Flame Hole Research Articles

A method of evaluating cell state based on data augmentation and ViT16

In this paper, based on the model of data augmentation and Vision Transformer 16 (ViT16), a method of assessment for electrolysis cell state is presented to get the real-time information of the current cell state, so as to improve current efficiency of process. Firstly, in order to solve the issue of the small sample data and improve classification accuracy, the method of data augmentation is performed on the flame hole images by using convolutional block attention module to improve auxiliary classifier generativhyhee adversarial network. Secondly, the deep feature data of the flame hole images is extracted by the method of ViT16, and the genetic algorithm is applied to eliminate the redundant feature data to improve the accuracy. Thirdly, the support vector machines model is employed to classify the feature data, and the aluminum cells are classified into cold, hot, and normal. Finally, the actual data are applied to the experiments of the above method, the results of experiments show that this method is better than other methods, and the accuracy of classifying the cell state is as high as 98.677%. This is of great significance for the guidance of aluminum electrolysis production process.

Extinction of solid diffusion flame in microgravity: Details of quenching and blowoff processes

Long duration microgravity experiments aboard the International Space Station determine diffusion flame extinction limits of PMMA spheres. Upon ignition from an electrically-heated coil, the 4-cm-diameter samples are exposed to forced flows ranging from 0.2 to 80 cm/s and an oxygen ranging from 13 to 28 % in one atmosphere total pressure. Extinction is reached as oxygen concentration gradually decreases by natural depletion. Five extinction tests are presented at different flow velocities and oxygen concentrations. Quenching at low velocity is observed with the flame tip shrinking upstream and blowoff is observed when a hole forms in the flame at the forward flow stagnation point. However, these processes are not quasistatic. The quenching motion involves periodic flame tip pulsating toward downstream and shrinking upstream with a continually decreasing flame size at the end of each cycle. In blowoff, the flame base pulsates between the flame hole and the downstream location, although with far fewer cycles compared to quenching. The pulsations appear to be the result of a premixed flame front spreading into a combustible mixture. Two specific cases are discussed in more details. In the first, at very low flow velocity (∼ 0.4 cm/s) and elevated oxygen, self-sustained flame tip cyclic pulsations are observed for a lengthy period (∼15 min). In the second case with a higher flow velocity (50 cm/s), the diffusion flame is stabilized at the shoulder of the spherical sample after local stagnation point blowoff. With steady decrease in ambient oxygen due to depletion, spinning flamelets are formed. The long-duration microgravity environment makes it possible to observe these interesting and detailed extinction processes.

An improved cell situation identification approach with convolutional neural network and wavelet extreme learning machine

In the aluminum reduction process, the flame hole is an influential index of the whole production situation, which reflects the distribution of the physical field of the reduction cell, the current efficiency and the lifespan of the aluminum reduction cell. Therefore, flame hole situation detection and identification are critical and significant in the whole process of aluminum electrolysis. However, in the practical industrial production, flame hole identification result is coming from the experimental operation workers, with the loss of workers and different experimental levels, the real-time measurement of the flame hole index is still a challenge beyond solution. This article develops a flame hole image classification method based on wavelet extreme learning machine. First, a deep feature set is extracted from the original images with a convolutional neural network. Then, a classification model based on an extreme learning machine with wavelet activation function is developed. Finally, the proposed convolutional neural network–weighted extreme learning machine model is applied to superheat degree real-time detection in the industrial electrolysis cell. The proposed method is evaluated on aluminum production which outperforms existing other superheat degree methods in accuracy and robustness.

