orders of magnitude with respect to previously published algorithms. Exemplifying results, such as scattering diagrams and cross-sectional curves, including the case of Gaussian beam illumination, are provided. Particular attention is paid to scattering at the rainbow angle for which approaches based on geometrical optics might fail to provide accurate enough results.

A nonlinear time-dependent collisional-radiative model for recombining argon is presented. Reactions involving Ar(2)+ are taken into account and their influence is discussed. It is shown that Ar(2)+ may increase the time to reach the quasi-steady-state by a factor of 100. The calculation of the recombination rate coefficient at the quasi-steady-state is presented. An analytical expression is derived and compared with existing literature values. The importance of the increase of the quasi-steady-state time is illustrated by comparisons of excited levels population densities distribution measured in a fast moving plasma where the mechanical time scale is sufficiently short to provide a time-dependent chemistry in a reference frame moving with the flow. The high sensitivity of the results towards the electron number density is pointed out. Finally, the influence of the processes involving Ar(2)+ on the excitation temperature is discussed.

Partially premixed flames are observed in nonpremixed turbulent combustion when fuel and oxidizer have mixed before burning. This combustion regime combines the properties of both premixed and diffusion flames. A procedure based on the resolved fields is proposed to associate premixed and diffusion flame descriptions in large eddy simulation. Using basic and well known subgrid modelling of premixed and diffusion flames, the proposed methodology is tested for flames lifted in a two-dimensional turbulent wake. Very recent experimental observations concerning the dynamics of the flow field at the turbulent flame base are reproduced.

Relying on van de Hulst’s localization principle, a localized approximation to the generalized Lorenz-Mie theory is introduced. The validation of this simple approximation is obtained from numerical comparisons the Rayleigh-Gans theory. Other comparisons concerning scattering profiles are carried out first with theoretical data published in the literature and later with experimental measurements. Original results are given for coal particles as an example of the versatility of the method.

Natural convection in cavities is an interesting subject for many researchers. Especially, in recent years, the number of articles written in this regard has grown enormously. This work provides a review of recent natural convection studies. At first, experimental studies were reviewed and, then, numerical studies were examined. Then, the articles were classified based on effective parameters. In each section, numerical studies were examined the parameters added to the cavity such as magnetic forces, fin, porous media and cavity angles. Moreover, studies on non-rectangular cavities were investigated. Free convection in enclosures depends more on the fluid velocity relative to the forced convection, leading to the opposite effect of some parameters that should essentially enhance rate of heat transfer. Nanoparticle addition, magnetic fields, fins, and porous media may increase forced convection. However, they can reduce free convection due to the reduction in fluid velocity. Thus, these parameters need more precision and sometimes need the optimization of effective parameters.

cited By 84

Turbulence is still one of the main challenges in accurate prediction of reactive flows. Therefore, the development of new turbulence closures that can be applied to combustion problems is essential. Over the last few years, data-driven modeling has become popular in many fields as large, often extensively labeled datasets are now available and training of large neural networks has become possible on graphics processing units (GPUs) that speed up the learning process tremendously. However, the successful application of deep neural networks in fluid dynamics, such as in subfilter modeling in the context of large-eddy simulations (LESs), is still challenging. Reasons for this are the large number of degrees of freedom in natural flows, high requirements of accuracy and error robustness, and open questions, for example, regarding the generalization capability of trained neural networks in such high-dimensional, physics-constrained scenarios. This work presents a novel subfilter modeling approach based on a generative adversarial network (GAN), which is trained with unsupervised deep learning (DL) using adversarial and physics-informed losses. A two-step training method is employed to improve the generalization capability, especially extrapolation, of the network. The novel approach gives good results in a priori and a posteriori tests with decaying turbulence including turbulent mixing, and the importance of the physics-informed continuity loss term is demonstrated. The applicability of the network in complex combustion scenarios is furthermore discussed by employing it in reactive and inert LESs of the Spray A case defined by the Engine Combustion Network (ECN).

A nonlinear time-dependent two-temperature collisional-radiative model for air plasma has been developed for pressures between 1kPa and atmospheric pressure to be applied to the flow conditions of space vehicle re-entry into the Earth’s atmosphere. The model consists of 13 species: N2, O2, N, O, NO, N2+, O2+, N+, O+, NO+, O2−, O− in their ground state and major electronic excited states and of electrons. Many elementary processes are considered given the temperatures involved (up to 10 000K). Time scales to reach the final nonequilibrium or equilibrium steady states are derived. Then we apply our model to two typical re-entry situations and show that O2− and O− play an important role during the ionization phase. Finally, a comparison with existing reduced kinetic mechanisms puts forward significant discrepancies for high velocity flows when the flow is in chemical nonequilibrium and smaller discrepancies when the flow is close to chemical equilibrium. This comparison illustrates the interest of using a time-dependent collisional-radiative model to validate reduced kinetic schemes for the relevant time scales of the flows studied.

A so-called localized approximation, allowing one to speed up the evaluation of beam shape coefficients in the generalized Lorenz–Mie theory for spheres, has been previously introduced and, in the case of Gaussian beams, rigorously justified. We examine and demonstrate the validity of this approximation for arbitrary shaped beams.