Emission reduction after starting an internal combustion engine

Abstract

A method of reducing emissions in an engine includes determining a temperature of an aftertreatment device. When the temperature is less than a dosing temperature, determining whether a load on the engine is above a load threshold. When the load is below the load threshold, injecting fuel with an early timing whereby the fuel burns in a partially-premixed combustion mode. When the load is above the load threshold, decreasing a fuel injection pressure to a lower value; and injecting the fuel with a delayed injection timing at the lower value, whereby the fuel burns in a diffusion-flame combustion mode.

Description

EMISSION REDUCTION AFTER STARTING AN INTERNAL COMBUSTION ENGINETechnical Field

[0001] The present application relates to emission reduction after starting an internal combustion engine, and in particular reductions in oxides of nitrogen after cold start.

[0002] The following discussion of the background to the invention is intended to facilitate an understanding of the present invention only. It should be appreciated that the discussion is not an acknowledgement or admission that any of the material referred to was published, known or part of the common general knowledge of the person skilled in the art in any jurisdiction as at the priority date of the invention.

[0003] A late cycle, direct injection (LCDI) operating mode in an internal combustion engine includes injecting a main fuel later during a compression stroke of the internal combustion engine, and in some circumstances early during a power stroke, and employing a pilot fuel to ignite the main fuel. The main fuel typically has a relatively low cetane number such that it does not auto-ignite due to the pressure and temperature environment in a combustion chamber of the internal combustion engine due to the heat of compression created during the compression stroke. The combustion of the pilot fuel is then used to create a pressure and temperature environment suitable for igniting the main fuel. The main fuel and the pilot fuel both bum in a diffusion-flame combustion mode.

[0004] The emissions of oxides of nitrogen (NOx) vary depending upon the type of main fuel employed. For example, the LCDI operating mode produces higher engine-out NOx emissions when the main fuel is hydrogen compared to when the main fuel is natural gas. It has been recognized that a majority of NOx emissions occur during and just after a cold start where engine components are below preferred operational temperatures and where the emissions increase as the ambient temperature decreases. For example, a majority of NOx emissions occur during a warm-up period for a selective catalytic converter (SCR) when the exhaust and SCR temperatures are too low for ureadosing. More stringent NOx regulations for future internal combustion engines are expected and will be a challenge to meet these future regulations in engines employing the LCDI mode.

[0005] The state of the art is lacking in techniques for emission reduction after starting an internal combustion engine, and in particular reductions in oxides of nitrogen after cold starting.

[0006] An improved method of reducing emissions in an internal combustion engine includes determining whether a temperature of an aftertreatment device is less than a dosing temperature. When the temperature of the aftertreatment device is less than the dosing temperature, determining whether a load on the internal combustion engine is above a load threshold. When the load is below the load threshold, (1) injecting fuel with an early timing calibration whereby the fuel bums in a partially-premixed combustion mode, or (2) decreasing a fuel injection pressure to a lower value compared to the fuel injection pressure during operation of the engine at the load when the temperature of the aftertreatment device is greater than the dosing temperature; and injecting the fuel with a delayed injection timing at the lower value of the fuel injection pressure, whereby the fuel bums in a diffusion-flame combustion mode. When the load is at or above the load threshold, decreasing the fuel injection pressure to the lower value compared to the fuel injection pressure during operation of the engine at or above the load threshold when the temperature of the aftertreatment device is greater than the dosing temperature; and injecting the fuel with a delayed inj ection timing at the lower value of the fuel inj ection pressure, whereby the fuel bums in a diffusionflame combustion mode. Oxides of nitrogen (NOx) emissions are reduced accordingly.

[0007] In some embodiments, a minimum value of the lower value of the fuel injection pressure increases with at least one of the load and a boost pressure. The load threshold can be 50% of a maximum rated load. The fuel can be ignited by at least one of a pilot fuel, a spark igniter, and a heated surface. The fuel can be a main fuel ignited by the pilot fuel. The main fuel can be injected between 120 crank angle degrees before top dead center during a compression stroke and 10 crank angle degrees before top dead center during the compression stroke when burned in the partially- premixed combustion mode. The main fuel can be injected between 30 crank angle degrees before top dead center during a compression stroke and 40 crank angle degrees after top dead center during a power stroke when burned in the diffusion-flame combustion mode. The injection timing can bedelayed such that an exhaust temperature is at an exhaust temperature limit. The main fuel can include hydrogen and the pilot fuel can include diesel, dimethyl ether, or kerosene.

