Combustion Flow Field Research Articles

A Deep Learning Framework for Supersonic Turbulent Combustion

The rapid simulation of supersonic turbulent combustion is a significant demand in scientific research and engineering applications for hypersonic vehicles. This paper proposes a deep learning framework for fast predicting unsteady turbulent combustion flow fields within the combustor of hypersonic vehicles. Based on convolutional neural networks and recurrent neural networks, this framework extracts spatial distribution characteristics of the flow fields and temporal evolution rules. And we enhance the traditional mean square error loss function by assigning loss weights to different channel data. Numerical simulations are conducted on the model scramjet combustor with various geometric structures to generate the dataset for training, and part of the untrained cases are used to verify the effectiveness. The results show that the proposed model, under different geometric structures, achieves high computational accuracy, with a correlation coefficient between the predicted results and the true values above 0.99. Considering the time cost of data transferring between heterogeneous systems, the model takes only 30 seconds to complete the calculation, representing an acceleration of at least two orders of magnitude compared to computational fluid dynamics. In the future, it can be applied to the rapid prediction of hypersonic vehicle performance and efficiently guide the optimal design of aircraft.

Numerical and experimental study of the combustion performance of an integrated inclined combustor

To shorten the length of an aero-engine combustor and reduce its weight, we propose an integrated inclined combustor. A combination of numerical calculations and experimental research methods is employed to conduct relevant research on the cold and combustion flow fields and ignition dynamic processes of the integrated inclined combustor. Results show that the inclination angle of the swirler has minimal effect on the total pressure loss and combustion efficiency of the integrated inclined combustor. However, as the swirler’s inclination angle increases, the uniformity of the outlet temperature distribution and recirculation flow rate of the primary combustion zone decrease. Furthermore, when the integrated inclined combustor is ignited, the flame does not propagate symmetrically to both sides in the circumferential direction but propagates faster along the inclination direction of the swirler.

Numerical study of aluminum combustion with agglomerate size distribution in solid rocket motor

The aluminum agglomerate size distribution plays an important role in influencing the particle distributed combustion in motor, which subsequently affects the performance of solid rocket motor significantly. In this work, a three-phase model to describe distributed combustion of aluminum agglomerates is established based on the Eulerian-Lagrangian method. Then, the agglomerate size, including mono size and distributed size, is studied to reveal its effect on aluminum combustion and motor flow field. The simulated results indicate that the increasing agglomerate mono size observes the obvious decrease of the average temperature inside the motor combustion chamber, implying the low combustion efficiency of large agglomerate size. When considering the agglomerate size distribution, the size distribution mode and the mean size D43 determine the combustion efficiency together. In particular, even the mean size is similar, with different distribution mode, like skewed distribution, bimodal or trimodal distribution, the combustion efficiency and flow field parameters are nonnegligible different. However, when the size distribution mode is the same and the peak range is similar, the mean size D43 becomes the only and predominant factor as the mono size.

Numerical simulation of the effects of pulsed injection on combustion and flow processes in a supersonic combustor

This paper describes two-dimensional numerical simulations of pulsed injection to a hydrogen-fueled air-breathing ramjet using the Reynolds-averaged Navier–Stokes method. We analyze the effects of sinusoidal pulsed injection on the start-up ignition time, start-up ignition range, combustion efficiency, and flow field evolution of the ramjet and compare the performance with that of steady injection. The results show that pulsed injection shortens the ignition time and optimizes the ignition equivalence ratio of hydrogen. Pulsed injection also improves the combustion efficiency of hydrogen, increasing the thrust of the engine by 7.79% compared with steady injection under the same equivalence ratio. We find that pulsed injection can cause oscillations in the ramjet combustion flow field, and show that the oscillation frequency is affected by the pulsed injection frequency.

Combustion flow field and thermal protection performance of a liquid oxygen/kerosene engine with pintle injectors

A pintle injector is a convenient method for adjusting propellant combustion and has received extensive attention in recent years. However, few studies have examined the effects of the injection direction and momentum ratio (ratio of liquid oxygen momentum to kerosene momentum) on the flow field, temperature field, and dual-belt liquid film cooling effect in the combustion chamber. This study verified a new computational fluid dynamics (CFD) engine combustion model based on a pintle injector through a hot-test experiment. A simulation was conducted using this model. With different injection directions, when the momentum ratio was increased, the trends in the combustion efficiency and chamber pressure growth were entirely opposite. For the axial kerosene and radial liquid oxygen injection direction of liquid film cooling, when the momentum ratio was 0.73–0.9, the proportion of the first ring belt liquid film flow was selected as 5–10 %, and the second ring belt liquid film flow proportion was selected as 5–7%. When the momentum ratio was less than 0.73, the liquid film flow proportions for both the first and second ring belts were selected as 5 %. When the total liquid film flow proportion was 10 %, the chamber pressure decreased by 0.2–0.3 MPa, and the combustion efficiency decreased by 8–10 %. This numerical study provides guidance for studying the combustion characteristics and cooling performance of pintle-type rocket combustion chambers.

