Evaluation of the approximated diffusion flamelet concept using fuels with different chemical complexity

The ability of flamelet models to reproduce turbulent combustion in devices such as diesel engines or gas turbines has enhanced the usage of these approaches in Computational Fluid Dynamics (CFD) simulations. The models based on turbulent look-up tables generated from counterflow laminar diffusion flames (DF model) permit drastic reduction of the computational cost of the CFD calculation. Nevertheless, for complex molecular fuels, such as n-heptane, the oxidation process involves hundreds of species and the calculation of the transport equations together with the ODE system that models the chemical kinetics for the DF solution becomes unaffordable for industrial devices where hundreds of flamelets are required. In this context, new hypotheses have to be introduced in order to reduce the computational cost maintaining the coherence of the combustion process. Recently, a new model known as Approximated Diffusion Flamelet (ADF) has been proposed with the aim of solving the turbulent combustion for complex fuels in a reduced time. However, the validity of this model is still an open question and has to be verified in order to justify subsequent CFD calculations. This work assesses the ADF model and its ability to reproduce accurately the combustion process and its main parameters for three fuels with different chemical complexity and boundary conditions by its comparison with the DF model. Results show that although some discrepancies arise, the ADF model has the ability to correctly describe the ignition delay and the combustion structure in the auto-ignition zone that is the most relevant one for industrial processes.     

Mathematical programming time-based decomposition algorithm for discrete event simulation

Mathematical programming has been proposed in the literature as an alternative technique to simulating a special class of Discrete Event Systems. There are several benefits to using mathematical programs for simulation, such as the possibility of performing sensitivity analysis and the ease of better integrating the simulation and optimisation. However, applications are limited by the usually long computational times. This paper proposes a time-based decomposition algorithm that splits the mathematical programming model into a number of submodels that can be solved sequentially to make the mathematical programming approach viable for long running simulations. The number of required submodels is the solution of an optimisation problem that minimises the expected time for solving all of the submodels. In this way, the solution time becomes a linear function of the number of simulated entities.     

Mathematical modeling and validation of mass transfer phenomenon in homogeneous charge compression ignition engines based on a thermodynamic multi zone model

The main purpose of the current study is mathematical modelling and validation of mass transfer phenomenon in homogeneous charge compression ignition engines. A validated multi-zone model coupled to a semi-detailed chemical kinetics is used to predict homogeneous charge compression ignition combustion and emissions. Heat and Mass transfer submodels are linked to the multi-zone model. Bulk flow and diffusion mass transfer between zones are considered. The results indicate that the diffusion mass transfer is negligible in homogeneous charge compression ignition engines. Bulk flow mass transfer plays a critical role in homogeneous charge compression ignition simulation and applying it in the multi-zone model leads to accurate prediction of the start of combustion, peak pressure and exhaust emissions. The results show that the maximum error changes from 90% to 5% in carbon monoxide prediction and from 98% to 14% in unburned hydrocarbons prediction, using the mass transfer submodel.     

Embedding optimization in computational science workflows

Workflows support the automation of scientific processes, providing mechanisms that underpin modern computational science. They facilitate access to remote instruments, databases and parallel and distributed computers. Importantly, they allow software pipelines that perform multiple complex simulations (leveraging distributed platforms), with one simulation driving another. Such an environment is ideal for computational science experiments that require the evaluation of a range of different scenarios “in silico” in an attempt to find ones that optimize a particular outcome. However, in general, existing workflow tools do not incorporate optimization algorithms, and thus whilst users can specify simulation pipelines, they need to invoke the workflow as a stand-alone computation within an external optimization tool. Moreover, many existing workflow engines do not leverage parallel and distributed computers, making them unsuitable for executing computational science simulations. To solve this problem, we have developed a methodology for integrating optimization algorithms directly into workflows. We implement a range of generic actors for an existing workflow system called Kepler, and discuss how they can be combined in flexible ways to support various different design strategies. We illustrate the system by applying it to an existing bio-engineering design problem running on a Grid of distributed clusters.     

Representing knowledge about linear programming formulation

This paper describes a system, LPFORM, that enables users to design linear programming models interactively, using graphics. An output of the system is an algebraic statement of the model and data references that are subsequently used to generate the input for a solver in the standard MPS format. The emphasis of this paper is on the types of knowledge one has on the submodels that make up larger models and how this knowledge can be organized.     

