Cinematographic stereoscopic PIV, combined with Taylor's frozen flow hypothesis, is used to generate three-dimensional (3D) quasi-instantaneous pseudo volumes of the three-component (3C) velocity field in the far field of turbulent nonpremixed jet flames at jet exit Reynolds number Reδ in the range 8,000-15,300. The effect of heat release, however, lowers the local (i.e., based on local properties) Reynolds number to the range 1,500-2,500. The 3D data enable computation of all nine components of the velocity gradient tensor ∇u from which the major 3D kinematic quantities, such as strain rate, vorticity, dissipation and dilatation, are computed. The volumetric PIV is combined with simultaneously acquired 10 Hz OH planar laser-induced fluorescence (PLIF). A single plane of the OH distribution is imaged on the center-plane of the volume and provides an approximate planar representation of the instantaneous reaction zone. The pseudo-volumes are reconstructed from temporally and spatially resolved kilohertz-rate 3C velocity field measurements on an end-view plane (perpendicular to the jet flame axis) invoking Taylor's hypothesis. The interpretation of the measurements is therefore twofold: the measurements provide a time-series representation of all nine velocity gradients on a single end-view plane or, after volumetric reconstruction, they offer a volumetric representation, albeit approximate, of the spatial structure of the flow. The combined datasets enable investigation of the fine-scale spatial structure of turbulence, the effect of the reaction zone on these structures and the relationship between the jet kinematics and the reaction zone. Emphasis is placed on the energy dissipation field and on the presence and role of dilatation. Statistics of the components of the velocity gradient tensor and its derived quantities show that these jet flames exhibit strong similarities to incompressible turbulent flows, such as in the distribution of the principal strain rates and strain-vorticity alignment. However, the velocity-gradient statistics show that these jet flames do not exhibit small-scale isotropy but exhibit a strong preference for high-magnitude radial gradients, which are attributed to regions of strong shear induced by the reaction zone. The pseudo volumes reveal that the intense-vorticity field is organized in two major classes of structures: tube-like away from the reaction zone (the classical worms observed in incompressible turbulence) and sheet-like in the vicinity of the local reaction zone. Sheet-like structures are, however, the dominant ones. Moreover, unlike incompressible turbulence where sheet-like dissipative structures enfold, but don't coincide with, clusters of tube-like vortical structures, it is observed that the sheet-like intense-vorticity structures tend to closely correspond to sheet-like structures of high dissipation. The primary reason for these features is believed to be due to the stabilizing effect of heat release on these relatively low local Reynolds number jet flames. It is further observed that regions of both positive and negative dilatation are present and tend to be associated with the oxidizer and fuel sides of the OH zones, respectively. These dilatation features are mostly organized in small-scale, short-lived blobby structures that are believed to be mainly due to convection of regions of varying density rather than to instantaneous heat release rate. A model of the dilatation field developed by previous researchers using a flamelet approximation of the reaction zone was used to provide insights into the observed features of the dilatation field. Measurements in an unsteady laminar nonpremixed jet flame where dilatation is expected to be absent support the simplified model and indicate that the observed structure of dilatation is not just a result of residual noise in the measurements, although resolution effects might mask some of the features of the dilatation field. The field of kinetic energy dissipation is further investigated by decomposing the instantaneous dissipation field into the solenoidal, dilatational and inhomogeneous components. Analysis of the current measurements reveals that the effect of dilatation on dissipation is minimal at all times (it contributes to the mean kinetic energy dissipation only by about 5-10%). Most of the mean dissipation arises from the solenoidal component. On average, the inhomogeneous component is nearly zero, although instantaneously it can be the dominant component. Two mechanisms are believed to be important for energy dissipation. Near the reaction zone, where the stabilizing effect of heat release generates layers of laminar-like shear and hence high vorticity, solenoidal dissipation (which is proportional to the enstrophy) dominates. In the rest of the ow the inhomogeneous component dominates in regions subjected to complex systems of nested vortical structures where the mutual interaction of interwoven vortical structures in intervening regions generates intense dissipation. / text