On the Analysis of Scale Effects of Flame Stabilization in Scramjet Combustors

Scale effects of flame stabilization in scramjet combustors at Mach 2.52 are investigated through experiments and numerical simulations based on two geometrically similar combustors. As the equivalence ratio increases, the combustion mode will change from Scram to Dual to Ram. However, the larger combustor has a delayed mode transition. The scale effects of flame stabilization are also reflected in the changes of heat release, precombustion shock train length, jet penetration depth, and reaction zone location. The combustion flowfields of different-scale combustors exhibit numerous differences under different combustion modes. However, after shortening the isolator of the proportionally scaled small-scale combustor, the combustion flowfield is closer to that of the large-scale combustor. Further analysis reveals the reason for the inconsistent scale effects of flame stabilization under different combustion modes. The probable reason is that the boundary layer, which does not change proportionally with the combustor scale, has different effects on the combustion under the three modes. Compared with the Scram and Ram modes, the scale effect in the Dual mode is the most significant because the strong interaction between heat release and the boundary layer amplifies the slight difference in the relative thickness of the boundary layer.

Experimental study of rotating direction for dual-axial swirlers on the flow field and combustion characteristics of aero-engine combustor

The flow field structure for aero-engine combustor is one of the most important factors for combustion performance. To investigate how the rotating direction of dual-axial swirlers affects the flow field and combustion characteristics, particle image velocimetry (PIV) experiments were conducted in the primary combustion and intermediate zones. Additionally, planar laser induced fluorescence (PLIF) was used to map the distribution of kerosene and OH in the primary combustion zone. The results show that the swirling jets in co-swirl exhibit greater jet momentum, resulting in a longer primary recirculation zone and a larger deflection angle of the primary jets. This ultimately leads to a higher recirculation rate compared to counter-swirl. The primary jets define the boundary of the flow field structure. The rotating direction primarily affects the flow field structure upstream of the primary jets, exerts a relatively weaker influence on the intermediate zone, which is mainly affected by the deflection direction of the primary jets, and essentially has no impact on the dilution jets. Co-swirl has a relatively stable flow field than counter-swirl, whereas the flow field in counter-swirl is more chaotic, featuring a greater number of vortex cores in the primary recirculation zone. The coupling between the primary and swirling jets affects the mass transfer direction in the intermediate zone, which is reflected in the fact that the mass transfer is from top-left to bottom-right in co-swirl, while it is in the opposite direction in counter-swirl. OH is predominantly found in the swirling shear layers, the primary and secondary combustion zones, whereas kerosene is mainly concentrated in the shear layers of swirling jets. The kerosene distribution in co-swirl shows a larger and more concentrated expansion angle compared to counter-swirl. In contrast, the OH distribution downstream of counter-swirl is broader, with more intense combustion due to enhanced kerosene decomposition and fuel-air mixing. The smaller primary recirculation zone in counter-swirl forces the primary combustion zone to extend downstream, leading to a more intense reaction in the secondary recirculation zone.

Numerical study on the duty cycle of a pulsed jet in the combustion flow field

Numerical study on the duty cycle of a pulsed jet in the combustion flow field

CO formation characteristics in a centrally staged swirl combustor

Based on a validated numerical simulation methodology, this study focused on a centrally staged swirl combustor to investigate the combustion flow field inside the chamber under typical operating conditions. The CO formation characteristics of each region in the combustor were analyzed. In terms of elementary reactions, this study analyzed crucial reaction pathways contributing to CO formation in distinct regions utilizing the eddy dissipation concept model. Results reveal the occurrence of CO formation in the flame front and its rapid oxidation to form CO2 in the post-flame zone. CO2 partially dissociates into CO at high temperatures. Furthermore, air film quenching substantially impacts CO formation and emission, and air film greatly affects CO oxidation more than its formation. Local cooling and dilution effects from the air film contribute to the freezing of CO oxidation reactions. Therefore, a substantial quantity of CO remains unreacted, leading to increased CO emissions at the combustion chamber outlet.

