Search Results

Multi-dimensional computational efforts using comprehensive and skeletal kinetics have been made to investigate the cool flame region in HCCI combustion. The work was done in parallel to an experimental study that showed the impact of the negative temperature coefficient and the cool flame on the start of combustion using different fuels, which is now the focus of the simulation work. Experiments in a single cylinder CFR research engine with n-butane and a primary reference fuel with an octane number of 70 (PRF 70) were modeled. A comparison of the pressure and heat release traces of the experimental and computational results shows the difficulties in predicting the heat release in the cool flame region. The behavior of the driving radicals for two-stage ignition is studied and is compared to the behavior for a single-ignition from the literature. Model results show that PRF 70 exhibits more pronounced cool flame heat release than n-butane.

A procedure has been developed to build system level predictive models that incorporate physical laws as well as information derived from experimental data. In particular a soot model was developed, trained and tested using experimental data. It was seen that the model could fit available experimental data given sufficient training time. Future accuracy on data points not encountered during training was estimated and seen to be good. The approach relies on the physical phenomena predicted by an existing system level phenomenological soot model coupled with ‘weights’ which use experimental data to adjust the predicted physical sub-model parameters to fit the data. This approach has developed from attempts at incorporating physical phenomena into neural networks for predicting emissions. Model training uses neural network training concepts.

Multiple injection strategies have been revealed as an efficient means to reduce diesel engine NOx and soot emissions simultaneously, while maintaining or improving its thermal efficiency. Empirical soot models widely adopted in engine simulations have not been adequately validated to predict soot formation with multiple injections. In this work, a multiple-step phenomenological (MSP) soot model that includes particle inception, surface growth, oxidation, and particle coagulation was revised to better describe the physical processes of soot formation in diesel combustion. It was found that the revised MSP model successfully reproduces measured soot emission dependence on the start-of-injection timing, while the two-step empirical and the original MSP soot models were less accurate. The revised MSP model also predicted reasonable soot and intermediate species spatial profiles within the combustion chamber.

Accurate and useful mathematical models of physical processes can be made when we understand all of the phenomena involved. This paper reviews our understanding of the fluid flow, heat transfer and thermodynamic processes occurring in engines and the status of mathematical models expressing this understanding. Thermodynamic single system rate models are found to be extremely useful in predicting power and fuel consumption performance but of limited value in predicting emission performance. Multiple-zone, nonequilibrium models are essential for predicting emissions but are limited in accuracy by computer capacity and our understanding of engine phenomena which vary rapidly both with space and time. The need for and ability of new types of instrumentation, primarily optical, to increase our understanding of engine phenomena and improve our models is discussed.

A modification of the computational fluid dynamics code KIVA-II is presented that allows computations to be made in complex engine geometries. An example application is given in which three versions of KIVA-II are run simultaneously. Each version considers a separate block of the computational domain, and the blocks exchange boundary condition information with each other at their common interfaces. The use of separate blocks permits the connectedness of the overall computational domain to change with time. The scavenging flow in the cylinder, transfer pipes (ports), and exhaust pipe of a ported two-stroke engine with a moving piston was modeled in this way. Results are presented for three engine designs that differ only in the angle of their boost ports. The calculated flow fields and the resulting fuel distributions are shown to be markedly different with the different geometries.

Three distinct types of diesel particulate matter (PM) are generated in selected engine operating conditions of a single-cylinder heavy-duty diesel engine. The three types of PM are trapped using typical Cordierite diesel particulate filters (DPF) with different washcoat formulations and a commercial Silicon-Carbide DPF. Two systems, an external electric furnace and an in-situ burner, were used for regeneration. Furnace regeneration experiments allow the collected PM to be classified into two categories depending on oxidation mechanism: PM that is affected by the catalyst and PM that is oxidized by a purely thermal mechanism. The two PM categories prove to contribute differently to pressure drop and transient filtration efficiency during in-situ regeneration.

An experimental study was performed to investigate diesel particulate filter (DPF) performance during filtration with the use of real-time measurement equipment. Three operating conditions of a single-cylinder 2.3-liter D.I. heavy-duty diesel engine were selected to generate distinct types of diesel particulate matter (PM) in terms of chemical composition, concentration, and size distribution. Four substrates, with a range of geometric and physical parameters, were studied to observe the effect on filtration characteristics. Real-time filtration performance indicators such as pressure drop and filtration efficiency were investigated using real-time PM size distribution and a mass analyzer. Types of filtration efficiency included: mass-based, number-based, and fractional (based on particle diameter). In addition, time integrated measurements were taken with a Rupprecht & Patashnick Tapered Element Oscillating Microbalance (TEOM), Teflon and quartz filters.

