Search Results

Carbon dioxide (CO2 or R744) is a promising next-generation refrigerant for mobile air-conditioning applications (MAC), which has the advantages of good heating performance in cold climates and environmental-friendly properties. This paper presents a simulation model of an integrated internal heat exchanger (IHX) and accumulator (Acc) using the finite volume method. The results are validated by a group of experimental data collected with different transcritical R744 mobile air-conditioner and heat pump (MHP) systems, and the error was within ±10%. The impacts of refrigerant mass flow rate and operating temperatures on the heat transfer rate of the IHX, improvement on refrigeration capacity and the liquid level in the Acc were studied. Results show that the net benefits of IHX are significant in AC mode, while it helps preventing flooding of the compressor in MHP mode.

A geometrically accurate, three-dimensional finite element model of a Diesel engine exhaust valve and cylinder head assembly has been developed to analyze the effect of cylinder head interactions on exhaust valve stresses. Results indicate that a multi-lobed stress pattern occurs around the exhaust valve head due to cylinder head deformation, stiffness variations, and thermal asymmetry. Consequently, peak valve bending and hoop stresses from the three-dimensional model are 48% and 40% higher, respectively, than for the two-dimensional, axisymmetric model. These results indicate the degree of model complexity required for more accurate analyses of exhaust valve operating stresses.

This paper describes a ‘toolbox’ for modeling liquid cooling system networks within vehicle thermal management systems. Components which can be represented include pumps, coolant lines, control valves, heat sources and heat sinks, liquid-to-air and liquid-to-refrigerant heat exchangers, and expansion tanks. Network definition is accomplished through a graphical user interface, allowing system architecture to be easily modified. The elements of the toolbox are physically based, so that the models can be applied before hardware is procured. The component library was coded directly into MATLAB / SIMULINK and is intended for control system development, hardware-in-the-loop (HIL) simulation, and as a system emulator for on-board diagnostics and controls purposes. For HIL simulation and on-board diagnostics and controls, it is imperative that the model run in real-time.

A numerical study of micro-explosion in multi-component droplets is presented. The homogeneous nucleation theory is used in describing the bubble generation process. A modified Rayleigh equation is then used to calculate the bubble growth rate. The breakup criterion is then determined by applying a linear stability analysis on the bubble-droplet system. After the explosion/breakup, the atomization characteristics, including Sauter mean radius and averaged velocity of the secondary droplets, are calculated from conservation equations. Micro-explosion can be enhanced by introducing biodiesel into the fuel blends of ethanol and tetradecane. Micro-explosion is more likely to occur at high ambient pressure. However, increasing the ambient temperature does not have a significant effect on micro-explosion. There exists an optimal composition in the liquid mixture for micro-explosion.

An experimental study was performed at the University of Illinois at Urbana-Champaign Low-Speed Wind Tunnel to quantify the performance and flowfield effects of two-element open-wheel-race-car front wing configurations. Four distinct configurations were tested in- and out-of-ground effect and at various speeds (Reynolds numbers), angles of attack, and flap positions. A splitter plate was installed in the wind tunnel to act as the ground plane. Data presented include balance force measurements, surface pressure data, and downstream flow measurements using a seven-hole probe. Results show that these elementary factors in the design of race-car front wings have a significant effect on wing performance and behavior of the downstream flowfield.

The purpose this paper is to delineate why there exists a trade-off between biomechanical realism and algorithmic efficiency for human motion simulation models, and to illustrate how empirical human movement data and findings can be integrated with novel modeling techniques to overcome such a realism-efficiency tradeoff. We first review three major classes of biomechanical models for human motion simulation. The review of these models is woven together by a common fundamental problem of redundancy—kinematic and/or muscle redundancy. We describe how this problem is resolved in each class of models, and unveil how the trade-off arises, that is, how the computational demand associated with solving the problem is amplified as a model evolves from small scale to large scale, or from less realism to more realism.

Low aspect ratio wings are used extensively on open-wheeled race cars to generate aerodynamic downforce. Consequently, a great deal of effort is invested in obtaining wing profiles that provide high values of lift coefficient. If the wings are designed using 2-D methods, then it is necessary to take into account the change in operating angle of a typical airfoil section that occurs when it operates in the downwash generated by the wing. Accounting for this change during the design phase will ensure that the airfoil sections are optimized for their intended operating conditions. The addition of endplates to the wing serves to counteract the magnitude of the change in operating angle by effectively producing an increase in wing aspect ratio. During the design process at UIUC, an empirical method was used to provide an estimate of the effective aspect ratio of the wing and endplate combination.