Experiments in numerical methods for a problem in combustion modeling

In this paper, we have shown that the numerical method of lines can be used effectively to solve time dependent combustion models in one spatial dimension. By the numerical method of lines (NMOL), we mean the reduction of a system of partial differential equations to a system of ordinary differential equations (ODE's), followed by the solution of this ODE system with an appropriate ODE solver. We used finite differences for the spatial discretization and a variant of the GEAR package for the ODE's.We have presented various solution methods of interest for the nonlinear algebraic system in this setting; that is, in the corrector iteration section of the GEAR package applied to combustion models. These methods include Newton/block SOR (SOR denotes successive over-relaxation), block SOR/Newton, Newton/block-diagonal Jacobian, Newton/kinetics-only Jacobian, and Newton/block symmetric SOR. These methods have in common their lack of frequent use in ODE software and their eady applicability to partial differential equations in more than one spatial dimension.Finally, we have given the results of numerical tests, run on the CDC-7600 and Cray-1 computers. By so doing, we indicate the more promising nonlinear system solvers for the NMOL solution of combustion models.     

A domain-decomposing parallel sparse linear system solver

The solution of large sparse linear systems is often the most time-consuming part of many science and engineering applications. Computational fluid dynamics, circuit simulation, power network analysis, and material science are just a few examples of the application areas in which large sparse linear systems need to be solved effectively. In this paper, we introduce a new parallel hybrid sparse linear system solver for distributed memory architectures that contains both direct and iterative components. We show that by using our solver one can alleviate the drawbacks of direct and iterative solvers, achieving better scalability than with direct solvers and more robustness than with classical preconditioned iterative solvers. Comparisons to well-known direct and iterative solvers on a parallel architecture are provided.     

A so-called Cluster Benders Decomposition approach for solving two-stage stochastic linear problems

The optimization of stochastic linear problems, via scenario analysis, based on Benders decomposition requires appending feasibility and/or optimality cuts to the master problem until the iterative procedure reaches the optimal solution. The cuts are identified by solving the auxiliary submodels attached to the scenarios. In this work, we propose the algorithm named scenario Cluster Benders Decomposition (CBD) for dealing with the feasibility cut identification in the Benders method for solving large-scale two-stage stochastic linear problems. The scenario tree is decomposed into a set of scenario clusters and tighter feasibility cuts are obtained by solving the auxiliary submodel for each cluster instead of each individual scenario. Then, the scenario cluster based scheme allows to identify tighter feasibility cuts that yield feasible second stage decisions in reasonable computing time. Some computational experience is reported by using CPLEX as the solver of choice for the auxiliary LP submodels at each iteration of the algorithm CBD. The results that are reported show the favorable performance of the new approach over the traditional single scenario based Benders decomposition; it also outperforms the plain use of CPLEX for medium-large and large size instances.     

A domain-decomposing parallel sparse linear system solver

The solution of large sparse linear systems is often the most time-consuming part of many science and engineering applications. Computational fluid dynamics, circuit simulation, power network analysis, and material science are just a few examples of the application areas in which large sparse linear systems need to be solved effectively. In this paper, we introduce a new parallel hybrid sparse linear system solver for distributed memory architectures that contains both direct and iterative components. We show that by using our solver one can alleviate the drawbacks of direct and iterative solvers, achieving better scalability than with direct solvers and more robustness than with classical preconditioned iterative solvers. Comparisons to well-known direct and iterative solvers on a parallel architecture are provided.     

Numerical Studies of Viscoelastic Flow Using the Software OpenFOAM

Viscoelastic fluids are a special class of non-Newtonian fluids. There are several types of viscoelastic fluid models, and all of them have a complex rheological response in comparison to Newtonian fluids. This response can be viewed as a combination of viscous and elastic effects and non-linear phenomena. This complex physics makes a numerical simulation a rather challenging task, even in simple test-cases. Studies presented in this paper are numerical studies of the viscoelastic fluid flow in several test cases. These studies have been done in OpenFOAM, an open-source CFD package. Implementation of viscoelastic models and a solver is only available in a community driven version of software (OpenFOAM-ext). One of the goals of research in this paper was to test the solver and models on some simple test cases. We considered start-up and pulsating flows of viscoelastic fluid in a channel and a circular pipe. The important thing is that an analytical solution can be found in these cases, making in possible to test all aspects of numerical simulation in OpenFOAM. Obtained results showed an excellent agreement with the analytical solution for both velocity and stress components. These results encouraged authors' motivation and a choice to use OpenFOAM for simulation of viscoelastic flows. We hope that our research will make a contribution to the OpenFOAM community. Our plan for the further research is a simulation of blood flow in arteries with the viscoelastic solver. (© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Elementary submodels of parametrizable models

