Session Q25: Reacting Flows: Combustion Kinetics and Modeling

Skeletal Reaction Models for Atmospheric and High Pressure Combustion of Methane

Skeletal reaction mechanisms for atmospheric and high pressure combustion of methane are generated from the foundational fuel chemistry model (FFCM-1) via the forced-optimally time dependent (f-OTD) methodology. In the f-OTD methodology, the sensitivity matrix, i.e., a large matrix containing all local sensitivities of a system, is modeled as the multiplication of two low-rank time-dependent matrices. The evolution equations of these matrices are derived from the governing equations of the system. For skeletal mechanism reduction, the sensitivity of mass fractions and temperature with respect to the reaction rates are considered. These modeled sensitivities are computed for the auto-ignition problem with different sets of initial temperatures, pressures, and equivalence ratios. The calculated sensitivities are then analyzed to rank the most sensitive species. A series of skeletal mechanisms with different levels of accuracy in reproducing the results of the detailed kinetics model, i.e. FFCM-1 are then produced. This skeletal reduction technique is conducted both for atmospheric and high pressure combustion of methane. The performances of the generated models are compared against FFCM-1 based on their abilities in predicting ignition delays, laminar flame speeds, and diffusion flame extinctions. Results suggest a skeletal model with 27 species can accurately reproduce the results of FFCM-1 for both atmospheric and high pressure test cases.

    

Reduced order model for stiffness removal in chemically reacting flow simulations

The computation of finite-rate chemical kinetics forms the bottleneck of modern reacting flow simulations due to the inherent stiffness of the system. The operator splitting approach has been introduced to eliminate the time step requirements due to stiffness. However, integrating the operator split chemistry ordinary differential equations (ODEs) is computationally intensive and limits the overall performance of the modern reacting flow simulations. In this work, we introduce a machine learning-based autoencoder approach to eliminate the stiffness of the system. Specifically, the operator split chemistry ODEs are transformed into a latent space using autoencoders and integrated using the neural ordinary differential equation (neural ODE) approach. This approach results in elimination of the stiffness and thus a relaxed time step requirement. Furthermore, the dynamics of the reduced variables in the latent space, obtained from autoencoders and principal component analysis, are compared for two different chemistry mechanisms: i) H2-air and ii)CH4-air. The results obtained from the constant pressure batch reactor integration with H2 and CH4 mixtures show a good agreement with Cantera.   

Development, comparison, and validation of an all-reactions and all-species H2/O2/N2 mechanism.

This presentation focuses on a comprehensive H₂/O₂/N₂ chemical kinetics mechanism and its validation for combustion at both atmospheric plus high-pressure and high-temperature operating conditions. This work is motivated by the lack of validated H₂/Air detailed kinetic mechanisms available for the prediction of NO_x species at high-pressure and high-temperature conditions. Only core H₂/O₂ mechanisms are available at these conditions but those do not enable the assessment of NO_x levels. On another hand, detailed mechanisms with NO_x as well as other species are available but only validated under atmospheric conditions. This presentation is centered on filling this gap to address the prediction of all species including NO_x at relevant operating conditions. The present mechanism development is based on available state-of-the-art experimental measurements and kinetics data. The developed mechanism consists of 33 species and 224 reactions. The method to develop this mechanism has three steps. The first step is to gather an experimental database including fundamental measurement data over a wide range of initial combustion conditions and various experimental devices. The second step is to update the key reaction rates and includes new reaction pathways to form a revised H₂/O₂ core mechanism. The evaluation of four widely utilized NO_x chemistry at both atmospheric and high-pressure conditions for stirred reactor and in-flame data is made to select a baseline. Finally, the last step is to combine both the updated core mechanism with the assessed nitrogen chemistry. The obtained mechanism is validated on a comprehensive data set for a wide range of inlet temperature and pressure for three selected targets: laminar flame speed, ignition time delay, and NO_x level. This process leads to a novel all-reactions and all-species mechanism for H₂/Air and H₂/O₂ mixtures combusting in future thermal-powered aircraft under high pressure and high-temperature conditions with hydrogen fuel.   

Development and Analysis of a Multiphase Model-Informed Closure Relation for Steady Detonation Behavior of Energetic Materials

The steady detonation behavior of energetic materials (EM) in response to shock loading is a key component of performance characterization. The reactive Euler equations serve as the theoretical foundation of many hydrocodes that are used to predict the behavior of EM at engineering scales. An expression referred to as a closure relation is required to ensure a unique solution to the system equations. However, the form of this closure relation, as well as its ability to faithfully represent the underlying physics, continues to be a topic of significant debate within the energetics community. Unlike traditional pressure-dependent reactive burn models such as ignition-and-growth, the use of more recent models for energetics may lead to inaccurate predictions if the selection of closure relation is not appropriate. Here, we discuss the results of a unified mixture/multiphase modeling framework to express the closure relation in terms of simplified inter-constituent interactions. This allows us to pose a new closure relation that better reflects the transport processes between mixture components. We examine the effects of this new closure model on the reaction zone structure for ideal and non-ideal EM, and compare the results to those using conventional closures.   