In this work, an efficient and unified combustion model is introduced to simulate the flame propagation, diffusion-controlled combustion, and chemically-driven ignition in both SI and CI engine operation. The unified model is constructed upon a G-equation model which addresses the premixed flame propagation. The concept of the Livengood-Wu integral is used with tabulated ignition delay data to account for the chemical kinetics which is responsible for the spontaneous ignition of fuel-air mixture. A set of rigorously defined operations are used to couple the evolution of the G scalar field and the Livengood-Wu integral. The diffusion-controlled combustion is simulated equivalent to applying the Burke-Schumann limit. The combined model is tested in the simulation of the premixed SI combustion in a constant volume chamber, as well as the CI combustion in a conventional small bore diesel engine.

Ignition location is one of the important factors that affect the thermal efficiency, exhaust emissions and knock sensitivity in premixed-charge ignition engines. However, the ignition initiation locations of pre-chamber generated turbulent jet ignition, which is a promising ignition enhancement method, are not clearly understood due to the complex physics behind it. Motivated by this, the ignition initiation location of a transient turbulent jet in a constant volume combustor is analyzed by the use of computational fluid dynamics (CFD) simulations. In the CFD simulations of this work, commercial codes KIVA-3 V release 2 and an in-house-developed chemical solver with a detailed mechanism for H2/air mixtures are used. Comparisons are performed between simulated and experimental ignition initiation locations, and they agree well with one another. A detailed parametric study of the influence of in-cylinder temperature on the ignition initiation location is also performed.

An efficient multi-component fuel droplet vaporization model has been developed in this work using discrete approach. The precise modeling of droplet vaporization process is divided into two parts: vapor-phase and liquid-phase sub-models. Temporal evolution of flow inside the droplet is considered to describe the transient behavior introduced by the slow diffusion process. In order to account for the internal circulation motion, surface regression and finite diffusion without actually resolving the spatial governing equations within the liquid phase, a set of ordinary differential equations is applied to describe the evolution of the non-uniform distributions of universal diffusional variables, i.e. temperature and species mass fraction. The differences between the droplet surface and bulk mean states are modeled by constructing a quasi-1D frame; the effect of the internal circulations is taken into consideration by using the effective diffusivity rather than physical diffusivity.

As engine researchers are facing the task of designing more powerful, more fuel efficient and less polluting engines, a large amount of research has been focused towards homogeneous charge compression ignition (HCCI) operation for diesel engines. Ignition timing of HCCI operation is controlled by a number of factors including intake temperatures, exhaust gas recirculation (EGR) and injection timing to name a few. This study focuses on the computational modeling of an optically accessible high-speed direct-injection (HSDI) small bore diesel engine. In order to capture the phenomena of HCCI operation, the KIVA computational code package has been outfitted with an improved and optimized Shell autoignition model, the extended Zeldovich thermal NOx model, and soot formation and oxidation models. With the above named models in place, several cases were computed and compared to experimentally measured data and captured images of the DIATA test engine.

This paper describes a numerical study on air/fuel preparation process in a direct-injected spark-ignition engine under partial load stratified conditions. The fuel is represented as a mixture of four components with a distillation curve similar to that of actual gasoline, and its vaporization processes are simulated by two recently formulated multicomponent vaporization models for droplet and film, respectively. The models include major mechanisms such as non-ideal behavior in high-pressure environments, preferential vaporization, internal circulation, surface regression, and finite diffusion in the liquid phase. A spray/wall impingement model with the effect of surface roughness is used to represent the interaction between the fuel spray and the solid wall. Computations of single droplet and film on a flat plate were first performed to study the impact of fuel representation and vaporization model on the droplet and film vaporization processes.

Cyclic loading is not as damaging as static loading of ceramics at high temperatures. Microcrack growth retardation has been established as a mechanism for increasing the durability of ceramics at high temperatures. A combined experimental and theoretical approach provides a mechanistic understanding of the deformation and failure processes in ceramic materials at high temperatures. Results demonstrate that the high temperature behavior of some ceramic material systems are controlled by the behavior of the grain boundary phase whose response is considerably different under static and cyclic loading.