We introduce the notion of F-parametrizable model and prove some general results on elementary submodels of F-parametrizable models. Using this notion, we can uniformly characterize all elementary submodels for the field of real numbers and for the group of all permutations on natural numbers in the first order language as well as in the language of hereditarily finite superstructures. Assuming the constructibility axiom, we obtain a simpler characterization of elementary submodels of F-parametrizable models and prove some additional properties of the structure of their elementary submodels.     

Model order reduction by projection applied to the universal Reynolds equation

The approach presented in this paper yields a reduced order solution to the universal Reynolds equation for incompressible fluids, which is valid in lubrication as well as in cavitation regions, applied to oil-film lubricated journal bearings in internal combustion engines. The extent of cavitation region poses a free boundary condition to the problem and is determined by an iterative spatial evaluation of a superposed modal solution. Using a Condensed Galerkin and Petrov–Galerkin method, the number of degrees of freedom of the original grid is reduced to obtain a fast but still accurate short-term prediction of the solution. Based on the assumption that a detailed solution of a previous combustion cycle is available, a basis and an optimal test space for the Galerkin method is generated. The resulting reduced order model is efficiently exploited in a time-saving evaluation of the Jacobian matrix describing the elastohydrodynamic coupling in a multi-body dynamics simulation using flexible components. Finally, numerical results are presented for a single crankshaft main bearing of typical dimensions.     

A compressible two-phase flow framework for Large Eddy Simulations of liquid-propellant rocket engines

Atomization of liquid jets is a key feature of many propulsion systems, such as jet engines, internal combustion engines or liquid-propellant rocket engines (LRE). As it controls the characteristics of the spray, atomization has a great influence on the complex interaction between phenomena such as evaporation, turbulence, acoustics and combustion. In this context, Computational Fluid Dynamics is a promising way to bring better understanding of dynamic phenomena involving atomization, such as e.g. high-frequency combustion instabilities in LRE. However the unsteady simulation of primary atomization in reactive compressible two-phase flows is very challenging, due to the variety of the spatial and temporal scales, as well as to the high density, velocity and temperature gradients which require robust and efficient numerical methods. To address this issue, a numerical strategy is proposed in this paper, which is able to describe the dynamics of the whole chain of mechanisms from the liquid injection to its atomization and combustion. Primary atomization is modeled by a coupling between a homogeneous diffuse interface model and a kinetic-based Eulerian model for the spray. This strategy is successfully applied to the unsteady simulation of an operating point of the Onera’s Mascotte test bench, representative of one coaxial injector of LRE operating under subcritical conditions. The dynamics of the liquid core is retrieved and the flame shape as well as Sauter mean diameters are in good agreement with experimental results. These results demonstrate the ability of the strategy to deal with the harsh conditions of cryogenic combustion, and provide a promising framework for future studies of combustion instabilities in LRE.     

Investigating communication and step-size behaviour for co-simulation of hybrid physical systems

This paper discusses the method of cooperative simulation of discrete and continuous models with the Building Controls Virtual Test Bed, a software environment that allows coupling different simulation programs. In the course of a project aiming the energy optimization in cutting factories, models of machines of differing complexity and a building containing them have to be implemented to further simulate the thermal processes. Since all partial models require individual modelling approaches, solver time steps, solvers or even simulators, the method of co-simulation is considered. The partial models will be implemented with Modelica, MATLAB, Simulink and Simscape and accessed with the co-simulation tool BCVTB. The simulation results show that this method of co-simulation can be sufficient for the needs of describing thermal systems with large time constants but has to be found insufficient for simulations requiring high accuracy and variable step solvers in the overall simulation.     