Study of combustion flow field in a scramjet engine under inflow oscillation variations based on OpenFOAM

In this study, a compressible supersonic numerical model was established based on the open-source computational fluid dynamics software OpenFOAM. The supersonic air inflow was set in sinusoidal variation and gas hydrogen was employed as fuel for the DLR engine numerical simulation. The SST k-ω model and a 9-species 27-reaction hydrogen/oxygen chemical reaction model was employed as the turbulence and reaction mechanical kinetics models respectively. The effect of continuous variable inflow on the combustion performance for the scramjet engine was analyzed and the results showed that the combustion flame inside the scramjet engine exhibited significant partitioning phenomena around the combustion stabilization zones under periodic continuous variable inflow conditions. It was found that intense combustion zones were formed at both upstream and downstream flow as the induction of intersecting shock waves. When the flame was stabilized in the engine, the initiation position of the intense combustion zone at the upstream was transferred to the distance of twice the strut length, while the initiation position of the intense combustion zone at the downstream was anchored at a distance of about five times the strut length.

Influence of fluidic shock control on the combustion flow field of a dual-mode scramjet

To improve the resistance capability of a dual-mode scramjet isolator, fluidic shock control devices with different control parameters were designed. First, the effects of the control parameters on the position distribution, morphological structure, evolution process of the shock train in the isolator and the resistance capability of the isolator were numerically simulated. Then, cold- and hot-state tests of the scramjet through a direct-connected pulse combustion wind tunnel were carried out at a simulated flight Mach number of 4 to obtain the ignition and combustion process in the scramjet combustor under fluidic shock control through measurements by high-speed schlieren and pressure transducers. Finally, the influence of fluidic shock control on supersonic combustion was analysed by comparing the flow field structure and combustion process with and without a control device at different equivalent ratios and under different fuel injection schemes. The results showed that the fluidic shock control parameters with a great influence on the resistance capability of the isolator were the angle and width of the suction slot and the angle of the injection slot and that the resistance capability of the isolator could be effectively improved by reasonably designing the control parameters. Under fluidic shock control, the morphological structure and positional distribution of the shock train showed an obvious hysteresis evolution process near the injection slot and the suction slot in the isolator with increasing and decreasing back pressures, and this hysteresis effect could further improve the combustion performance of the combustor.

Intelligent reconstruction of unsteady combustion flow field of scramjet based on physical information constraints

To alleviate the problem of high-fidelity data dependence and inexplicability in pure data-driven neural network models, physical informed neural networks (PINNs) provide a new learning paradigm. This study constructs an efficient, accurate, and robust PINN framework for predicting unsteady combustion flow fields based on Navier–Stokes (NS) equation constraints. To achieve fast prediction of a multi-physical field in a scramjet combustion chamber, we propose a U-shaped residual neural network model based on feature information fusion. The model uses a residual neural network module as the backbone, uses jump connection to improve model generalization, and uses the U-shaped structure to fuse the receptive field features with different scales to enhance the feature expression ability of the model. To prevent improper assumptions from leading to wrong method constraints, we consider the flow characteristic mechanism of each physical field to constrain the neural network and verify its accuracy through numerical simulation of the unsteady flow field in the scramjet combustor with Mach number (Ma) 2.0. This method can accurately predict the multi-physical field of unsteady turbulent combustion based on the time, space, Ma and turbulent eddy viscosity coefficients of a small number of samples. Specially, the proposed physical driven and data driven fusion proxy model can predict the unsteady combustion flow field in milliseconds. It has important reference value to solve the problem of low calculation efficiency of a traditional numerical simulation method of a combustion process.

Advances in Femtosecond Coherent Anti-Stokes Raman Scattering for Thermometry

The combustion process is complex and harsh, and the supersonic combustion flow field is also characterized by short duration and supersonic speed, which makes the real-time diagnostic technology for the transient environment extremely demanding. It is of great significance to realize high time-resolved accurate measurement of temperature, component concentration, and other parametric information of the combustion field to study the transient chemical reaction dynamics of the combustion field. Femtosecond CARS spectroscopy can effectively avoid the collision effect between particles in the measurement process and reduce the influence of the non-resonant background to improve the measurement accuracy and realize the time-resolved measurement on a millisecond scale. This paper introduces the development history of femtosecond CARS spectroscopy, points out its advantages and disadvantages, and looks forward to the future development trend to carry out high time-resolved measurements, establish a database of temperature changes in various complex combustion fields, and provide support for the study of engine mechanisms.