A two dimensional CFD model has been developed to study the mechanism of soot deposition on the porous wall surface in a Diesel Particulate Filter (DPF). The goal is to improve understanding of the soot deposition and its interaction with the hydrodynamic behaviour of the device. The KIVA3V CFD code and pre-processor were modified to simulate a single channel in a DPF. The code was extended to solve the conservation equations in porous media materials, to account for the sticking of particles on a porous surface and to evaluate the increasing resistance to the flow as the soot inside the trap accumulates. The code is already configured to track Lagrangian particles and these were modified to represent the soot particles in the flow. The code pre-processor was modified to allow definition of a double-symmetric geometry and to specify porous cells.

A Soot-NOx Trap (SNT) is a combinatorial aftertreatment device intended to decrease both particulate and NOx emissions simultaneously. A system-level Soot-NOx Trap model was developed by adding Lean NOx Trap kinetics to a 1D Diesel Particulate Filter model. The hybrid model was validated against each parent model for the limiting cases, then exercised to investigate the interacting redox behavior. Modulations in temperature and exhaust air-fuel ratio were investigated for their ability to facilitate particulate oxidation and NOx reduction in the trap.

The development of the Diesel Exhaust Filtration Analysis system (DEFA), which utilizes a rectangular wafer of the same substrate material as used in a full-scale Diesel Particulate Filter (DPF), is presented in this paper. Washcoat variations of the wafer substrate (bare, washcoat, and catalyzed washcoat) were available for testing. With this setup, the complications of flow and temperature distribution that arise in the full-scale DPF can be significantly minimized while critical parameters that affect the filtration performance can be fully controlled. Cold flow experiments were performed to test the system's reliability, and determine the permeability of each wafer type. A Computational Fluid Dynamics (CFD) package was utilized to ensure the flow uniformity inside the filter holder during the cold flow test.

Three Cordierite diesel particulate filters (DPF) with variations in the washcoat† (bare, washcoat-only, and catalyzed washcoat) were filled with equal amounts of PM (∼2 g/l) from a single steady-state engine operating condition (30% load, 1800 rpm). Two regeneration systems were used: an electrical furnace to extract the kinetic parameters by performing Temperature Programmed Oxidation (TPO) experiments and an inline burner to study how DPF washcoat variations affect active regeneration performance. Detailed emissions measurements were performed upstream and downstream of the DPF during the filtration and regeneration processes to quantify DPF filtration and regeneration performance. These measurements included gaseous emission, PM mass concentration, and particle size distribution.

Empirical models for engine-out oxides of Nitrogen (NOx) and smoke emissions have been developed for the purpose of minimizing transient emissions while maintaining transient response. Three major issues have been addressed: data acquisition, data processing and modeling method. Real and virtual transient parameters have been identified for acquisition. Accounting for the phase shift between transient engine events and transient emission measurements has been shown to be very important to the quality of model predictions. Several methods have been employed to account for the transient transport delays and sensor lags which constitute the phase shift. Finally several different empirical modeling methods have been used to determine the most suitable modeling method for transient emissions. These modeling methods include several kinds of neural networks, global regression and localized regression.

A first law based regression model for estimating mean value engine torque on-board a diesel engine is presented. The model uses first law terms across the engine control volume in a regression built from least squares to predict engine torque. Torque information is often required by the engine ECM for torque based control and torque broadcast purposes. In the absence of real-time torque measurement torque estimation is usually achieved through look-up tables or empirical models. Given the increase in engine operating parameters as well as engine operating regimes as a result of emission control and exhaust aftertreatment technologies, accurate torque estimation has become more challenging as well as necessary.

An investigation of high speed direct injection (DI) compression ignition (CI) engine combustion fueled with gasoline injected using a triple-pulse strategy in the low temperature combustion (LTC) regime is presented. This work aims to extend the operation ranges for a light-duty diesel engine, operating on gasoline, that have been identified in previous work via extended controllability of the injection process. The single-cylinder engine (SCE) was operated at full load (16 bar IMEP, 2500 rev/min) and computational simulations of the in-cylinder processes were performed using a multi-dimensional CFD code, KIVA-ERC-Chemkin, that features improved sub-models and the Chemkin library. The oxidation chemistry of the fuel was calculated using a reduced mechanism for primary reference fuel combustion chosen to match ignition characteristics of the gasoline fuel used for the SCE experiments.