Implementation of EOQ and Lambert W function in 1-D engine simulation model for optimizing fuel injection in GDI engine

The aim of this study was to implement Economic Order Quantity method (EOQ) together with the Lambert W function in a 1-D engine simulation model in order to develop a fuel control strategy for a Gasoline direct injection (GDI) engine. Previous work of the co-author demonstrated the possibility of optimizing fuel injection quantity in GDI engine using the EOQ that is commonly used in supply chain of perishable products. This work extends the previous work and implements it in a 1-D, crank angle resolved, engine simulation model for the application of model based calibration process. The present work uses a validated engine simulation model, which is based on predictive combustion modelling approach, and couples the 1-D engine simulation model with SIMULINK to add the evaporation, wall- wetting and heat transfer models. It employs FORTRAN subroutines to modify the internal code of the 1-D simulation software in order to add crank angle resolved evaporation model. Finally, EOQ with Lambert W function was added to the model using MATLAB with special attention to the decimal control for the solution. This study demonstrated that EOQ and Lambert W functions together are a suitable method to develop fuel control strategy for a model based calibration procedure when implemented in crank angle resolved 1-D simulation model.     

GasTurbnLab: a multidisciplinary problem solving environment for gas turbine engine design on a network of nonhomogeneous machines

Gas turbine engines are very complex (with 20–40,000 parts) and have extreme operating conditions. The important physical phenomena take place on scales from 10–100 microns to meters. A complete and accurate dynamic simulation of an entire engine is enormously demanding. Designing a complex system, like a gas turbine engine, will require fast, accurate simulations of computational models from multiple engineering disciplines along with sophisticated optimization techniques to help guide the design process. In this paper, we describe the architecture of an agent-based software framework for the simulation of various aspects of a gas turbine engine, utilizing a “network” of collaborating numerical objects through a set of interfaces among the engine parts. Moreover, we present its implementation using the Grasshopper agent middleware and provide simulation results that show the feasibility of the computational paradigm implemented.     

Application of Enthalpy Deficit Flamelet Model in Spray Combustion Simulation北大核心CSCD

An OpenFOAM-based solver for spray combustion simulation with the large eddy simulation (LES) and the flamelet generated manifold (FGM) method, was developed. A simple reduction of the temperature was employed to account for the evaporative heat loss. The solver was firstly validated against the Sydney piloted ethanol spray flame benchmark EtF7. The predicted mean gas temperature and droplet statistics correspond well with the experimental data and have similar accuracy to the spray flamelet model. The turbulence-chemistry interaction modeling may have a larger influence on the simulation accuracy. Then a realistic gas turbine slinger combustor was simulated with 2 sets of operating conditions. The simulation results reveal different flame characteristics of the 2 working conditions. The predicted total pressure losses are close to the measured values. © 2023 Editorial Office of Applied Mathematics and Mechanics. All rights reserved.     

Modelling redundancy allocation for a fuzzy random parallel-series system

Due to subjective judgment, imprecise human knowledge and perception in capturing statistical data, the real data of lifetimes in many systems are both random and fuzzy in nature. Based on the fuzzy random variables that are used to characterize the lifetimes, this paper studies the redundancy allocation problems to a fuzzy random parallel-series system.Two fuzzy random redundancy allocation models (FR-RAM) are developed through reliability maximization and cost minimization, respectively. Some properties of the FR-RAM are obtained, in which an analytical formula of reliability with convex lifetimes is derived and the sensitivity of the reliability is discussed. To solve the FR-RAMs, we first address the computation of reliability. A random simulation method based on the derived analytical formula is proposed to compute the reliability with convex lifetimes. As for the reliability with nonconvex lifetimes, the technique of fuzzy random simulation together with the discretization method of fuzzy random variable is employed to compute the reliability, and a convergence theorem of the fuzzy random simulation is proved. Subsequently, we integrate the computation approaches of the reliability and genetic algorithm (GA) to search for the approximately optimal redundancy allocation of the models. Finally, some numerical examples are provided to illustrate the feasibility of the solution algorithm and quantify its effectiveness.     

Flexible composition and execution of large scale applications on distributed e-infrastructures

Computer simulation is finding a role in an increasing number of scientific disciplines, concomitant with the rise in available computing power. Marshalling this power facilitates new, more effective and different research than has been hitherto possible. Realizing this inevitably requires access to computational power beyond the desktop, making use of clusters, supercomputers, data repositories, networks and distributed aggregations of these resources. The use of diverse e-infrastructure brings with it the ability to perform distributed multiscale simulations. Accessing one such resource entails a number of usability and security problems; when multiple geographically distributed resources are involved, the difficulty is compounded. In this paper we present a solution, the Application Hosting Environment,³ which provides a Software as a Service layer on top of distributed e-infrastructure resources. We describe the performance and usability enhancements present in AHE version 3, and show how these have led to a high performance, easy to use gateway for computational scientists working in diverse application domains, from computational physics and chemistry, materials science to biology and biomedicine.