[0008] The method can further include determining whether a temperature of the aftertreatment device is less than a first low temperature threshold; determining whether a time-averaged exhaust temperature is less than the dosing temperature. When the temperature of the aftertreatment device is less than the first low temperature threshold and the time-averaged exhaust temperature is less than the dosing temperature, decreasing an amount of air within a combustion chamber in the internal combustion engine after an intake valve is closed for respective engine load and engine speed conditions such that a fuel / air equivalence ratio is increased, whereby an exhaust temperature increases causing an increase in a rate of rise of the temperature of the aftertreatment device.

[0009] The method can further include actuating a throttle valve to decrease the amount of air within the combustion chamber. The method can further include actuating the wastegate valve to decrease the amount of air within the combustion chamber by decreasing a boost pressure. The method can further include returning at least a portion of exhaust to a combustion chamber of the internal combustion engine for a subsequent engine cycle. The exhaust can be returned by an external exhaust gas recirculation loop. The exhaust can be returned by internal exhaust gas recirculation. An amount of recirculated exhaust gas is at least 15% of total exhaust gas.

[0010] The method can further include, when the temperature of the aftertreatment device increases to or above the dosing temperature, increasing the fuel injection pressure; and injecting the fuel without the delayed timing calibration at the increased fuel injection pressure; whereby the fuel bums in the diffusion-flame combustion mode. The method can further include injecting a diesel exhaust fluid into an exhaust upstream of the aftertreatment device wherein oxides of nitrogen (NOx) emissions are reduced. The diesel exhaust fluid can be urea. The aftertreatment device can include a catalytic converter. The catalytic converter can be a selective catalytic reduction converter.

[0011] The accompanying drawings, which are incorporated into and constitute a part of the specification, illustrate specific embodiments of the apparatus, systems, and methods and, togetherwith the general description above, and the detailed description of the specific embodiments, serve to explain the principles of the apparatus, systems, and methods.

[0012] FIG. 1 is a schematic view of an internal combustion engine according to an embodiment.

[0013] FIG. 2 is a cross-sectional view of a cylinder of the internal combustion engine of FIG. 1.

[0014] FIG. 3 is a schematic view of an internal combustion engine according to an embodiment.

[0015] FIG. 4 is a schematic view of an internal combustion engine according to an embodiment.

[0016] FIG. 5 is a schematic view of an internal combustion engine according to an embodiment.

[0017] FIG. 6 is a chart view of fuel-injection mass-flow for fuel injectors of FIG. 1, FIG. 3, FIG.4, or FIG. 5 for a late cycle, direct injection operating mode according to an embodiment.

[0018] FIG. 7 is a chart view of fuel-injection mass-flow for fuel injectors of FIG. 1, FIG. 3, FIG.4, or FIG. 5 for a mid-cycle, direct injection operating mode according to an embodiment.

[0019] FIG. 8 is a flow chart of an emission reduction algorithm after starting the internal combustion engine of FIGS. 1, 3, 4, and 5 according to an embodiment.

[0020] FIG. 9 is a flow chart of an emission reduction algorithm after starting the internal combustion engine of FIGS. 1, 3, 4, and 5 according to an embodiment.

[0021] FIG. 10 is a chart illustrating the effect of a late cycle, direct injection operating mode with low main rail pressure PMR and a mid-cycle, direct injection operating mode with different pulse separation PSEP timing on brake thermal efficiency.

[0022] FIG. 11 is a chart illustrating the effect of the late cycle, direct injection operating mode with low main rail pressure PMR and the mid-cycle, direct injection operating mode with different pulse separation PSEP timing on NOx emissions.