Large-eddy simulation/probability density function modeling of a pre-chamber turbulent jet ignition

To circumvent the adverse effects of climate change and to achieve net zero emissions by the year 2050, it is essential to have a steady progress towards reducing the carbon footprints. Turbulent Jet Ignition (TJI) is an innovation in which the spark plug of a spark-ignited combustion engine is replaced by a small volume pre-chamber to produce a high-temperature jet causing main-chamber ignition. TJI enables more robust lean combustion and thereby increases fuel efficiency. In this study, computational fluid dynamics simulations of an experimental TJI rig are conducted by using the combined Large-Eddy Simulation (LES) and transported Probability Density Function (PDF) method to assess the model's predictive capability for the relevant lean premixed combustion regime. A consistent Eulerian Monte-Carlo Fields (EMCF) approach is used to approximate the multi-dimensional transported PDF equation. A machine learning assisted mixing model is employed for the micro-mixing process in the stochastic differential equations. The predicted pre-chamber flame propagation and the lean main-chamber ignition is validated by using the available experiments data.   

A study of the fourth order joint statistical moment for dimensionality reduction of combustion datasets

Principal Component Analysis (PCA) is a widely used dimensionality reduction technique that uses the eigenvectors of the second-order joint statistical moment, i.e., the covariance matrix to transform thethermochemical state space. However, combustion simulations are often characterized by thin reaction zones that occupy small fractions of the computational domain. These zones are, statistically speaking, represented by a small set of samples which can be viewed as extreme valued events that break the assumption of an underlying Gaussian distribution. Furthermore, it is also reasonable that the underlying distribution of the thermochemical scalars be considered as non-Gaussian. Thus, it can be conjectured that the ability of PCA based reduced manifolds to capture important local chemical dynamics may not be optimal. In this work, the principal vectors needed for dimensionality reduction are obtained through a singular value decomposition of a matricized representation of the fourth order joint statistical moment, namely the Co-Kurtosis tensor. The performance of the proposed dimensionality reduction procedure is evaluated, and compared against PCA, for two combustion datasets namely the spontaneous ignition of premixed ethylene in a homogeneous reactor and that of a homogeneous charge compression ignition, for two reduced manifolds and the linear reconstruction technique. It is found that the proposed method captures the chemical dynamics, as represented by the heat release rate, approximately 1.85 times better (averaged over all results) than PCA.   

A phenomenological model for the impact of nanosecond repetitively pulsed discharges on a laminar methane-air flame

Nanosecond repetitively pulsed (NRP) discharges are a promising technique to enhance combustion efficiency and control. Numerical studies are essential to improve the understanding of complex plasma-combustion interaction. Limited by the prohibitive computational cost of fully coupled detailed plasma mechanism and combustion chemistry, a phenomenological model taken from literature is further developed to study the behavior of a laminar methane-air flame under NRP discharges. The phenomenological model focuses on two channels through which the electric energy is deposited: 1) the ultrafast heating and ultrafast dissociation of O₂ coming from the relaxation of electronically excited N₂; and 2) the slow gas heating coming from the relaxation of vibrational states of N₂. The energy fraction deposited to these two channels is governed by the reduced electric field (E/N) which cannot be accurately predicted without resolving ion transport. Electric field is instead determined by solving the static poisson equation between two pin electrodes with three tested geometries. The predicted flame displacement under plasma qualitatively matches the experimental result. The roles played by chemical and thermal effects are strongly dependent on the E/N profile. Higher prediction of E/N magnitude at the preheating zone results in stronger dissociation effect and modifies the flame morphology more than what a lower E/N prediction concludes.   

Constrained model of the stretched flame speed in laminar and turbulent premixed flames

Coupling with thermo-diffusive effects, the flame stretch alters flame structures and thereby the burning velocity. We propose an analytical model for the stretched flame speed. Constraints on the maximum/minimum limits of the stretched flame speed make this model able to represent the nonlinear variation of the stretched flame speed from weak to strong flame stretch. Validations against different configurations, including counterflow flames and inwardly/outwardly propagating spherical flames, show that the constrained model well captures the flame speed response to the flame stretch. The model can be employed for laminar flame speed extrapolation and turbulent premixed flame modeling. Extrapolations based on experimental and numerical data show that the present model improves the accuracy of laminar flame speed extrapolation from stretched flames, especially for lean hydrogen flames. This is due to the strong nonlinearity caused by the thermo-diffusive effects, which is well captured by the proposed model. We also validate the model using the direct numerical simulation (DNS) data of turbulent premixed flames at weak and moderate turbulence intensities. Comparisons show reasonable agreements between model predictions and the DNS data on the local flame displacement speed.