Electrically assisted combustion has been studied in order to control or improve flame characteristics, and emphasizing efficiency and emission regulation. Many phenomenological observations have been reported on the positive impact of electric fields on flame, however there is a lack of detailed physical mechanisms for interpreting these. To clarify the effects of electric fields on flame, I have investigated flame structure, soot formation, and flow field with ionic wind electrical current responses in nonpremixed counterflow flames. The effects of direct current (DC) electric field on flame movement and flow field was also demonstrated in premixed Bunsen flames. When a DC electric field was applied to a lower nozzle, the flames moved toward the cathode side due to Lorentz force action on the positive ions, soot particles simultaneously disappeared completely and laser diagnostics was used to identify the results from the soot particles. To understand the effects of an electric field on flames, flow visualization was performed by Mie scattering to check the ionic wind effect, which is considered to play an important role in electric field assisted combustion. Results showed a bidirectional ionic wind, with a double-stagnant flow configuration, which blew from the flame (ionic source) toward both the cathode and the anode. This implies that the electric field affects strain rate and the axial location of stoichiometry, important factors in maintaining nonpremixed counterflow flames; thus, soot formation of the counterflow flame can also be affected by the electric field. In a test of premixed Bunsen flames having parallel electrodes, flame movement toward the cathode and bidirectional ionic wind were observed. Using PIV measurement it was found that a created radial velocity caused by positive ions (i.e. toward a cathode), was much faster than the velocity toward the anode. Even in a study of alternating current (AC) electric fields, bidirectional ionic wind could be observed, regardless of applied frequencies. Therefore, the effect of ionic wind cannot be considered negligible under both DC and AC electric fields. Detailed explanations for electrical current, flame behavior, and flow characteristics under various conditions are discussed herein.

The performance of a high-power small-scale combustion-based fluidic actuator for high-speed flow control applications is characterized with specific focus on comparisons between premixed and nonpremixed operating modes for the device. Momentary (pulsed) actuation jets are produced by the ignition of a mixture of gaseous fuel and oxidizer within a small (cubic centimeter scale) combustion chamber. The combustion process yields a high pressure burst (1 to 3 ms in duration in the typical configurations) and the ejection of a high-speed exhaust jet. The actuation frequency can be continuously varied by independently controlling the flow rate of the fuel/oxidizer and the spark ignition frequency up to a maximum determined by the operating characteristics of the actuator. The actuator performance is characterized by both its peak thrust and net total impulse, with increases in peak jet momentum often two to three orders of magnitude above the baseline steady jet. Results for operation of the device in both premixed and nonpremixed modes are presented and analyzed, with nonpremixed operation typically yielding higher pressures and greater frequency ranges in the present configurations.

An experimental study of sooting characteristics of laminar underventillated ethylene non-premixed flames in hot vitiated environments was performed using a modified Wolfhard-Parker co-flowing slot burner. The burner could be operated in "single mode" with a cold air/oxygen mixture as the oxidizer for the non-premixed flame or in varying degrees of "dual mode" where the products of lean premixed hydrogen/air/oxygen flames were used as the oxidizer for the non-premixed flame. Premixed flame stoichiometries of 0.3 and 0.5 were considered for the dual mode cases. Dual mode operation of the burner was intended to simulate the conditions of fuel rich pockets of gas burning in the wake of previously burned fuel/air mixture as typically found in real nonpremixed combustion devices. Dual mode operation introduced competing thermal and chemical effects on soot chemistry. Experimental conditions were chosen to match peak nonpremixed flame temperatures among the cases by varying oxidizer inert (N2) concentration to minimize the dual mode thermal effect. In addition the molecular oxygen (post premixed flame for dual mode cases) and ethylene fuel flow rates were held constant to maintain the same overall equivalence ratio from case to case. Thermocouple thermometry utilizing a rapid insertion technique and radiation corrections yielded the gas temperature field. Soot volume fractions were measured simultaneously with temperature using Thermocouple Particle Densitometry (TPD). Soot volume fraction, particle size and particle number density fields were measured using laser light scattering and extinction. Gas velocities were measured using Particle Imaging Velocimetry (PIV) on the non-premixed flame centerline by seeding the ethylene flow and calculated in the oxidizer flow stream. Porous sinters in the oxidizer slots prevented oxidizer particle seeding required for PIV measurements. In general as the degree of dual mode operation was increased (i.e. increasing stoichiometry of the premixed flames) soot volume fractions decreased, particle sizes increased and soot particle number densities decreased. This trend is suspected to be result of water vapor elevating OH concentrations near the flame front in dual mode operation reducing soot particle nucleation early in the flame by oxidizing soot precursors. The larger particle sizes measured at later stages of dual mode flames are suspected to be the result of lower competition for surface growth species for the lower particle number densities in those flames. Integrated soot volume fraction and particle number fluxes at various heights in the flame decreased with increasing degree of dual mode operation. / Master of Science