LES/PDF modeling of swirl-stabilized non-premixed methane/air flames with local extinction and re-ignition

Turbulent non-premixed flames with local extinction and re-ignition exhibit multiple combustion modes including ignition waves, diffusion flames, partially premixed flames, and ignition-assisted partially premixed flames. The mechanisms of local extinction and re-ignition are not well understood and numerical modeling of multi-mode combustion is a challenging task. In this work, a specially designed swirl-burner was used to study local extinction and re-ignition of non-premixed turbulent methane/air flames. High speed Particle Image Velocimetry (PIV) and laser induced fluorescence of OH radicals (OH-PLIF) measurements along with Large Eddy Simulation (LES) were carried out to investigate the mechanisms of extinction and re-ignition processes in the burner. LES is based on a transported probability density function model within the framework of Eulerian Stochastic Fields (PDF-ESF). It is found that local extinction occurs when the scalar dissipation rate around the stoichiometric mixture fraction is high. The characteristic time scale for local extinction and re-ignition in the present flames is an order of magnitude longer than the characteristic time scale of diffusion/extinction of laminar flamelets. There are two mechanisms for flame hole re-ignition in the present flames. First, under low degree of local extinction conditions (i.e., for small flame holes surrounded by flames) the flame hole re-ignition is due to the mechanism of turbulent flame folding. Second, under high degree of extinction conditions (i.e., with large regions of extinction and lifted flames), re-ignition of the locally extinguished flame is due to the mechanism of ignition assisted partially premixed flame propagation. The results show that the PDF-ESF model is capable of simulating the quenching and re-ignition process found in the experiments.

A semi-supervised Laplacian extreme learning machine and feature fusion with CNN for industrial superheat identification

The superheat degree (SD) in industrial aluminum electrolysis cell is a critical index that can maintain the energy balance, improve the current efficiency and improve production. However, the existing SD identification is mainly relying on artificial experience and the accuracy of SD is far from satisfactory. Further, artificial costs and physical equipment are expensive and time-consuming. In this paper, we propose a deep soft sensor method for SD detection. First, CNN is utilized for flame hole image feature extraction. Second, a semi-supervised extreme learning machine (ELM) that integrates Laplacian regularization is further used for SD classification. The main contributions of the paper are: (1) The proposed CNN-LapsELM utilizes the CNN for flame hole image feature extraction and then ELM for further classification, which fully takes advantage of CNN’s ability for complex feature extraction, ELM’s excellent generalization ability, and high computation efficiency. (2) Both the labeled and unlabeled samples are utilized for the CNN-LapsELM training process. It fully leverages the information contained in unlabeled data. At the same time, Laplacian regularization is utilized for learning the manifold structure of hole image samples, so the performance of the proposed CNN-LapsELM are improved. (3) The proposed CNN-LapsELM algorithm improves the generalization ability and robustness. The comparison result demonstrates that the CNN-LapsELM is superior to the existing SD identification and the accuracy is 87%.

Transient interactions between a premixed double flame and a vortex

The interaction between a laminar flame and a vortex is an important study for understanding the fundamentals of turbulent combustion. In the past, however, flame-vortex interactions have been investigated only for high-temperature flames. In this study, the impact of a vortex on a premixed double flame, which consists of a coupled cool flame and a hot flame, is examined experimentally and computationally using dimethyl ether/oxygen/ozone mixtures. The double flame is first shown to occur near the extinction limit of the hot flame. The differences between steady-state cool flames, double flames, and hot flames are explored in a one-dimensional counterflow configuration. The transient interactions between double flames and impinging vortices are then investigated experimentally using a micro-jet and numerically in two-dimensional transient modeling. It is seen that the vortex can extinguish the near-limit hot flame locally, resulting in a lone cool flame. At higher vortex intensities, the cool flame may also be extinguished after the extinction of the hot flame. It is found that there can be three different transient flame structures coexisting at the same time: an extinguished flame hole, a cool flame, and a double flame. Moreover, flame curvature is shown to play an important role in determining whether the vortex weakens or strengthens the cool flame and double flame.