[0023] FIG. 12 is a chart illustrating the effect of main rail pressure PMR on brake thermal efficiency for various engine operating conditions.

[0024] FIG. 13 is a chart illustrating the effect of main rail pressure PMR on NOx emissions for various engine operating conditions.

[0025] FIG. 14 is a chart illustrating the effect of wastegating and throtling on inlet manifold pressure.

[0026] FIG. 15 is a chart illustrating the effect of wastegating and throttling on brake thermal efficiency.

[0027] FIG. 16 is a chart illustrating the effect of wastegating and throtling on exhaust temperature.

[0028] FIG. 17 is a chart illustrating the effect of uncooled and cooled exhaust gas recirculation on brake thermal efficiency for various operating conditions.

[0029] FIG. 18 is a chart illustrating the effect of uncooled and cooled exhaust gas recirculation on NOx emissions for various operating conditions.Detailed

[0030] Referring to FIG. 1, there is shown a schematic view of internal combustion engine 10 according to an embodiment where only elements relevant to the disclosure of the present invention are illustrated, and other elements not illustrated in FIG. 1 that can be employed in internal combustion engines can also be employed by engine 10. Engine 10 can be used for a vehicle, and can also be employed in marine, locomotive, mine haul, power generation, stationary or other applications. Engine 10 includes fuel system 15 and air system 16 that supply fuel and air, respectively, to the engine. Fuel system 15 can store, pressurize, regulate a pressure of, and deliver fuel to engine 10. Air system 16 can compress and cool intake air, prepare air / exhaust-gas mixtures, and deliver air or air / exhaust-gas mixtures to engine 10. Fuel system 15 and air system 16 are now discussed in more detail.

[0031] Fuel system 15 includes main-fuel system 17 and pilot-fuel system 18 in the illustrated embodiment, and each will now be described in turn. Main-fuel system 17 can store the main fuel,regulate a pressure of the main fuel, and deliver the main fuel to engine 10. Main-fuel supply 20 stores the main fuel. In the illustrated embodiment, the main fuel is a gaseous fuel that is stored as a compressed gas at a rated fdl pressure. In some embodiments, the main fuel is a gaseous fuel stored as a cryogenic fluid. In some embodiments, the main fuel is a liquid fuel that can be stored substantially at atmospheric pressure. A gaseous fuel is any fuel that is in the gas state at standard temperature and pressure, which is defined herein as a temperature of 0 degrees Celsius (0 °C) and an absolute pressure of 100,000 Pascals (100 kPa), respectively. A liquid fuel is any fuel that is in a liquid state at standard temperature (0 °C) and pressure (100 kPa). Pilot-fuel supply 90 stores the pilot fuel substantially at atmospheric pressure. Although engine 10 can consume a variety of types of gaseous fuel, hydrogen, methane, natural gas, or mixtures thereof are exemplary main fuels.

[0032] The main fuel is fluidly communicated to pressure regulator 50 from main-fuel supply 20 through shut-off valve 40. Pressure regulator 50 can regulate main rail pressure PMR in main rail 70. Accumulator 60 can provide a predetermined volume of the main fuel at main rail pressure PMR to reduce pressure fluctuations in main rail 70 as engine 10 operates, and in other embodiments accumulator 60 can be a delivery pipe appropriately sized for supplying main fuel from pressure regulator 50 to main rail 70. Main rail pressure PMR can be measured by main-rail pressure sensor 80, which generates a signal representative of main rail pressure PMR that is received by controller 100 and can measure the main rail pressure PMR directly in main rail 70, in accumulator 60, or a delivery pipe between pressure regulator 50 and main rail 70. Pressure regulator 50 can selectively regulate main rail pressure PMR to two or more predetermined rail pressures and can be commanded by controller 100 to select which predetermined rail pressure to regulate main rail pressure PMR to in main rail 70. In exemplary embodiments, pressure regulator 50 can employ mechanical pressure regulation or electronic pressure regulation. In other exemplary embodiments pressure regulator can be a fuel injector that is commanded by controller 100 to inject fuel into accumulator 60 and main rail 70. Regulator-bypass valve 42 can be employed to deliver the main fuel from main-fuel supply 20 to main rail 70 without regulating a pressure of the main fuel, such that the main rail pressure PMR is equal to the main supply pressure PMS. In other embodiments regulator-bypass valve 42 is not required such that the main fuel is always pressure regulated, or pressure regulator 50 is not required such that the main fuel is not pressure regulated.