Combustion-generated carbon black nano particles, or soot, have both positive and negative effects depending on the application. From a positive point of view, it is used as a reinforcing agent in tires, black pigment in inks, and surface coatings. From a negative point of view, it affects performance and durability of many combustion systems, it is a major contributor of global warming, and it is linked to respiratory illness and cancer. Laser-Induced Incandescence (LII) was used in this study to measure soot volume fractions in four steady and twenty-eight pulsed ethylene diffusion flames burning at atmospheric pressure. A laminar coflow diffusion burner combined with a very-high-speed solenoid valve and control circuit provided unsteady flows by forcing the fuel flow with frequencies between 10 Hz and 200 Hz. Periodic flame oscillations were captured by two-dimensional phase-locked LII images and broadband luminosity images for eight phases (0°- 360°) covering each period. A comparison between the steady and pulsed flames and the effect of the pulsation frequency on soot volume fraction in the flame region and the post flame region are presented. The most significant effect of pulsing frequency was observed at 10 Hz. At this frequency, the flame with the lowest mean flow rate had 1.77 times enhancement in peak soot volume fraction and 1.2 times enhancement in total soot volume fraction; whereas the flame with the highest mean flow rate had no significant change in the peak soot volume fraction and 1.4 times reduction in the total soot volume fraction. A correlation (ƒv Reˉ1 = a+b· Str) for the total soot volume fraction in the flame region for the unsteady laminar ethylene flames was obtained for the pulsation frequency between 10 Hz and 200 Hz, and the Reynolds number between 37 and 55. The soot primary particle size in steady and unsteady flames was measured using the Time-Resolved Laser-Induced Incandescence (TIRE-LII) and the double-exponential fit method. At maximum frequency (200 Hz), the soot particles were smaller in size by 15% compared to the steady case in the flame with the highest mean flow rate.

On s’intéresse ici aux erreurs de modélisation liées à l’usage de modèles de flammelette sous-maille en combustion turbulente non prémélangée. Le but de cette thèse est de développer une stratégie d’estimation d’erreur a posteriori pour déterminer le meilleur modèle parmi une hiérarchie, à un coût numérique similaire à l’utilisation de ces mêmes modèles. Dans un premier temps, une stratégie faisant appel à un estimateur basé sur les résidus pondérés est développée et testée sur un système d’équations d’advection-diffusion-réaction. Dans un deuxième temps, on teste la méthodologie d’estimation d’erreur sur un autre système d’équations, où des effets d’extinction et de réallumage sont ajoutés. Lorsqu’il n’y a pas d’advection, une analyse asymptotique rigoureuse montre l’existence de plusieurs régimes de combustion déjà observés dans les simulations numériques. Nous obtenons une approximation des paramètres de réallumage et d’extinction avec la courbe en «S», un graphe de la température maximale de la flamme en fonction du nombre de Damköhler, composée de trois branches et d’une double courbure. En ajoutant des effets advectifs, on obtient également une courbe en «S» correspondant aux régimes de combustion déjà identifiés. Nous comparons les erreurs de modélisation liées aux approximations asymptotiques dans les deux régimes stables et établissons une nouvelle hiérarchie des modèles en fonction du régime de combustion. Ces erreurs sont comparées aux estimations données par la stratégie d’estimation d’erreur. Si un seul régime stable de combustion existe, l’estimateur d’erreur l’identifie correctement ; si plus d’un régime est possible, on obtient une fac˛on systématique de choisir un régime. Pour les régimes où plus d’un modèle est approprié, la hiérarchie prédite par l’estimateur est correcte. / We are interested here in the modeling errors of subgrid flamelet models in nonpremixed turbulent combustion. The goal of this thesis is to develop an a posteriori error estimation strategy to determine the best model within a hierarchy, with a numerical cost at most that of using the models in the first place. Firstly, we develop and test a dual-weighted residual estimator strategy on a system of advection-diffusion-reaction equations. Secondly, we test that methodology on another system of equations, where quenching and ignition effects are added. In the absence of advection, a rigorous asymptotic analysis shows the existence of many combustion regimes already observed in numerical simulations. We obtain approximations of the quenching and ignition parameters, alongside the S-shaped curve, a plot of the maximal flame temperature as a function of the Damköhler number, consisting of three branches and two bends. When advection effects are added, we still obtain a S-shaped curve corresponding to the known combustion regimes. We compare the modeling errors of the asymptotic approximations in the two stable regimes and establish new model hierarchies for each combustion regime. These errors are compared with the estimations obtained by using the error estimation strategy. When only one stable combustion regime exists, the error estimator correctly identifies that regime; when two or more regimes are possible, it gives a systematic way of choosing one regime. For regimes where more than one model is appropriate, the error estimator’s predicted hierarchy is correct.

No description available.