Gas turbine combustion chambers contain numerous smallscale features that help to dampen acoustic waves and alter the acoustic mode shapes. This damping helps to alleviate problems such as thermoacoustic instabilities. During computational fluid dynamics simulations (CFD) of combustion chambers, these small-scale features are often neglected as the corresponding increase in the mesh cell count augments significantly the cost of simulation while the small physical size of these cells can present problems for the stability of the solver. In problems where acoustics are prevalent and critical to the validity of the simulation, the neglected small-scale features and the associated reduction in overall acoustic damping can cause problems with spurious, nonphysical noise and prevents accurate simulation of transients and limit cycle oscillations. Low-order dynamical systems (LODS) and artificial neural networks (ANNs) are proposed and tested in their ability to represent a simple two-dimensional acoustically forced simulation of an orifice at multiple frequencies. These models were built using compressible CFD, using OpenFOAM, of an orifice placed between two ducts. The acoustic impedance of the orifice has been computed using the multi-microphone method and compared to a commonly used analytical model. Following this, the flow field downstream of the orifice has been modelled using both a LODS and ANN model. Both methods have shown the ability to closely represent the simulated dynamical flows at much lower computational cost than the original CFD simulation. Such models may also assist in the accurate simulation of flame quenching due to cooling flows or the design of effusion cooled aerodynamic surfaces such as nozzle guide vanes (NGVs) and turbine blades.
The present work details the analysis of the aerodynamics of an experimental swirl stabilized burner representative of gas turbine combustors. This analysis is carried out using High Frequency PIV (HFPIV) measurements in a reactive situation. While this information is usually available at a rather low rate, temporally resolved PIV measurements are necessary to better understand highly turbulent swirled flows, which are unsteady by nature. Thanks to recent technical improvements, a PIV system working at 12 kHz has been developed to study this experimental combustor flow field. Statistical quant…
Combustion instabilities are a major concern in the design of Liquid Rocket Engines (LREs) and gas turbines. During this PhD work, several directions were explored to understand and mitigate their effects. First, more efficient and robust numerical methods for their prediction in complex combustors were designed. In this matter, a novel type of modal expansion, named a frame expansion and comparable to the classical Galerkin expansion, was introduced to build more accurate acoustic Low-Order Models (LOMs), able to account for the full geometrical complexity of industrial combustors. In particu…
In practical combustion systems, the high nonlinearity of the turbulent combustion process can result in unexpected behavior even for nominal operating conditions. This includes flame flashback in premixed gas turbines, engine unstart in scramjets, and failure of engines to ignite at high altitudes. Many of these events have catastrophic consequences. Hence, understanding the nature of such irregular excursions of combustion systems is important to ensure robust operation. The focus of this work is to identify the sources of such rare excursions from a dynamical systems perspective. Specifical…
The accurate and robust simulation of transcritical real-fluid effects is crucial for many engineering applications, such as fuel injection in internal combustion engines, rocket engines and gas turbines. For example, in diesel engines, the liquid fuel is injected into the ambient gas at a pressure that exceeds its critical value, and the fuel jet will be heated to a supercritical temperature before combustion takes place. This process is often referred to as transcritical injection. The largest thermodynamic gradient in the transcritical regime occurs as the fluid undergoes a liquid-like to a…
The accurate prediction of small scales in underresolved flows is still one of the main challenges in predictive simulations of complex configurations. 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. In fact, the successful application of deep neural networks in fluid dynamics, such as for underresolved reactive flows, is still challenging. This work advances …
The high cost of high-resolution computational fluid/flame dynamics (CFD) has hindered its application in combustion related design, research and optimization. In this study, we propose a new framework for turbulent combustion simulation based on the deep learning approach. An optimized deep convolutional neural network (CNN) inspired from a U-Net architecture and inception module is designed for constructing the framework of the deep learning solver, named CFDNN. CFDNN is then trained on the simulation results of hydrogen combustion in a cavity with different inlet velocities. After training,…
Spectral effects of radiating gases on the ignition in a multiswirl staged model combustor using full-spectrum k distribution method -- A Large Eddy Simulation Investigation
Large eddy simulation (LES) has become the de-facto computational tool for modeling complex reacting flows, especially in gas turbine applications. However, readily usable general-purpose LES codes for complex geometries are typically academic or proprietary/commercial in nature. The objective of this work is to develop and disseminate an open source LES tool for low-Mach number turbulent combustion using the OpenFOAM framework. In particular, a collocated-mesh approach suited for unstructured grid formulation is provided. Unlike other fluid dynamics models, LES accuracy is intricately linked …
Stanford University（斯坦福大学）；SLAC National Accelerator Laboratory（SLAC国家加速器实验室）
Many practical combustion systems such as those in rockets, gas turbines, and internal combustion engines operate under high pressures that surpass the thermodynamic critical limit of fuel-oxidizer mixtures. These conditions require the consideration of complex fluid behaviors that pose challenges for numerical simulations, casting doubts on the validity of existing subgrid-scale (SGS) models in large-eddy simulations of these systems. While data-driven methods have shown high accuracy as closure models in simulations of turbulent flames, these models are often criticized for lack of physical …
In lean premixed combustors, flame stabilization is an important operational concern that can affect efficiency, robustness and pollutant formation. The focus of this paper is on flame lift-off and re-attachment to the nozzle of a swirl combustor. Using time-resolved experimental measurements, a data-driven approach known as cluster-based reduced order modeling (CROM) is employed to 1) isolate key flow patterns and their sequence during the flame transitions, and 2) formulate a forecasting model to predict the flame instability. The flow patterns isolated by the CROM methodology confirm some o…
Combustion in automotive and aerospace applications employing diesel, gas turbine and liquid rocket engines is preceded by injection and mixing of fuel and oxidizer at high pressures, often exceeding mixture critical values. Experimental observations indicate that the jets injected at supercritical pressures exhibit significantly different dynamics than the jets at subcritical conditions, owing to the lack of distinct liquid and gas phases in supercritical state. As a result, the averaged flow quantities such as the potential core length, jet spatial growth rate and velocity decay profiles dif…
Flame-wall interaction of thermodiffusively unstable hydrogen/air flames -- Part II: Parametric variations of equivalence ratio, temperature, and pressure
Accurate simulations of high-pressure transcritical fuel sprays are essential for the design and optimization of next-generation gas turbines, internal combustion engines, and liquid propellant rocket engines. Most important and challenging is the accurate modelling of complex real-gas effects in high-pressure environments, especially the hybrid subcritical-to-supercritical mode of evaporation during the mixing of fuel and oxidizer. In this paper, we present a novel modeling framework for high-fidelity simulations of reacting and non-reacting transcritical fuel sprays. In this method, the high…
Direct Numerical Simulation of Thermoacoustically Unstable Flames Via Non-Drifting Acoustic Delay Characteristic Boundary Conditions
School of Engineering, University of Manchester（曼彻斯特大学工程学院）；Department of Mathematics, University of Manchester（曼彻斯特大学数学系）；School of Engineering, Cardiff University（卡迪夫大学工程学院）
The estimation of model parameters with uncertainties from observed data is a ubiquitous inverse problem in science and engineering. In this paper, we suggest an inexpensive and easy to implement parameter estimation technique that uses a heteroscedastic Bayesian Neural Network trained using anchored ensembling. The heteroscedastic aleatoric error of the network models the irreducible uncertainty due to parameter degeneracies in our inverse problem, while the epistemic uncertainty of the Bayesian model captures uncertainties which may arise from an input observation's out-of-distribution natur…