Numerical study of combustion flow field characteristics of industrial gas turbine under different fuel blending conditions

Numerous studies have shown that the mixture of fuel and oxidant significantly influences the flow characteristics and pollutant emissions within the combustion flow field. This study investigates the effects of various methane fuel blending ratios (0.19–0.41) on combustion flow, emission characteristics, performance parameters, and field synergies through numerical simulation. Combustion simulation employs the SST k-ω and flame-generated manifold models. The findings indicate that as the fuel blending ratio increases from 0.19 to 0.41, the temperature at the front end of the cyclone gradually rises, resulting in an average outlet temperature increase of 165 K. The outlet temperature uniformity index (γT) shows minimal variation, while the Pattern Factor (PF) and NO emissions at the outlet gradually increase. Additionally, the fuel blending ratio (α) exhibit a limited impact on CO emissions at the outlet. When α ≥ 0.39, the peak temperature is located in the cyclonic region, resulting in an extremely heterogeneous outlet temperature distribution. Under various operational conditions, NO is primarily concentrated in the high-temperature region near the wall, and the pollutant emission merit function (fe) gradually increase with α. The optimum field synergy and overall performance for the low-resistance heat transfer are attained at α = 0.37. Significant velocity shocks are observed in the near-wall regions on both sides of the combustion chamber, when α ≥ 0.35, based on DMD analysis. To achieve low NO emission, uniform outlet temperature, and ensure safe, long-term operation of the gas turbine, the fuel blending ratio of 0.3 to 0.35 is advised.

Experimental study on combustion flow field characteristics of swirl combustor

The impact of the swirl number on the flow field of a single-stage swirl combustor is investigated using the particle image velocimetry technology. The variations in recirculation zone size, pulsating region, turbulent distribution, vorticity, and Reynolds stress within the combustor are summarized through quantitative analysis of the flow field. Experimental results indicate the following: (1) Under the same air mass flow rate, the length of the recirculation zone in the combustion state is shorter than that in the cold state. (2) The length of the recirculation zone and the axial vortex spacing display a decreasing trend as the swirl number increases, while the width of the recirculation zone demonstrates an increasing trend. (3) For the single-stage swirl combustor, the primary pulsating region is at the swirling jet area at the exit of the swirl. As the swirl number increases, the standard deviation of radial velocity fluctuations and turbulent kinetic energy also increase. (4) The strong shear region of the single-stage swirl combustor can be divided into inner and outer shear layers based on the vorticity distribution and the Q criterion. The vortices in the inner and outer shear layers exhibit opposite orientations according to the vorticity distribution. Overall, the research results can provide basic experimental data for numerical simulation of swirl combustion.

Investigating the influence of combustion flow field position on intelligent enhancement in moiré fringe analysis using deep learning

Moiré tomography, holds a pivotal role in visualizing structures and inverting key parameters for flow fields, with its robust stability and extensive measuring range. And, extracting information from moiré fringes with the highest possible accuracy and efficiency remains one of the most challenging problems. In this study, an intelligent moiré fringe analysis method based on U-Net is proposed, which can accurately and efficiently extract wrapped phase information for moiré fringes. To verify the feasibility of the proposed method, three sets of experiments were conducted, resulting in the acquisition of more than 3600 frames moiré fringes. By training the fringes, three prediction models are obtained, and these models are subsequently utilized for prediction. The results manifest that the average RMSE of the predictions for the corresponding three groups is less than 0.220, when the training set and the test set have matching flow fields in the optical path. However, the prediction effect is not ideal if there is no match, which means the predictive performance is influenced by the position of the flow field in the optical path in the training set. Therefore, it is essential to consider the influence of the flow field position in the process of achieving intelligence in moiré tomography. In a word, the pertinent conclusions drawn in this paper provide valuable insights for enhancing the intelligence level of moiré tomography.

Supersonic combustion field evolution prediction in scramjet engine using a deblurring multi-scale attention network

Traditional deep learning technologies often overlook crucial spatiotemporal information when reconstructing supersonic combustion flow fields. These models can only reconstruct the current flow field, resulting in poor visual quality due to the atomization effect. This research introduces experimental studies on scramjet combustion flows with variable injection pressures at the inlet of a scramjet Mach 2.5, creating image datasets of flame evolution processes over different predicted time spans (5, 10, and 15 ms). It develops a multi-scale attention algorithm with deblurring (DB-MSAN) capabilities to predict the flame image after a certain period using the wall pressure signal from the previous moment. The algorithm employs two multi-head attention channels at different scales to extract timing information, encouraging the model to focus on the anisotropy of the input data and to identify more significant features. The DB module utilizes the adapted atmospheric scattering model to jointly estimate the transmission map and atmospheric light, achieving an exceptional defogging effect with a minimal computational expense. The model’s overall performance is assessed regarding efficiency and accuracy, comparing DB-MSAN’s performance differences with those of Chen’s and the ResNet16 models across multi-span, large-span, various test sets, and sparse wall measurement points. Experimental results indicated that, compared to other methods, DB-MSAN achieves a maximum peak signal-to-noise index improvement of 76.10 % and an SSIM index improvement of 155.26 % when predicting average-quality images. The prediction accuracy remains stable, even with sparse pressure input. In addition, DB-MSAN strikes an optimal balance between the number of parameters and model accuracy; it represents only 16.97 % of the volume compared to Chen’s model, which is 585 MB.

