Crank-angle resolved cylinder pressure data is valuable for characterizing engine performance and various techniques have been developed for post-processing the pressure traces to understand the rate of heat release and its overall impact on engine performance. However, many of these techniques rely on accurate knowledge of the compression ratio, which may not be well-known because of uncertainties in component dimensions for new and rebuilt engines. Additionally, uncertainties in cylinder pressure referencing and top dead center (TDC) offset can lead to variation in the calculation of these parameters. A method was developed to estimate the compression ratio and heat transfer sensitivity for large bore diesel engines using GT-Power and experimental cylinder pressure traces. An injector cutout method was used on a 228.6mm bore 16-cylinder engine to record motoring cylinder pressure traces for an individual cylinder. The cylinder pressure traces were pegged thermodynamically by matching the slope of a 40-deg window of the compression trace with that of a GT-power simulation of a similar condition. Once the cylinder pressure was properly referenced, it was found that the compression ratio of the cylinder could be estimated by matching the slope of the compression trace over a larger crank angle window. Additionally, it is shown that the location of peak cylinder pressure is dependent on heat transfer and if the location of peak cylinder pressure relative to top dead center is accurately known, then the heat transfer coefficients in GT-Power can be estimated. For an engine where the exact compression ratio may not be known due to variations in hardware dimensions (for both new and rebuilt engines), this method provides a simple path to estimating compression ratio. Furthermore, by measuring the exact location of TDC and comparing that to the location of peak cylinder pressure, heat transfer can be estimated.

The predictive ability of artificial neural networks where a large number of experimental data are available, has been studied extensively. Studies have shown that ANN models are capable of accurately predicting NO x emissions from engines under various operating conditions and different fuel types when trained well. One of the major advantages of an ANN model is its ability to relearn when new experimental data is available, thus continuously improving its accuracy. The present work explored the potential of an ANN model to predict NO x emissions for various engine configurations outside its training envelop. This work also looked into quantifying the amount of new data required to improve the accuracy of the model when exposed to unknown conditions. The chosen ANN model was constructed using data from a high-speed direct injection diesel engine and is capable of accurate NO x emissions over a wide range of operating conditions. The optimized network utilized 14 input parameters and is using 6 neurons in a single hidden layer feed-forward neural network. Experimental data from the various engine configurations tested, were then used to predict NO x from the existing ANN model. The results indicate that when the new data are within the baseline training envelop, the ANN model is capable of accurate NO x prediction even when there are substantial changes in engine configuration such as piston material. Similar results were also observed when the injector nozzle is changed. However, the model’s performance drops significantly when new data, outside the baseline training envelop, were employed indicating that additional training is required. As such, various methods for retraining the ANN model were explored with the selected method showing the best compromise between new-data accuracy and old-data accuracy retention. The retrained ANN model developed was found to be an effective tool in predicting NO x emissions for different engine configurations and operating conditions.

Intake valve seat recession and subsequent valve leakage were experienced during the late development phase of a new DOHC aluminum engine. This type of failure was not observed in the earlier development phase during engine break-in. To resolve this valve burning issue, a three-dimensional finite element model was used to study the thermal/mechanical behavior of this engine. A parallel experimental program was conducted to evaluate the effect of the hardware changes between the early and late engine development phases on engine performance. The experimental effort also provided engine temperature and measured distortion data for validating the analytical model. The finite element analysis was able to identify the root cause of this condition and to propose structural modifications to remedy this problem rapidly by predicting the distortion and stress distribution within the cylinder head.