Effect of Acoustic Excitation on Lean Blowoff in Turbulent Premixed Bluff Body Flames

ABSTRACTThe lean blowoff characteristics of a premixed air-methane flame were investigated in a ducted combustor with a bluff body according to acoustic excitation. The blowoff equivalence ratio increases with the Reynolds number and changes depending on the extent of the recirculation zone. Using the relation between the Damköhler number and the Reynolds number, it was confirmed that the flow velocity at the downstream tip of the bluff body and the laminar flame speed are decisive blowoff factors. Although a periodic flame hole appeared far from the blowoff only with acoustic excitation, the blowoff observed by OH radical chemiluminescence occurred using a similar process regardless of the excitation. The recirculation zone collapses and the flame becomes small when it is close to the blowoff. Then, the flame is locally extinguished downstream from the bluff body and the recirculation zone completely collapses. Eventually, the unburned gas does not ignite and the flame is extinguished. The blowoff equivalence ratio rapidly increases at specific acoustic excitation frequencies. This was investigated using proper orthogonal decomposition analysis, the two-microphone method, and phase-lock particle imaging velocimetry measurement. Resonance occurs when the excitation frequency approaches the harmonic frequency of the combustor and it increases the velocity fluctuation in the combustor and the infiltration velocity of the unburned gas in the shear layer of the recirculation zone. Consequently, because the burning velocity must have a larger value corresponding to the enhanced mixture velocity for a sustained flame, the blowoff occurs at a higher equivalence ratio.

A computational approach to flame hole dynamics using an embedded manifold approach

We present a new numerical algorithm for tracking the evolution of flame holes in diffusion flames. The key element is the solution of an evolution equation for a flame state field defined on a complex moving surface. The surface itself can evolve in time and is defined implicitly as a level set of an associated Cartesian scalar field. The surface coordinates, or parameterization, do not need to be determined explicitly. Instead, the numerical method employs an embedding technique where the evolution equation is extended to the Cartesian space. In our application, the flame state field represents the chemical activity state of a diffusion flame; i.e. quenched and burning regions of the flame surface. We present a formulation that describes the formation, propagation, and growth of flame holes using edge-flame modeling in laminar and turbulent diffusion flames. The evolution equation is solved using a high-order finite-volume WENO method and a new extension algorithm defined in terms of propagation PDEs. The complete algorithm is demonstrated by tracking the dynamics of flame holes in a turbulent reacting shear layer and its applicability is also demonstrated in a turbulent reacting lifted jet simulation.

Experimental Characterization of Methane Inverse Diffusion Flame

This article presents 10-kHz images of OH-PLIF simultaneously with 2-D PIV measurements in an inverse methane diffusion flame. Under a constant fuel flow rate, the central air jet Re was varied, leading to air to fuel velocity ratio, Vr, to vary from 8.3 to 66.5. Starting from Vr = 20.7, the flame is commonly characterized by three distinct zones. The length of the lower fuel entrainment region is inversely proportional to Vr. The flames investigated resemble a string shear layer confining this zone, and converging into the second distinct region, the flame neck zone. The third region is the rest of the flame, which spreads in a jet-like manner. The inverse diffusion flames exhibit varying degrees of partial premixing, depending upon on the velocity ratio Vr, and this region of partial premixing evolves into a well-mixed reaction zone along the flame centerline. The OH distribution correlated with the changes in the mean characteristics of the flow through reduction in the local Reynolds number due to heat release. The existence of a flame suppresses or laminarizes the turbulence at early axial locations and promotes fluctuations at the flame tip for flames with Vr < 49.8. In addition, the flame jet width can be correlated to the OH distribution. In upstream regions of the flames, the breaks in OH are counterbalanced by flame closures and are governed by edge flame propagation. These local extinctions were found to occur at locations where large flow structures were impinging on the flame and are associated with a locally higher strain rate or correlated to the local high strain rates at the flame hole edges without this flow impinging. Another contributor to re-ignition was found to be growing flame kernels. As the flames approach global blow-off, these kernels become the main mechanism for re-ignition further downstream of the flames. At low Vr, laminarization within the early regions of the flame provides an effective shield, preventing the jet flow from penetrating through the flame.

