Search Results

The possibility to operate current diesel engines in dual-fuel mode with the addition of an alternative fuel is fundamental to accelerate the energy transition to achieve carbon neutrality. The simulation of the dual- fuel combustion process with 0D/1D combustion models is fundamental for the performance prediction, but still particularly challenging, due to chemical interactions of the mixture. The authors defined a novel data-driven workflow for the development of combustion reaction mechanisms and used it to generate a dual-fuel mechanism for Ammonia and Diesel Primary Reference Fuels (DPRF) suitable for efficient combustion simulations in heavy duty engines, with variable cetane number Diesel fuels. A baseline reaction mechanism was created by merging the detailed ammonia mechanism by Glarborg et al. with reaction pathways for n- hexadecane and 2,2,4,4,6,8,8-heptamethylnonane from a well-established multi-component fuel mechanism.

We propose a novel dual-fuel combustion model for simulating heavy-duty engines with the G-Equation. Dual-Fuel combustion strategies in such engines features direct injection of a high-reactivity fuel into a lean, premixed chamber which has a high resistance to autoignition. Distinct combustion modes are present: the DI fuel auto-ignites following chemical ignition delay after spray vaporization and mixing; a reactive front is formed on its surroundings; it develops into a well-structured turbulent flame, which propagates within the premixed charge. Either direct chemistry or the flame-propagation approach (G- Equation), taken alone, do not produce accurate results. The proposed Dual-Fuel model decides what regions of the combustion chamber should be simulated with either approach, according to the local flame state; and acts as a “kernel” model for the G- Equation model. Direct chemistry is run in the regions where a premixed front is not present.

Thermal cabin comfort is the largest consumer of battery energy second only to propulsion in Battery Electric Vehicles (BEV’s). Accurate prediction of thermal comfort in the vehicle cabin with fast turnaround times will allow engineers to study the impact of various thermal comfort technologies and develop energy efficient Heating, Ventilation and Air Conditioning (HVAC) systems. In this study a novel data-driven model based on physics-guided Sparse Identification of Nonlinear Dynamics (SINDy) method was developed to predict Equivalent Homogeneous Temperature (EHT), Mean Radiant Temperature (MRT) and cabin air temperature under transient conditions and drive cycles. EHT is a recognized measure of the total heat loss from the human body that can be used to characterize highly non-uniform thermal environments such as a vehicle cabin. The SINDy model was trained on drive cycle data from Climatic Wind Tunnel (CWT) for a representative Battery Electric Vehicle.

Spray-wall interactions in diesel engines have a strong influence on turbulent flow evolution and mixing, which influences the engine’s thermal efficiency and pollutant-emissions behavior. Previous optical experiments and numerical investigations of a stepped-lip diesel piston bowl focused on how spray-wall interactions influence the formation of squish-region vortices and their sensitivity to injection timing. Such vortices are stronger and longer-lived at retarded injection timings and are correlated with faster late-cycle heat release and soot reductions, but are weaker and shorter-lived as injection timing is advanced. Computational fluid dynamics (CFD) simulations predict that piston bowls with more space in the squish region can enhance the strength of these vortices at near-TDC injection timings, which is hypothesized to further improve peak thermal efficiency and reduce emissions. The dimpled stepped-lip (DSL) piston is such a design.

G-Equation models represent propagating flame fronts with an implicit two-dimensional surface representation (level-set). Level-set methods are fast, as transport source terms for the implicit surface can be solved with finite-volume operators on the finite-volume domain, without having to build the actual surface. However, they include approximations whose practical effects are not properly understood. In this study, we improved the numerics of the FRESCO CFD code’s G-Equation solver and developed a new method to simulate kernel growth using signed distance functions and the analytical sphere-mesh overlap. We analyzed their role for simulating propane/air flames, using three well-established constant-volume configurations: a one-dimensional, freely propagating laminar flame; a disc-shaped, constant-volume swirl combustor; and torch-jet flame development through an orifice from a two-chamber device.

A one-dimensional, non-equilibrium, compressible law of the wall model is proposed to increase the accuracy of heat transfer predictions from computational fluid dynamics (CFD) simulations of internal combustion engine flows on engineering grids. Our 1D model solves the transient turbulent Navier-Stokes equations for mass, momentum, energy and turbulence under the thin-layer assumption, using a finite-difference spatial scheme and a high-order implicit time integration method. A new algebraic eddy-viscosity closure, derived from the Han-Reitz equilibrium law of the wall, with enhanced Prandtl number sensitivity and compressibility effects, was developed for optimal performance. Several eddy viscosity sub-models were tested for turbulence closure, including the two-equation k-epsilon and k-omega, which gave insufficient performance.