With the increasing recognition that variable valve actuation (VVA) in its various forms is a powerful tool for optimizing the performance of internal combustion engines, more and more production systems are being designed and implemented throughout the industry. However, as these control systems become more capable of altering lift, timing, duration, and even the number of valve events, the complexity of designing algorithms and calibrating them becomes enormous. In addition, without prior knowledge of an engine’s response to these algorithms, designing a cost-effective mechanism which provides adequate but not over-reaching capability is difficult. Ricardo has developed methodology for timestep coupled simulations which enables the use of one-dimensional (1-D) gas dynamics simulation of engine performance (WAVE™) coupled to a simulation of the valve actuation mechanism constructed in MATLAB ® and AMESim ® . This arrangement allows valve motion input to the 1-D code to be controlled either manually or by a VVA controller simulation, allowing such engine parameters as torque, fuel consumption, NVH, and EGR rates to be monitored as a function of valve timing strategy. This method allows the examination of such engine development concerns as tolerances, valve velocities and accelerations, and interactions with other engine controls to be studied without the costs, leadtimes, or hardware reliability problems that are associated with prototyping a VVA system. In addition, the interfacing of the valve control/engine performance simulation combination with the Design of Experiments optimization software iSIGHT allows the control system space to be explored automatically, without the brute force numerical search required to examine all permutations of the control strategies. The output of this procedure is an array of requirements which can be quickly translated into a specification document which will guide hardware and controls design efforts.

In this work, an external air and EGR emulation facility has been designed that can replicate a wide range of boosting and EGR delivery systems to a multi-cylinder engine platform. The facility works by removing the incumbent air path and replacing it with externally boosted fresh air that is conditioned using a transient flow and temperature controller. The facility also recycles the actual exhaust gases from the engine whilst removing the constraints of required pressure differences to drive this flow. The resulting system is able to control the boundary conditions of intake air flow - pressure and temperature, engine back-pressure and EGR flow rate independently. Three testing approaches have been described that allow to obtain valuable data across a wide range of the engine map (based on an example of a 1.0L direct injection gasoline engine) also beyond its typical hardware related limits. The facility is designed to be used as part of an engine design optimisation process. The facility generates data of the engine combustion system independently of the associated air path subsystems and excites the boundary conditions beyond those that would be expected from a specific air path design. The data is then used to populate 1D engine models which can be confidently used to predict the performance of a specific air path hardware combination and control strategy.

Engine knock remains one of the major barriers to further improve thermal efficiency of Spark Ignition (SI) engines. Knock can be suppressed by lowering the compression ratio, or retarding the spark ignition timing, however, at an expense of efficiency penalty. SI engine is usually operated at knock-limited spark advance (KLSA) to achieve possibly maximum efficiency with given engine hardware and fuel properties, such as Research Octane Number (RON), Motor Octane Number (MON), and heat of vaporization, etc. Co-optimization of engine design and fuel properties is promising to improve the engine efficiency and predictive CFD models can be used to facilitate this optimization process. However, difficulties exist in predicting KLSA in CFD simulations. First, cyclic variability of SI engine demands that multi-cycle results are required to capture the extreme conditions. Secondly, Mach Courant-Friedrichs-Lewy (CFL) number of 1 is desired to accurately predict the knock intensity (KI), resulting in unaffordable computational cost, especially for multi-cycle simulations. In this study, a new approach to numerically predict KLSA using large Mach CFL number of 50 is proposed. This approach is validated against experimental data for a boosted Direct Injection Spark Ignition (DISI) engine at multiple loads and spark timings. G-equation combustion model coupled with well-mixed chemical kinetic model are used to predict the turbulent flame propagation and end-gas auto-ignition, respectively. Simulations run for 10 consecutive engine cycles at each condition. The results show good agreement between model predictions and experiments in terms of cylinder pressure, combustion phasing and cyclic variation. Engine knock is predicted with early spark ignition timing, indicated by significant pressure wave oscillation and end-gas heat release. Maximum Amplitude of Pressure Oscillation (MAPO) analysis is performed to quantify the KI, and the slope change point in KI extrema is used to indicate the KLSA accurately. Using a smaller Mach CFL number of 5 also results in the same conclusions thus demonstrating that this approach is insensitive to the Mach CFL number. The use of large Mach CFL number allows us to achieve fast turn-around time for multi-cycle engine CFD simulations.