부력의 영향을 최소화한 조건에서 대향류 확산화염의 화염 소화에 관한 실험적 연구

Experiments were conducted to clarify role of the outermost edge flame on low-strain-rate flame extinction in buoyancy-suppressed non-premixed methane flames diluted with He and <TEX>$N_2$</TEX>. The use of He curtain flow produced a microgravity level of <TEX>$10^{-2}-10^{-3}g$</TEX> in <TEX>$N_2$</TEX>- and He-diluted non-premixed counterflow flame experiments. The critical He and <TEX>$N_2$</TEX> mole fractions at extinction with a global strain rate were examined at various burner diameters (10, 20, and 25 mm). The results showed that the extinction curves differed appreciably with burner diameter. Before the turning point along the extinction curve, low-strain-rate flames were extinguished via shrinkage of the outermost edge flame with and without self-excitation. High-strain-rate flames were extinguished via a flame hole while the outermost edge flame was stationary. These characteristics could be identified by the behavior of the outermost edge flame. The results also showed that the outermost edge flame was not influenced by radiative heat loss but by convective heat addition and conductive heat losses to the ambient He curtain flow. The numerical results were discussed in detail. The self-excitation before the extinction of a low-strain-rate flame was well described by a dependency of the Strouhal number on global strain rate and normalized nozzle exit velocity.

Role of the outer-edge flame on flame extinction in nitrogen-diluted non-premixed counterflow flames with finite burner diameters

This study of nitrogen-diluted non-premixed counterflow flames with finite burner diameters investigates the important role of the outer-edge flame on flame extinction through experimental and numerical analyses. It explores flame stability diagrams mapping the flame extinction response of nitrogen-diluted non-premixed counterflow flames to varying global strain rates in terms of burner diameter, burner gap, and velocity ratio. A critical nitrogen mole fraction exists beyond which the flame cannot be sustained; the critical nitrogen mole fraction versus global strain rate curves have C-shapes for various burner diameters, burner gaps, and velocity ratios. At sufficiently high strain-rate flames, these curves collapse into one curve; therefore, the flames follow the one-dimensional flame response of a typical diffusion flame. Low strain-rate flames are significantly affected by radial conductive heat loss, and therefore flame length.Three flame extinction modes are identified: flame extinction through shrinkage of the outer-edge flame with or without oscillations at the outer-edge flame prior to the extinction, and flame extinction through a flame hole at the flame center. The extinction modes are significantly affected by the behavior of the outer-edge flame. Detailed explanations are provided based on the measured flame-surface temperature and numerical evaluation of the fractional contribution of each term in the energy equation. Radial conductive heat loss at the flame edge to ambience is the main mechanism of extinction through shrinkage of the outer-edge flame in low strain-rate flames. Reduction of the burner diameter can extend the flame extinction mode by shrinking the outer-edge flame in higher strain-rate flames.

Heat transfer efficiency evaluation for outward and inward multi-flame-hole gas burner

The purpose of this study is to understand the factor that influence the heating efficiency of the outward and inward multi-hole gas burner. The flame-hole angle and the distance from flame hole to heating object are chosen as the experimental parameters. The measurement of the flame temperature distribution is carried out on each experimental condition. The observation of combustion flame, by the Schlieren method, is done from the purpose to understand the combustion phenomenon on the heating efficiency. LPG (Liquefied petroleum gas) is used for the test fuel gas. The compositions of LPG are propane 97.5vol%, butane 0.2vol% and methane + ethylene 2.3vol%. The optimum ranges of the flame-hole angle and the distance from flame hole to heating object are clarified. The experimental correlation equations for the outward and inward multi-flame-hole gas burner are proposed.

