The main objective of this research is to develop a kinetic model reduction framework that enables incorporation of detailed chemistry with realistic flow simulation. Comprehensive computational fluid dynamics tools and detailed kinetic mechanisms have been developed, and the fully integration of these two components has been recognized as an imperative necessity to represent realistic systems. However, integrating detailed chemistry in complex flow simulation is expensive and oftentimes prohibitive. Thus this work is driven by the premise to reduce the computational intensity introduced by including detailed chemistry in realistic flow simulation and meanwhile retain acceptable accuracy. The work in this dissertation is focused on the development of an efficient yet accurate kinetic reduction method that enables dynamic reduction within the context of reactive simulations. Excessive computational intensity introduced by the integration of detailed chemistry in reactive flow simulation stems from the large size of detailed kinetic models. The kinetic model reduction method proposed in this dissertation is to address the following two unique aspects: (i) effective reduction of model size; and (ii) efficient integration of the reduction method dynamically during reactive flow simulation without introducing significant overhead. The proposed method is based on an element flux analysis approach which provides an indicator to quantify element transitions between species. The element flux can be further implemented to retrieve useful information from the kinetic network and identify active species under given reaction conditions, which constitute the fundamental of kinetic analysis and redundancy identification in mechanism reduction. It is demonstrated in this research that element flux analysis gives rise to both an effective and efficient dynamic mechanism reduction method, as well as a useful kinetic analysis tool. The proposed approaches can be extended to multiple disciplines since a large number of applications in novel fuel development, engine design, and petro chemistry require the efficient modeling of reactive flows.

Computational fluid dynamics simulations of chemically reacting flows, which incorporate detailed kinetic models, exhibit a very large computational expense, which hinders the ability of these programs to be used as a design tool for combustion devices. However, this issue can be corrected by simplifying the chemistry and thus decreasing the size of the kinetic model. Kinetic model reduction is an active area of research, and this thesis adds to the literature by making further progress in developing a method capable of generating small, accurate reduced kinetic models in an efficient amount of time.

The primary objective of this research is to characterize fuel combustion in reactive flow simulations using advanced kinetic modeling and mechanism reduction tools. Since incorporating detailed chemical kinetic model in the realistic reactive flow simulations is a computationally challenging task due to the large size of detailed kinetic mechanism, it is of great interest to develop approaches for simplifying the kinetic models and reducing computational costs in reactive flow simulations. In this dissertation, we first extend the previously developed on-the-fly reduction approach to the characterization of complex biodiesel combustion using detailed biodiesel surrogate mechanism. Major combustion characteristics such as ignition, emission, as well as engine performance for biodiesel compared with conventional fossil fuels are studied. Although the incorporation of detailed biodiesel combustion mechanism in complex reactive flow simulation is enabled, the simulation is still highly time-consuming. To further alleviate the computational intensity, a hybrid reduction scheme coupling the on-the-fly reduction with global quasi-steady-state approximation (QSSA) is developed. The proposed hybrid reduction scheme is demonstrated in various reactive flow simulations including zero-dimensional PFR model, multidimensional HCCI engine CFD model, and realistic gas phase injector CFD simulations. A flux-based quasi-steady-state (QSS) species selection procedure is introduced to facilitate the demonstration of hybrid scheme. Finally, a novel computational framework integrating automated mechanism generation and on-the-fly reduction is proposed and implemented using a stepwise integration. The proposed framework is then demonstrated in methane oxidation case studies and shows a new way of conducting reactive flow simulation without having an actual mechanism before the simulation starts. The integration of automated mechanism generation and on-the-fly reduction is a promising technique to perform reactive flow simulations and has the potential to reduce the computational cost of the simulations. The work in this dissertation provides powerful tools and important insight for the incorporation of detailed chemical kinetics in the reactive flow simulations.

This monograph addresses the state of the art of reduced order methods for modeling and computational reduction of complex parametrized systems, governed by ordinary and/or partial differential equations, with a special emphasis on real time computing techniques and applications in computational mechanics, bioengineering and computer graphics. Several topics are covered, including: design, optimization, and control theory in real-time with applications in engineering; data assimilation, geometry registration, and parameter estimation with special attention to real-time computing in biomedical engineering and computational physics; real-time visualization of physics-based simulations in computer science; the treatment of high-dimensional problems in state space, physical space, or parameter space; the interactions between different model reduction and dimensionality reduction approaches; the development of general error estimation frameworks which take into account both model and discretization effects. This book is primarily addressed to computational scientists interested in computational reduction techniques for large scale differential problems.

Contributed articles presented in the International Conference on Advances in the Theory of Ironmaking and Steelmaking; organized by the Dept. of Material Engineering, IISc., Bangalore.

This work details efforts to reduce the cost of using detailed chemical kinetic modeling in realistic reactive-flow simulations, utilizing analytical Jacobian evaluation and vectorized-computing on the central processing unit (CPU), graphics processing unit (GPU) and other hardware-accelerators. The first part of this thesis investigated GPU-based ordinary differential equation (ODE) methods for stiff chemical kinetics. A fifth-order implicit Runge--Kutta method and two fourth-order exponential integration methods were implemented for the GPU and paired with the analytical chemical kinetic Jacobian software pyJac. The performance of each algorithm was compared with a commonly used CPU-based implicit integrator CVODEs. The implicit Runge--Kutta method running on a single Tesla C2075 GPU was equivalent to CVODEs running on 12-38 CPU cores for integration of hydrogen and methane kinetic models using a smaller global integration time-step, however the performance of the GPU-solver degraded at a larger time-steps due to thread divergence and higher memory traffic. The second part of this work investigated the performance of vectorized evaluation of constant-pressure/volume thermochemical source-term and sparse/dense chemical kinetic Jacobians using single-instruction, multiple-data (SIMD) and single-instruction, multiple thread (SIMT) paradigms; the developed codes were additionally incorporated into pyJac. A new formulation of the chemical kinetic governing equations was derived and verified, resulting in greatly increased Jacobian sparsities. Significant speedups were found for shallow-vectorized OpenCL source-rate evaluation as compared with a parallel OpenMP code, increasing for sparse and dense chemical kinetic Jacobian evaluation. Further, the developed work was shown to be orders of magnitude faster than a simple first-order finite-difference Jacobian approach. Finally, several CPU-vectorized linearly-implicit Rosenbrock solvers were adapted for use with pyJac, and validated against CVODEs. The open-source computational fluid dynamics code OpenFOAM was extended to utilize the vectorized solvers, and the eddy dissipation concept combustion model was adapted for their use. The OpenFOAM-coupled vectorized solver was validated over a range of zero-dimensional homogeneous ignition problem against Cantera, before its performance and precision were compared to built-in OpenFOAM solvers for a case modeling the Sandia Flame D; a speedup of 12-15x was found for the vectorized solver.

Physiology of Sugarcane looks at the development of a suite of well-established and developing biofuels derived from sugarcane and cane-based co-products, such as bagasse. Chapters provide broad-ranging coverage of sugarcane biology, biotechnological advances, and breakthroughs in production and processing techniques. This single volume resource brings together essential information to researchers and industry personnel interested in utilizing and developing new fuels and bioproducts derived from cane crops.