The prospect of analysis-driven pre-calibration of a modern diesel engine is extremely valuable in order to significantly reduce hardware investments and accelerate engine designs compliant with stricter EPA fuel economy regulations. Advanced modeling tools, such as CFD, are often used 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. However, the effectiveness of CFD simulation tools for in-cylinder engine combustion is often compromised by the complexity, accuracy, and computational overhead of detailed chemical kinetics necessary for combustion calculations. The standard approach has been to use skeletal kinetic mechanisms (∼50 species) which consume acceptable computational time but with degraded accuracy. In this work, a comprehensive demonstration and validation of the analytical pre-calibration process is presented for a passenger car diesel engine using CFD simulations with CONVERGE™ and a GPU-based chemical kinetics solver (Zero-RK, developed at Lawrence Livermore National Laboratory) on high performance computing resources to enable the use of detailed kinetic mechanisms. Diesel engine combustion computations have been conducted over 600 operating points spanning in-vehicle speed-load map, using massively parallel ensemble simulation sets on the Titan supercomputer located at the Oak Ridge Leadership Computing Facility. The results with different mesh resolutions have been analyzed to compare differences in combustion and emissions (NOx, Carbon Monoxide CO, Unburned Hydrocarbons UHC, and Smoke) with actual engine measurements. The results show improved agreement in combustion and NOx predictions with a large n-heptane mechanism consisting of 144 species and 900 reactions with refined mesh resolution; however; agreement in CO, UHC and Smoke remain a challenge.

This paper is one of the first publications documenting the performance optimization of a port fuel injected, spark ignited, internal combustion engine for an unmanned aerial vehicle hybrid powertrain. An optimized hybrid powertrain enables long-duration quadcopter flight and provides vertical takeoff and landing capability for fixed-wing aircraft. This technology can improve flight times and lessen runway requirements, helping to solve two of the biggest problems faced by the booming UAV industry. The main activities contributing to this hybrid powertrain effort are the development of a small engine test facility, design of powertrain vehicle integration hardware, and the testing and characterization of both electric motor-generators and fuel-injected two-stroke engines. This paper investigates engine testing completed to date including experimental setup, testing procedures, data post processing, and results.

To enable sustainable power generation through increasing shares of renewable energy, it is necessary to find flexible solutions that use conventional fossil fuels to compensate for volatile energy production from the wind and sun in order to stabilize the electrical grid. Modern large bore engines fueled by gas are already able to ramp up or shut down production quickly and also provide high efficiency throughout all load conditions. Nevertheless, transient capabilities of these engines must be improved even more in order to compete with diesel engines in applications with the highest transient requirements. To meet these demands, sophisticated actuators and control strategies are required. Testing of these components and strategies should already be conducted in an early development phase using rapid prototyping simulation and measurements on single cylinder engines instead of expensive multicylinder engine tests. The first section of this paper shows how engine controller functions for transient operation based on rapid prototyping models and real-time capable models can be derived and tested. This enables the capabilities of different control strategies to be quantified in order to improve transient performance in an early stage of development. The second section of the paper presents a methodology for transferring the transient behavior of a large multicylinder engine to a single cylinder test bed using a hardware-in-the-loop (HiL) approach with real time capable simulation models. A description of the demands on hardware and software is provided followed by a description of the overall system, after which the application of the real-time capable models on the real-time controllers of the test bed system is introduced. Finally, the models with measurement data from the single cylinder engine are compared with the multicylinder engine with a special focus on block loads and ramping the engine at constant speed.