Neural networks have been used for engine computations in the recent past. One reason for using neural networks is to capture the accuracy of multi-dimensional CFD calculations or experimental data while saving computational time, so that system simulations can be performed within a reasonable time frame. This paper describes three methods to improve upon neural network predictions. Improvement is demonstrated for in-cylinder pressure predictions in particular. The first method incorporates a physical combustion model within the transfer function of the neural network, so that the network predictions incorporate physical relationships as well as mathematical models to fit the data. The second method shows how partitioning the data into different regimes based on different physical processes, and training different networks for different regimes, improves the accuracy of predictions.

An integrated system model containing sub-models for diesel engine, emissions, and aftertreatment devices has been developed. The objective is to study engine-device and device-device interactions. The emissions sub-models used are for NOx and PM (particulate matter) prediction. The aftertreatment sub-models used include a diesel oxidation catalyst (DOC) and a diesel particulate filter (DPF). Controllers have also been developed to allow for transient simulations, active DPF regeneration, and prevention/control of runaway DPF regenerations. The integrated system-level model has been used to simulate DPF regeneration via exhaust fuel injection ahead of the DOC. In addition, the controller model can use intake throttling to assist in active DPF regeneration if needed. Regeneration studies have been done for both steady engine load and with load transients. High to low engine load transients are of particular interest because they can lead to runaway DPF regeneration.

An overall diesel engine and aftertreatment system model has been created that integrates diesel engine, exhaust system, engine emissions, and diesel particulate filter (DPF) models using MATLAB Simulink. The 1-D engine and exhaust system models were developed using WAVE. The engine emissions model combines a phenomenological soot model with artificial neural networks to predict engine out soot emissions. Experimental data from a light-duty diesel engine was used to calibrate both the engine and engine emissions models. The DPF model predicts the behavior of a clean and particulate-loaded catalyzed wall-flow filter. Experimental data was used to validate this sub-model individually. Several model integration issues were identified and addressed. These included time-step selection, continuous vs. limited triggering of sub-models, and code structuring for simulation speed. Required time-steps for different sub models varied by orders of magnitude.

An integrated system model containing sub-models for a multi-cylinder diesel engine, NOx and soot(PM) emissions, diesel oxidation catalyst (DOC) and diesel particulate filter (DPF) has been developed to simulate the engine and aftertreatment systems at transient engine operating conditions. The objective of this work is two-fold; ensure correct implementation of the integrated system level model and apply the integrated model to understand the fuel economy trade-off for various DPF regeneration strategies. The current study focuses on a 1.9L turbocharged diesel engine and its exhaust system. The engine model was built in GT-Power and validated against experimental data at full-load conditions. The DPF model is calibrated for the current engine application by matching the clean DPF pressure drop for different mass flow rates. Load, boost pressure, speed and EGR controllers are tuned and linked with the current engine model.

There is a significant global effort to study low temperature combustion (LTC) as a tool to achieve stringent emission standards with future light duty diesel engines. LTC utilizes high levels of dilution (i.e., EGR > 60% with <10%O2 in the intake charge) to reduce overall combustion temperatures and to lengthen ignition delay, This increased ignition delay provides time for fuel evaporation and reduces in-homogeneities in the reactant mixture, thus reducing NOx formation from local temperature spikes and soot formation from locally rich mixtures. However, as dilution is increased to the limits, HC and CO can significantly increase. Recent research suggests that CO emissions during LTC result from the incomplete combustion of under-mixed fuel and charge gas occurring after the premixed burn period [1, 2]1. The objective of the present work was to increase understanding of the HC/CO emission mechanisms in LTC at part-load.

An integrated system model containing sub-models for a diesel engine, NOx and soot emissions, and a diesel particulate filter (DPF) has been used to simulate stead-state engine operating conditions. The simulation results have been used to investigate the effect of DPF loading and passive regeneration on engine performance and emissions. This work is the continuation of previous work done to create an overall diesel engine/exhaust system integrated model. As in the previous work, a diesel engine, exhaust system, engine soot emissions, and diesel particulate filter (DPF) sub-models have been integrated into an overall model using Matlab Simulink. For the current work new sub-models have been added for engine-out NOx emissions and an engine feedback controller. The integrated model is intended for use in simulating the interaction of the engine and exhaust aftertreatment components.