Appropriate spray modeling in multidimensional simulations of diesel engines is well known to affect the overall accuracy of the results. More and more accurate models are being developed to deal with drop dynamics, breakup, collisions, and vaporization/multiphase processes; the latter ones being the most computationally demanding. In fact, in parallel calculations, the droplets occupy a physical region of the in-cylinder domain, which is generally very different than the topology-driven finite-volume mesh decomposition. This makes the CPU decomposition of the spray cloud severely uneven when many CPUs are employed, yielding poor parallel performance of the spray computation. Furthermore, mesh-independent models such as collision calculations require checking of each possible droplet pair, which leads to a practically intractable O(np2/2) computational cost, np being the total number of droplets in the spray cloud, and additional overhead for parallel communications.

Analysis-driven pre-calibration of a modern automotive engine is extremely valuable in significantly reducing hardware investments and accelerating engine designs compliant with stricter emission regulations. Advanced modelling tools, such as a Virtual Engine Model (VEM) using Computational Fluid Dynamics (CFD), are often used within the framework of a Design of Experiments for Powertrain Engineering (DEPE) with the goal of streamlining significant portions of the calibration process. The success of the methodology largely relies on the accuracy of analytical predictions, especially engine-out emissions. Results show excellent agreements in engine performance parameters (with R2 > 98%) and good agreements in NOx and combustion noise (with R2 > 87%), while the Carbon Monoxide (CO), Unburned Hydrocarbons (HC) and Smoke emissions predictions remain a challenge even with a large n-heptane mechanism consisting of 144 species and 900 reactions and refined mesh resolution.

A Machine Learning (ML) approach is presented to correlate in-cylinder images of early flame kernel development within a spark-ignited (SI) gasoline engine to early-, mid-, and late-stage flame propagation. The objective of this study was to train machine learning models to analyze the relevance of flame surface features on subsequent burn rates. Ultimately, an approach of this nature can be generalized to flame images from a variety of sources. The prediction of combustion phasing was formulated as a regression problem to train predictive models to supplement observations of early flame kernel growth. High-speed images were captured from an optically accessible SI engine for 357 cycles under pre-mixed operation. A subset of these images was used to train three models: a linear regression model, a deep Convolutional Neural Network (CNN) based on the InceptionV3 architecture and a CNN built with assisted learning on the VGG19 architecture.

Lagrangian spray modeling represents a critical boundary condition for multidimensional simulations of in-cylinder flow structure, mixture formation and combustion in internal combustion engines. Segregated models for injection, breakup, collision and vaporization are usually employed to pass appropriate momentum, mass, and energy source terms to the gas-phase solver. Careful calibration of each sub-model generally produces appropriate results. Yet, the predictiveness of this modeling approach has been questioned by recent experimental observations, which showed that at trans- and super-critical conditions relevant to diesel injection, classical atomization and vaporization behavior is replaced by a mixing-controlled phase transition process of a dense fluid. In this work, we assessed the shortcomings of classical spray modeling with respect to real-gas and phase-change behavior, employing a multicomponent phase equilibrium solver and liquid-jet theory.

In this work we studied the effects of piston bowl design on combustion in a small-bore direct-injection diesel engine. Two bowl designs were compared: a conventional, omega-shaped bowl and a stepped-lip piston bowl. Experiments were carried out in the Sandia single-cylinder optical engine facility, with a medium-load, mild-boosted operating condition featuring a pilot+main injection strategy. CFD simulations were carried out with the FRESCO platform featuring full-geometric body-fitted mesh modeling of the engine and were validated against measured in-cylinder performance as well as soot natural luminosity images. Differences in combustion development were studied using the simulation results, and sensitivities to in-cylinder flow field (swirl ratio) and injection rate parameters were also analyzed.

Sector mesh modeling is the dominant computational approach for combustion system design optimization. The aim of this work is to quantify the errors descending from the sector mesh approach through three geometric modeling approaches to an optical diesel engine. A full engine geometry mesh is created, including valves and intake and exhaust ports and runners, and a full-cycle flow simulation is performed until fired TDC. Next, an axisymmetric sector cylinder mesh is initialized with homogeneous bulk in-cylinder initial conditions initialized from the full-cycle simulation. Finally, a 360-degree azimuthal mesh of the cylinder is initialized with flow and thermodynamics fields at IVC mapped from the full engine geometry using a conservative interpolation approach. A study of the in-cylinder flow features until TDC showed that the geometric features on the cylinder head (valve tilt and protrusion into the combustion chamber, valve recesses) have a large impact on flow complexity.

Powertrain electrification could be a key enabler for compliance with future exhaust emission standards and carbon dioxide (CO2) emissions limits or a customer facing product differentiator. The main objective of this study was to assess the potential of electrified propulsion systems in achieving a substantial reduction in CO2 emissions when applied to a representative full-size heavy-duty (HD) truck compared to the baseline configuration. A representative full-size HD four-wheel drive (4WD) truck of adjusted loaded vehicle weight (ALVW) 4082 kg or 9000 lbs with a 6.6 L diesel engine was simulated with various electrified drive configurations over the combined US FTP-72 (Federal Test Procedure) cycle and the Highway Fuel Economy Test (HWFET). Every hybrid vehicle configuration used in the study was designed using representative battery pack and electric drive components.