Low temperature and dilute Homogenous Charge Compression Ignition (HCCI) and Spark Assisted Compression Ignition (SACI) can improve fuel economy and reduce engine-out NO x emissions to very low values, often less than 30 ppm. However, these combustion modes are unable to achieve stringent future regulations such as SULEV 30 without the use of lean aftertreatment. Though active selective catalytic reduction (SCR) with urea injection and lean NO x traps (LNT) have been investigated as options for lean gasoline engines, a passive TWC-SCR system is investigated in this work because it avoids the urea storage and dosing hardware of a urea SCR system, and the high precious metal cost of an LNT. The TWC-SCR concept uses periodic rich operation to produce NH 3 over a TWC to be stored on an SCR catalyst for subsequent NO x conversion during lean operation. In this work a laboratory study was performed with a modified 2.0 L gasoline engine that was cycled between lean HCCI and rich SACI operation, or between lean and rich SI (spark ignited) combustion, to evaluate NO x conversion and reduced fuel consumption. Different lambda values during rich operation and different times held in rich operation were investigated. Results are compared to a baseline case in which the engine is always operated at stoichiometric conditions. SCR system simulations are also presented that compare system performance for different levels of stored NH 3 . With the configuration used in this study, lean/rich HCCI/SACI operation showed a maximum NO x conversion efficiency of 10%, while lean/rich SI operation showed a maximum NO x conversion efficiency of 60%. However, if the low conversion efficiency of lean/rich HCCI/SACI operation could be improved through higher brick temperatures or additional SCR bricks, simulation results indicate TWC-SCR aftertreatment has the potential to provide near-zero SCR-out NO x concentration and increased system fuel efficiency. In these simulations, fuel efficiency improvement relative to stoichiometric SI were 7 to15% for lean/rich HCCI/SACI with zero tailpipe NO x and −1 to 5% for lean/rich SI with zero tailpipe NO x emissions. Although previous work indicated increased time for NH 3 to start forming over the TWC during rich operation, less NH 3 production over the TWC per fuel amount, and increased NH 3 slip over the SCR catalyst for advanced combustion systems, if NO x conversion efficiency could be enhanced, improvements in fuel economy and low engine-out NO x from advanced combustion modes would more than make up for these disadvantages.

The need for cost-effective fuel economy improvements has driven the introduction of automatic transmissions with an increasing number of gear ratios. Incorporation of interlocking dog clutches in these transmissions decreases package space and increases efficiency, as compared to conventional dry or wet clutches. Unlike friction based clutches, interlocking dog clutches require very precise rotational speed matching prior to engagement. Precise engine speed control is therefore critical to maintaining high shift quality. This research focuses on controlling the engine speed during a gearshift period by manipulating throttle position and combustion phasing. Model predictive control (MPC) is advantageous in this application since the speed profile of a future prediction horizon is known with relatively high confidence. The MPC can find the optimal control actions to achieve the designated speed target without invoking unnecessary actuator manipulation and violating hardware and combustion constraints. This research utilizes linear parameter varying (LPV) MPC to control the engine speed during the gearshift period. Combustion stability constraints are considered with a control oriented covariance of indicated mean effective pressure model (COV of IMEP). The proposed MPC engine speed controller is validated with a high-fidelity 0-dimensional engine model with crank angle resolution. Four case studies, based on simulation, investigate the impact of different MPC design parameters. They also demonstrate that the proposed MPC engine controller successfully achieves the speed reference tracking objective while considering combustion variation constraints.

Increasingly stringent fuel economy and CO 2 emission regulations provide a strong impetus for development of high efficiency engine technologies. Diesel engines dominate the heavy duty market and significant segments of the global light duty market due to their intrinsically higher thermal efficiency compared to spark ignited (SI) engine counterparts. Predictive simulation tools can significantly reduce the time and cost associated with optimization of engine injection strategies, and enable investigation over a broad operating space unconstrained by availability of prototype hardware. In comparison with 0-D/1-D and 3-D simulations, Quasi-D models offer a balance between predictiveness and computational effort, thus making them very suitable for enhancing the fidelity of engine system simulation tools. A most widely used approach for diesel engine applications is a multi-zone spray and combustion model pioneered by Hiroyasu and his group. It divides diesel spray into packets and tracks fuel evaporation, air entrainment, gas properties and ignition delay (induction time) individually during the injection and combustion event. However, original sub-models are not well suited for modern diesel engines, and the main objective of this work is to develop a multi-zonal simulation capable of capturing the impact of high-injection pressures and Exhaust Gas Recirculation (EGR). In particular, a new spray tip penetration sub-model is developed based on measurements obtained in a high-pressure, high-temperature constant volume combustion vessel for pressures as high as 1450 bar. Next, ignition delay correlation is modified to capture the effect of reduced oxygen concentration in engines with EGR, and an algorithm considering the chemical reaction rate of hydrocarbon-oxygen mixture improves prediction of the heat release rates. Spray and combustion predictions were validated with experiments on a single-cylinder diesel engine with common rail fuel injection, charge boosting, and EGR.