Effect of Gas Addition on Liquid Kerosene Combustion in a Scramjet Combustor

Experiments were conducted at Mach 2.52 to investigate the effect of gas addition on the mixing and combustion characteristics of the liquid kerosene in a scramjet combustor. Two different combustion modes (scram mode and ram mode) were achieved by varying the equivalence ratio. Flowfield diagnostic techniques, such as shadow visualization and wall pressure measurement, and Proper Orthogonal Decomposition analysis were employed to investigate the complex supersonic gas– liquid two-phase combustion flowfield. It is found that, compared to the pure high-pressure kerosene, the gas-added kerosene can achieve better combustion performance at low injection pressure. The gas addition increases the fuel-jet penetration depth and promotes flowfield fluctuations. The gas–liquid two-phase instability induces large-scale vortices within the fuel-jet shear layer. The diffusion and mixing of kerosene are enhanced by the large-scale vortices, and the rapid and intense combustion of kerosene is realized.

Application of UVAS and TDLAS-based multi-combustion-parameter diagnosis using computerized tomography

Fossil fuel combustion is the main source of mechanical power and will be accompanied by the production of a large number of toxic and harmful substances, such as NO and CO. In this study, Computed Tomography-Tunable Diode Laser Absorption Spectroscopy (CT-TDLAS) and Computed Tomography-Ultraviolet Absorption Spectroscopy (CT-UVAS) were used to establish a multi-combustion-parameter diagnosis system. A corrected H2O absorption database was applied by the wavelength at 1388 nm and 1343 nm. The spectra at 2325 nm were used to establish the CO absorption database. NO absorption database at 214 nm/226 nm was established by UVAS. The nonlinear least squares with the polynomial fitting method was applied to CT reconstruction. A custom CT cell was designed to fix the fiber and probes of the CT-TDLAS and CT-UVAS, and its rationality was evaluated by the Sum of squared difference (SSD) and Zero-mean normalization cross-correlation (ZNCC) through the comparison between CFD simulation and CT reconstruction. In the experiment, the CT reconstruction of the temperature, H2O, and CO showed significant distribution profile differences between two premixed flames. The NO detection function was demonstrated by standard NO gas, and the maximum average relative error was 4.99 %. The results show that the multi-combustion-parameter diagnosis system can measure the temperature, H2O, CO, and NO distribution of the combustion flow field at high frequency. The multi-combustion-parameter diagnosis system will assist with research and diagnosis on controlling NO generation and promoting combustion efficiency.

Initiation process of non-premixed continuous rotating detonation wave through Schlieren visualization

High-speed schlieren visualization on the initiation process of non-premixed air and hydrogen continuous rotating detonation (CRD) is carried out in a well-designed rounded-rectangle hollow combustor in this study. The CRD initiation processes of pre-detonator ignition and spark plug ignition are comprehensively revealed by synchronous high-speed schlieren images and high-frequency dynamic pressure results. The results show that the CRD undergoes three processes after ignition, including initiation process, adjustment process and stable propagation process. The CRD initiation process of pre-detonator ignition involves the upstream-propagation phase of leading shock wave and reaction zone and the coupling phase of circumferentially-propagating shock wave and reaction zone, while that of spark plug ignition includes the upstream-propagation phase of reaction zone, and the coupling phase of circumferentially-propagating pressure wave and reaction zone. The formation of deflagration combustion flowfield and re-establishment of combustible mixture layer after ignition are important prerequisites for CRD initiation. The strong interaction between circumferentially-propagating shock wave and post-shock wave reaction conduces to the rapid initiation of CRD by pre-detonator ignition, while the pressure wave gradually interacts with reaction leading to the detonation initiation by spark plug ignition. The CRD initiation time and adjustment time of pre-detonator ignition are shorter than those of spark plug ignition. This study provides detailed insight into the physics of CRD initiation process of pre-detonator and spark plug ignition, which can deepen the understanding of the initiation mechanism and provide practical guidance for the optimal design of CRDE ignitor.