Diesel piston bowl geometry can affect turbulent mixing and therefore it impacts heat-release rates, thermal efficiency, and soot emissions. The focus of this work is on the effects of bowl geometry and injection timing on turbulent flow structure. This computational study compares engine behavior with two pistons representing competing approaches to combustion chamber design: a conventional, re-entrant piston bowl and a stepped-lip piston bowl. Three-dimensional computational fluid dynamics (CFD) simulations are performed for a part-load, conventional diesel combustion operating point with a pilot-main injection strategy under non-combusting conditions. Two injection timings are simulated based on experimental findings: an injection timing for which the stepped-lip piston enables significant efficiency and emissions benefits, and an injection timing with diminished benefits compared to the conventional, re-entrant piston.

In light-duty direct-injection (DI) diesel engines, combustion chamber geometry influences the complex interactions between swirl and squish flows, spray-wall interactions, as well as late-cycle mixing. Because of these interactions, piston bowl geometry significantly affects fuel efficiency and emissions behavior. However, due to lack of reliable in-cylinder measurements, the mechanisms responsible for piston-induced changes in engine behavior are not well understood. Non-intrusive, in situ optical measurement techniques are necessary to provide a deeper understanding of the piston geometry effect on in-cylinder processes and to assist in the development of predictive engine simulation models. This study compares two substantially different piston bowls with geometries representative of existing technology: a conventional re-entrant bowl and a stepped-lip bowl. Both pistons are tested in a single-cylinder optical diesel engine under identical boundary conditions.

A swept-volume method of calculating the volume swept by the flame during each time step is developed and used to improve the calculation of fuel reaction rates. The improved reaction rates have been applied to the ignition model and coupled with the level set G-equation combustion model. In the ignition model, a single initial kernel is formed after which the kernel is convected by the gas flow and its growth rate is determined by the flame speed and thermal expansion due to the energy transfer from the electrical circuit. The predicted ignition kernel size was compared with the available experimental data and good agreements were achieved. Once the ignition kernel reaches a size when the fully turbulent flame is developed, the G-equation model is switched on to track the mean turbulent flame front propagation.

In light- and medium-duty diesel engines, piston bowl shape influences thermal efficiency, either due to changes in wall heat loss or to changes in the heat release rate. The relative contributions of these two factors are not clearly described in the literature. In this work, two production piston bowls are adapted for use in a single cylinder research engine: a conventional, re-entrant piston, and a stepped-lip piston. An injection timing sweep is performed at constant load with each piston, and heat release analyses provide information about thermal efficiency, wall heat loss, and the degree of constant volume combustion. Zero-dimensional thermodynamic simulations provide further insight and support for the experimental results. The effect of bowl geometry on wall heat loss depends on injection timing, but changes in wall heat loss cannot explain changes in efficiency.

Hybridization is a mainstream technology for automobiles, and its application is rapidly expanding in other fields. Marine propulsion is one such field that could benefit from electrification of the powertrain. In particular, for boats to sail in enclosed waterways, such as harbors, channels, lagoons, a pure electric mode would be highly desirable. The main challenge to accomplish hybridization is the additional weight of the electric components, in particular the batteries. The goal of this project is to replace a conventional 4-stroke turbocharged Diesel engine with a hybrid powertrain, without any penalty in terms of weight, overall dimensions, fuel efficiency, and pollutant emissions. This can be achieved by developing a new generation of 2-Stroke Diesel engines, and coupling them to a state-of-the art electric system. For the thermal units, two alternative designs without active valve train are considered: opposed piston and loop scavenged engines.

Modeling ignition kernel development in spark ignition engines is crucial to capturing the sources of cyclic variability, both with RANS and LES simulations. Appropriate kernel modeling must ensure that energy transfer from the electrodes to the gas phase has the correct timing, rate and locations, until the flame surface is large enough to be represented on the mesh by the G-Equation level-set method. However, in most kernel models, geometric details driving kernel growth are missing: either because it is described as Lagrangian particles, or because its development is simplified, i.e., down to multiple spherical flames. This paper covers the geometric aspects of kernel development, which makes up the core of a Triangulated Lagrangian Ignition Kernel model. One (or multiple, if it restrikes) spark channel is initialized as a one-dimensional Lagrangian particle thread.

This work reports a CFD study on a 2-stroke (2-S) opposed piston high speed direct injection (HSDI) Diesel engine. The engine main features (bore, stroke, port timings, et cetera) are defined in a previous stage of the project, while the current analysis is focused on the assembly made up of scavenge ports, manifold and cylinder. The first step of the study consists in the construction of a parametric mesh on a simplified geometry. Two geometric parameters and three different operating conditions are considered. A CFD-3D simulation by using a customized version of the KIVA-4 code is performed on a set of 243 different cases, sweeping all the most interesting combinations of geometric parameters and operating conditions. The post-processing of this huge amount of data allow us to define the most effective geometric configuration, named baseline.