In order to ensure that every portion of the emission control software in a vehicle works, all fault conditions must be tested. Simply simulating faults in the software of the engine controller and reporting it to the OBD II scanner is inadequate; the fault condition must be injected externally to the Engine Control Unit (ECU). In the case of hard-to-reproduce mechanical failures, this is a challenging task. This paper discusses the development of a system capable of emulating various faults that a fuel injector can have while operating as part of a complete working vehicle. For the ECU to operate properly, all fuel injectors must be present in the vehicle, be fully functional, and must represent an accurate electrical load to the ECU. Then, the induced faults must be seamlessly inserted into the running system in less than 10μs and removed before the subsequent injection event. This was accomplished with a variety of COTS hardware, a simple custom circuit, and the use of a large, flexible FPGA platform.

To meet the increasingly stringent emissions standards, Diesel engines need to include more active technologies with their associated control systems. Hardware-in-the-Loop (HiL) approaches are becoming popular when the engine system is represented as a real-time capable model to allow development of the controller hardware and software without the need for the real engine system. This paper focusses on the engine model required in such approaches. A number of semi-physical, zero-dimensional combustion modelling techniques are enhanced and combined into a complete model, these include — ignition delay, pre-mixed and diffusion combustion and wall impingement. In addition, a fuel injection model was used to provide fuel injection rate from solenoid energizing signals. The model was parameterized using a small set of experimental data from an engine dynamometer test facility and validated against a complete data set covering the full engine speed and torque range. The model was shown to characterize Rate of Heat Release (RoHR) well over the engine speed and load range. Critically the wall impingement model improved R 2 value for maximum RoHR from 0.89 to 0.96. This reflected in the model’s ability to match both pilot and main combustion phasing, and peak heat release rates derived from measured data. The model predicted indicated mean effective pressure and maximum pressure with R 2 values of 0.99 across the engine map. The worst prediction was for the angle of maximum pressure which had an R 2 of 0.74. The results demonstrate the predictive ability of the model, with only a small set of empirical data for training — this is a key advantage over conventional methods. The fuel injection model yielded good results for predicted injection quantity (R 2 = 0.99), and enables the use of the RoHR model without the need for measured rate of injection.

In Computational Fluid Dynamics (CFD) simulations of internal combustion engines, one of the critical modeling parameters is the valve setup. A standard workaround is to keep the valve opens at a certain clearance (minimum valve lift), while imposing a solid boundary to mimic valve closure. This method would yield a step change in valve lift during opening and closing event, and different valve event timing than hardware. Two parametric studies were performed to examine a) the effect of the minimum valve lift and b) the effect of grid resolution at the minimum valve lift on predicted in-cylinder flow fields in Reynolds Averaged Navier-Stokes (RANS) simulations. The simulation results were compared with the state-of-art PIV measurement from a two-valve transparent combustion chamber (TCC-3) engine. The comparisons revealed that the accuracy of flow simulation are sensitive to the choice of minimum valve lift and grid resolution in the valve seat region. In particular, the predicted in-cylinder flow field during the intake process was found to be very sensitive to the valve setup. A best practice CFD valve setup strategy is proposed as a result of this parametric studies. The proposed CFD valve setup was applied to Large Eddy Simulation (LES) of TCC-3 engine and preliminary results showed noticeable improvement already. Further evaluation of the valve setup strategy for LES simulations is on-going and will be reported in a separate report.

The gas exchange process is known to be a significant source of cyclic variability in Internal Combustion Engines (ICE). Traditionally, Large Eddy Simulations (LES) are expected to capture these cycle-to-cycle variations. This paper reports a numerical effort to establish best practices for capturing cyclic variability with LES tools in a Transparent Combustion Chamber (TCC) spark ignition engine. The main intention is to examine the sensitivity of cycle averaged mean and Root Mean Square (RMS) flow fields and Proper Orthogonal Decomposition (POD) modes to different computational hardware, adaptive mesh refinement (AMR) and LES sub-grid scale (SGS) models, since these aspects have received little attention in the past couple of decades. This study also examines the effect of near-wall resolution on the predicted wall shear stresses. LES is pursued with commercially available CONVERGE code. Two different SGS models are tested, a one-equation eddy viscosity model and dynamic structure model. The results seem to indicate that both mean and RMS fields without any SGS model are not much different than those with LES models, either one-equation eddy viscosity or dynamic structure model. Computational hardware results in subtle quantitative differences, especially in RMS distributions. The influence of AMR on both mean and RMS fields is negligible. The predicted shear stresses near the liner walls is also found to be relatively insensitive to near-wall resolution except in the valve curtain region.