대향류 확산화염에서 에지화염이 화염소화에 미치는 영향

대향류확산화염의 화염소화에 있어서 에지화염 역할에 관한 실험적 연구가 진행되었다. 속도비, 버너직경, 그리고 버너간격을 변화시키며 수행된 실험에서 전체신장률에 따른 화염소화 임계질소몰분율의 그래프는 c-커브 형태로 나타났다. 고신장률화염에서는 화염소화 임계질소몰분율의 그래프가 하나의 곡선으로 일치하였으며, 화염이 일차원의 응답특성을 갖는 것을 확인하였다. 화염 소화는 바깥 에지화염이 반경방향으로의 진동 후에 화염 중심으로 수축하며 소화하는 영역, 진동 없이 화염중심으로 수축하며 소화하는 영역, 그리고 바깥 에지부분의 수축과 진동 없이 화염중심에 화염 구멍이 생기며 소화하는 영역으로 세 가지 모드로 나타났다. 화염 표면온도 측정과 에너지 방정식의 각항을 수치해석 한 결과를 토대로 에지화염부분에서의 반경방향 전도 열손실이 에지화염의 불안정을 야기한다는 것과 전도를 통한 열 공급뿐만 아니라 대류를 통한 열 공급도 바깥 에지화염의 안정화에 기여한다는 것을 보였다. 그리고 반경방향의 전도열손실이 수축하며 소화하는 메커니즘의 지배적인 역할을 함을 보였다.

Prediction of the acoustic environment on the launch pad for the Ares I rocket, including the effects of drift and water suppression.

A model, PAD, has been developed for prediction of noise in the vicinity of launch vehicles, with specific application to the mobile launcher and tower for the Ares I launch vehicle. It follows the basic principles of a traditional NASA model [NASA Report No. SP‐8072 (1971)] but uses an updated rocket sound power source model developed by Sutherland [AIAA Paper No. 93‐4383 (1993)] together with updates for the distribution and directivity of sources along the rocket plume. The effect of the deflector on plume geometry and shielding by the launcher deck are included in the model. Overall and spectral (octave band or psd) levels are predicted at specific points or on a regular grid of lateral and vertical positions on and around the launcher. A basic version of PAD was used for nominal acoustic predictions for vertical lift with all exhaust flow going through the flame hole in the deck and no water suppression (AIAA Paper No. 2009‐3163). The model has since been extended to account for lateral drift of the ...

Behavior of Low Strain Rate Flame Disks in Counterflow Diffusion Flame

A study was conducted to clarify flame characteristics through the evaluation of critical mole fractions at flame extinction and edge-flame oscillation of low strain rate flames using the global strain rate, velocity ratio, and burner distance as experimental variables. The transition from a shrinking flame disk to a flame hole was verified through gradient measurements of the maximum flame temperature. Evidence of edge-flame oscillation in flame disks was also found using numerical simulations in zero and normal gravity. The main mechanisms of flame extinction and edge-flame oscillation were analyzed by comparing the energy fractions in the energy equation. For low strain rate flame disks, radial conduction heat loss rather than flame radiation was a significant contributor to flame extinction and even edge-flame oscillation. This was experimentally demonstrated by evaluating the critical mole fraction at flame extinction and edge-flame oscillation, as well as measurements of the flame temperature gradient along the flame disk surface. These results suggest that low strain rate flame responses are determined not only by one-dimensional flame responses, but also by multi-dimensional flame responses such as radial conduction heat loss. The results also show that extinction of low strain rate flames is more probably due to multi-dimensional heat losses than to radiative extinction.

예연소실을 갖는 점화플러그의 연소 특성에 관한 연구

The new concept spark plug was developed to study its influence on the combustion characteristics of SI engine. It has pre-ignition chamber in the lower end of spark plug and flame hole, in which fresh mixture gas can be introduced without any fuel supply system. This spark plug was tested with a commercial SI engine. Fuel consumption rate, emission gas and MBT timing were measured in the engine dynamometer for various flame hole numbers, hole positions, hole sizes of the pre-ignition chamber of the spark plug. And average flame propagation speed was measured by using the head gasket ionization probe in single cylinder engine. The new concept spark plug induces fast bum in combustion compared with the conventional spark plug, and MBT(Minimum advance for Best Torque) timing was retarded about crank angle. The flame hole number, hole direction and volume of pre-ignition chamber were found to influence the combustion characteristics.