Neural networks are massively parallel computational models for knowledge representation and information processing. The capabilities of neural networks, namely learning, noise tolerance, adaptivity, and parallel structure make them good candidates for application to a wide range of engineering problems including diagnostics problems. The general approach in developing neural network based diagnostic methods is described through a case study. The development of an acoustic wayside train inspection system using neural networks is described. The study is aimed at developing a neural network based method for detection defective wheels from acoustic measurements. The actual signals recorded when a train passes a wayside station are used to develop a neural network based wheel defect detector and to study its performance. Signal averaging and scoring techniques are developed to improve the performance of the constructed neural inspection system.

Induction hardening of mild steel components often results in significant improvements in the static and cyclic load capability, with comparatively small increases in cost. Members subjected primarily to torsional loading are a relevant subset of the broad range of induction hardened components. Due to the variation of material properties and residual stresses, failures are “initiated” at the traditional geometric locations predicted for homogeneous materials and also at subsurface sites. The introduction of shear based fatigue parameters has necessitated the consideration of the residual stress as a three dimensional quantity, especially when analyzing subsurface failures. Not considering the tensoral nature of the residual stress can lead to serious errors when estimating fatigue life, and for larger magnitude loadings, the residual stress field may relax.

A dynamic model for the springless electrohydraulic valvetrain has been developed. The model speeds up the valvetrain development process by simulating effects of parameter changes, thus minimizing the number of hardware variations. It includes dynamic characteristics of check valves that enable energy recovery, hydraulic snubbers that limit seating velocity of the engine valves, and leakage in the control solenoids. A good match of the experimental data has been obtained for a single valve system, and the model calibration and validation have been completed. The known parameters are used together with some unknown calibration constants which have been tuned to match the experimental data. The simulation results for a twin valve system are also presented. The model applications for system performance analysis and for the closed-loop control of the engine valve lift are described. The cyclic variability of the experimental data is also discussed.

Service loading histories have the same general character for an individual route and the magnitudes vary from driver to driver. Both the magnitude and character of the loading history change from route to route and a linear scaling of one loading history does not characterize the variability of usage over a wide range of operating conditions. In this paper a technique for measuring and extrapolating cumulative exceedance diagrams to quantify the distribution of service loading in a vehicle is described. Monte Carlo simulations are coupled with the local stress strain approach for fatigue to obtain distributions of service loading. Fatigue life estimates based on the original loading histories are compared to those obtained from statistical descriptions of exceedance diagrams.

The blow-by phenomenon is seldom acquainted with diesel engines, but for a small bore HSDI optical diesel engine, the effects are significant. A difference in peak pressure up to 25% can be observed near top-dead-center. To account for the pressure differences, a 0-D crevice flow model with a dynamic ring pack model was incorporated into the KIVA code to determine the amount of blow-by. The ring pack model will take into account the forces acting on the piston rings, the position of the piston rings, and the pressure located at each region of the crevice volume at every time step. The crevice flow model takes into consideration the flow through the circumferential gap, ring gap, and the ring side clearance. As a result, the cylinder mass, trapped mass in the crevice regions, and the blow-by values are known. Validation of the crevice model is accomplished by comparing the in-cylinder motoring pressure trace with the experimental motoring data.

High strength materials have desirable mechanical properties but often cannot be machined economically, which results in unacceptably high finished component cost. MADI™ (machinable austempered ductile iron) overcomes this difficultly and provides the highly desirable combination of high strength, excellent low temperature toughness, good machinability and attractive finished component cost. The Machine Tool Systems Research Laboratory at the University of Illinois at Urbana-Champaign performed extensive machinability testing and determined the appropriate tools, speeds and feeds for milling and drilling (https://netfiles.uiuc.edu/malkewcz/www/MADI.htm). This paper provides the information necessary for the efficient and economical machining of MADI™ and provides comparative machinability data for common grades of ductile iron (EN-GJS-400-18, 400-15, 450-10, 500-7, 600-3 & 700-2) for comparison.

A multicomponent fuel film vaporization model using continuous thermodynamics is developed for multidimensional spray and wall film modeling. The vaporization rate is evaluated using the turbulent boundary-layer assumption and a quasi-steady approximation. Third-order polynomials are used to model the fuel composition profiles and the temperature within the liquid phase in order to predict accurate surface properties that are important for evaluating the mass and moment vaporization rates and heat flux. By this approach, the governing equations for the film are reduced to a set of ordinary differential equations and thus offer a significant reduction in computational cost while maintaining adequate accuracy compared to solving the governing equations for the film directly.