Low natural gas prices have made the fuel an attractive alternative to diesel and other common fuels, particularly in applications that consume large quantities of fuel. The North American rail industry is examining the use of locomotives powered by dual fuel engines to realize savings in fuel costs. These dual fuel engines can substitute a large portion of the diesel fuel with natural gas that is premixed with the intake air. Engine knock in traditional premixed spark-ignited combustion is undesirable but well characterized by the Methane Number index, which quantifies the propensity of a gaseous fuel to autoignite after a period of time at high temperature. Originally developed for spark-ignited engines, the ability of the methane number index to predict a fuel’s “knock” behavior in dual fuel combustion is not as fully understood. The objective of this effort is to evaluate the ability of an existing methane number algorithm to predict rapid combustion in a dual fuel engine. Sets of specialized natural gas fuel blends that, according to the MWM methane number algorithm, should have similar knock characteristics are tested in a dual fuel engine and induced to experience rapid combustion. Test results and CFD analysis reveal that rapid or aggressive combustion rates happen late in the dual fuel combustion event with this engine hardware configuration. The transition from normal combustion to late rapid combustion is characterized by changes in the heat release rate profiles. In this study, the transition is also represented by a shift in the crank angle location of the combustion’s peak heat release rate. For fuels of similar methane number that should exhibit similar knock behavior, these transitions occur at significantly different relative air-fuel ratios, demonstrating that the existing MWM methane number algorithm, while excellent for spark-ignited engines, does not fully predict the propensity for rapid combustion to occur in a dual fuel engine within the scope of this study. This indicates that physical and chemical phenomena present in rapid or aggressive dual fuel combustion processes may differ from those in knocking spark-ignited combustion. In its current form a methane number algorithm can be used to conservatively rate dual fuel engines. It is possible that derivation of a new reactivity index that better predicts rapid combustion behavior of the gaseous fuel in dual fuel combustion would allow ratings to be less conservative.