ON COMBUSTION AROUND A POROUS BLOWING REACTANT SPHERE IN A LOW REYNOLDS NUMBER CROSSFLOW

The low Reynolds number limit of flow past a blowing reactant sphere serves to illustrate a powerful theoretical approximation for the conservation equations of combustion. For this approximation to work, the crossflow rate measured by the Reynolds number must be at least one order of magnitude smaller than the blowing Stefan flow rate from the porous sphere (droplet). This approximation, which retains in the equations only the irrotational component of the velocity field, enables analytical solutions to be derived for the mixture fraction and all other scalar variables. The mixture fraction solution permits predictions to be made for the flame including its shape and its position, and their dependence on both crossflow and radial sphere blowing. The porous sphere can also be treated as a transient, decaying fuel droplet in a quasi-steady gas. Using this simplification, an analytical droplet burning rate formula is derived and compared with an empirical correlation formula. Diffusion flame hole development is examined as a function of the crossflow.

Simulation of a diffusion flame in turbulent mixing layer by the flame hole dynamics model with level-set method

The partial quenching structure of diffusion flames, arising from the phenomenon of turbulent flame lift off, is investigated in a turbulent mixing layer by the method of flame hole dynamics. Modification of the flame hole dynamics by including the level-set method is specifically aimed to properly take into account the effect of slow flame-edge response near the crossover scalar dissipation rate at which the edge propagation speed vanishes. Simulating the flame hole dynamics with the level-set method results in two major improvements. The first improvement is observed in stabilizing lifted turbulent diffusion flames. The three necessary conditions for lifted stabilization are proposed and numerically tested to show that rapid variation of the edge propagation speed near the crossover scalar dissipation rate helps lifted stabilization. Secondly, an improvement in the statistical properties of the stationary turbulence reacting state is observed in that (1) the lift-off height is found to be higher because of the streamwise flow pushing the turbulent edge front to the downstream direction and (2) the partial burning probability, conditioned with the scalar dissipation rate, exhibits a realistic smooth transition across the crossover scalar dissipation rate.

Modelling of lifted turbulent diffusion flames in a channel mixing layer by the flame hole dynamics

The partial quenching structure of turbulent diffusion flames in a turbulent mixing layer is investigated by the method of flame hole dynamics as an effort to develop a prediction model for the turbulent flame lift off. The essence of the flame hole dynamics is derivation of the random walk mapping, from the flame-edge theory, which governs expansion or contraction of the quenching holes initially created by the local quenching events. The numerical simulation for the flame hole dynamics is carried out in two stages. First, a direct numerical simulation is performed for a constant-density fuel–air channel mixing layer to obtain the background turbulent flow and mixing fields, from which a time series of two-dimensional scalar-dissipation-rate array is extracted. Subsequently, a Lagrangian simulation of the flame hole random walk mapping, projected to the scalar dissipation rate array, yields a temporally evolving turbulent extinction process and its statistics on partial quenching characteristics. In particular, the probability of encountering the reacting state, while conditioned with the instantaneous scalar dissipation rate, is examined to reveal that the conditional probability has a sharp transition across the crossover scalar dissipation rate, at which the flame edge changes its direction of propagation. This statistical characteristic implies that the flame edge propagation instead of the local quenching event is the main mechanism controlling the partial quenching events in turbulent flames. In addition, the conditional probability can be approximated by a heavyside function across the crossover scalar dissipation rate.

Flame holes and flame disks on the surface of a diffusion flame

Flame holes and flame disks in a laminar axisymmetric counterflow configuration are numerically investigated for unity Lewis number, with the strain rate as the control parameter. The temporal evolution of the topological structure of flame holes and flame disks is described in detail for different representative strain rates. It is found that corresponding to each given strain rate, there exists a critical hole (disk) radius , with the strain rate (or equivalently, Damkohler number) as a parameter. For the flame hole case, it is found that in the final stage of collapse of a hole, the edge-flame velocity is essentially proportional to the inverse of the hole radius, except when the strain rate is very close to the extinction limit. Flame interactions induced by overlapping of pre-heat zones are mainly responsible for the acceleration of the edge-flame velocity when the hole radius approaches zero, and it is further enhanced by the focusing effects of hole curvature in the plane of the stoichiometric surface. For the flame disk, the increasing heat loss rate plays a major role on the acceleration of the shrinking speed when the disk radius approaches zero.