Future projections in global transportation fuel use show a demand shift towards diesel and away from gasoline. At the same time greenhouse gas regulations will drive higher vehicle fuel efficiency and lower well-to-wheel CO 2 production. Naphtha, a contributor to the gasoline stream and requiring less processing at the refinery level, is an attractive candidate to mitigate this demand shift while lowering the overall greenhouse gas impact. In this work, low cetane and high volatility gasoline-like fuels have shown potential to achieve high fuel efficiency with low engine-out emissions in a production commercial vehicle engine. This study investigates the combustion and emissions performance of two low cetane naphtha fuels (Naphtha 1: RON59; Naphtha 2: RON69) and one ultra-low sulfur diesel (ULSD) in a model year (MY) 2013, six-cylinder, heavy-duty diesel engine. The engine is equipped with a single-stage variable geometry turbocharger (VGT) and a fuel injection system that is capable of 2500 bar fuel injection pressure. The engine has a stock geometric compression ratio of 18.9. To date, most studies in this area have been conducted using single-cylinder research engines. Aramco aims to better understand the implications on hardware and software design in a multi-cylinder engine with a production engine air system. Engine testing was focused on the Heavy-Duty Supplemental Emissions Test (SET) “B” speed over a load sweep from 5 to 15 bar BMEP. At each operating point, NOx sweeps were conducted over wide ranges (e.g., 0.2 → 3 g/hp-hr) to understand the implications of fuel reactivity as well as other properties on combustion behavior under both high temperature mixing-controlled combustion and low temperature premixed combustion. At 10–15 bar BMEP, mixing-controlled combustion dominates the engine combustion process. Under a compression ratio of 18.9, cylinder pressure and temperature are sufficiently high to suppress the reactivity (cetane number) difference between ULSD and the low cetane naphtha fuels. As a result, the three test fuels showed similar ignition delay under high temperature and pressure conditions. Nevertheless, naphtha fuels still exhibited notable soot reduction compared to ULSD. Under mixing-controlled combustion, this is likely due to their lower aromatic content and higher volatility. At 10 bar BMEP, Naphtha 1 generated less soot than Naphtha 2 since it contains less aromatics and is more volatile. When operated at light load, in a less reactive thermal environment, the lower reactivity naphtha fuels led to longer ignition delays than ULSD. As a result, the soot benefit of naphtha fuels was enhanced. Overall, naphtha fuels and ULSD had similar fuel efficiency. Utilizing the soot benefit of the naphtha fuels, engine-out NOx was calibrated from the production level of 3–4 g/hp-hr down to 2–2.5 g/hp-hr over the twelve non-idle SET steady-state modes. At this reduced NOx level, naphtha fuels were still able to maintain a soot advantage over ULSD and remain “soot-free” (smoke ≤ 0.2 FSN) while achieving diesel-equivalent fuel efficiency. Finally, partially premixed compression ignition (PPCI) low temperature combustion (LTC) operation (NOx ≤ 0.2 g/hp-hr; smoke ≤ 0.2 FSN) was achieved with both of the naphtha fuels at 5 bar BMEP through a late injection approach with high injection pressure. Under high EGR dilution, Naphtha 2 showed an appreciably longer ignition delay than Naphtha 1, resulting in a soot reduction benefit. Early injection PPCI operation cannot be attained with the stock engine compression ratio due to excessive pressure rise rates. Although the late injection PPCI operation offered a significant NOx benefit over mixing-controlled combustion operation, it led to lower fuel efficiency with undesirably late combustion phasing. This points the research towards a lower engine compression ratio and an air system upgrade to promote high efficiency PPCI LTC operation.

Over time, environmental protection standards have become more strict and complex. Nitrogen oxides (NO x ) are regulated pollutants produced by combustion in a diesel engine. In this project, camshaft timing modifications were studied as a way of reducing NO x emission levels while using low cost hardware. Different valve timing strategies were proposed and modelled using engine simulation. This project was based on two concepts. The first was to open the intake valve during the exhaust stroke, thus expelling burnt gases from the cylinder into the intake manifold and then later re-admitting these gases into the cylinder during the intake stroke of the next cycle. The second was to open the exhaust valve during the intake stroke, allowing burnt gases from the exhaust manifold to enter the cylinder at the same time as the fresh charge enters. Both technologies studied were able to recirculate the exhaust gases without an external EGR system. The EGR amount was controlled by either an intake throttle or an exhaust throttle. The amount of EGR was predicted using engine simulation. The brake-specific fuel consumption (BSFC) and brake-specific NO x (BSNO x ) trade off was the main criterion used to select the best technology, although other features such as predicted manifold pressures and engine-out soot were also considered. The results indicate that, by using increased amounts of EGR while varying the intake or exhaust throttle position, NO x emissions can be reduced with a slight BSFC penalty. These methods are thus a low cost means of reducing engine-out NO x emissions.

The combustion system of a large bore four stroke diesel engine was developed using the Analysis Led Design (ALD) process. A combustion strategy of particulate matter (PM) control in-cylinder and oxides of nitrogen (NO x ) controlled through urea based selective catalytic reduction (SCR) of the engine exhaust, was adopted. Combustion analysis tools and single cylinder engine (SCE) tests were used to determine key combustion parameters like in-cylinder swirl, fuel injector configuration and, combustion chamber design to meet the EPA Tier4 final emission regulations at the different notch settings of the locomotive test cycle. Experimental validation of the ALD recommended hardware was accomplished on the SCE and the best combination was selected. Engine out PM and NO x target compliance was demonstrated on the multi cylinder diesel engine (MCE) over the composite locomotive notch cycle. Selective fuel Injector strategy was applied to control excessive unburned hydrocarbons (UHC) at low load engine operating conditions within the locomotive notch cycle.