[0033] Pilot-fuel system 18 can store a pilot fuel, pressurize the pilot fuel, regulate a pressure of the pilot fuel, and deliver the pilot fuel to engine 10. The pilot fuel is employed to ignite the main fuel in engine 10 and is typically a liquid fuel, although it is not required to be a liquid fuel. In the illustrated embodiment, pilot-fuel system 18 is configured for a liquid fuel. Pilot-fuel supply 90 stores the pilot fuel substantially at atmospheric pressure. Main fuels typically have a relatively low cetane number and are difficult to auto-ignite in the pressure and temperature environment typically found in combustion chambers of internal combustion engines during a compression stroke. Accordingly, the pilot fuel is employed to ignite the main fuel, where the pilot fuel has a cetane number high enough such that autoignition and combustion of the pilot fuel occurs in the pressure and temperature environment typically found in combustion chambers during the compression stroke. An exemplary pilot fuel is diesel fuel, which is stored in liquid form substantially at atmospheric pressure in pilotfuel supply 90; however, other pilot fuels are contemplated such as dimethyl either (DME) and kerosene. Low-pressure pump 110 pressurizes and fluidly supplies the pilot fuel from pilot-fuel supply 90 through shut-off valve 120 to high-pressure pump 130 at a suitable pressure for further pressurization. High-pressure pump 130, which can include an inlet metering valve and a common rail pump, pressurizes the pilot fuel to pilot delivery pressure PPD in delivery conduit 140. Pressure regulator 150 regulates pilot delivery pressure PPD down to pilot rail pressure PPR in pilot rail 160. In the illustrated embodiment, pressure regulator 150 regulates pilot rail pressure PPR in pilot rail 160 with respect to main rail pressure PMR (known as pilot follows main) such that a differential pressure between pilot rail pressure PPR and main rail pressure PMR is within a predetermined range. The differential pressure between pilot rail pressure PPR and main rail pressure PMR is known as system bias pressure. Pilot delivery pressure sensor 170 generates a signal representative of pilot delivery pressure PPD that is received by controller 100. Pilot rail pressure sensor 180 generates a signal representative of pilot rail pressure PPR that is received by controller 100. In other embodiments, main storage pressure PMS can be down regulated as a function of pilot rail pressure PPR to main rail pressure PMR (known as main follows pilot) and pilot rail pressure PPR can be substantially equal to pilot delivery pressure PPD, although it is not required to be and the pilot delivery pressure PPD can still be down regulated to pilot rail pressure PPR.

[0034] Main rail 70 and pilot rail 160 deliver the main fuel and the pilot fuel, respectively, to each fuel injector 190 associated with respective cylinders 200. In the illustrated embodiment there are six cylinders 200 and respective fuel injectors 190; however, in other embodiments there can bea single cylinder and fuel injector or another plurality of cylinders and fuel injectors. Each fuel injector 190 is a concentric needle, dual-fuel, in-cylinder injector that directly injects the main fuel into cylinder 200 and directly injects the pilot fuel into cylinder 200 separately and independently of the main fuel. Fuel injector 190 can be like fuel injectors disclosed in the Applicant’s United States patent no. 10,294,908 issued on May 21, 2019, and / or United States patent no. 10,502,169 issued on December 10, 2019.

[0035] Air system 16 fluidly communicates and conditions air from air intake 210 to intake manifold 280 that distributes the intake air to cylinders 200. Turbocharger 220 selectively pressurizes the intake air when exhaust gas is fluidly communicated through turbine 230 thereby driving compressor 240 to pressurize the intake air. Turbocharger 220 can be a variable geometry turbocharger, and in other embodiments can be a dual stage turbocharging system. Compressed intake air leaving an outlet of compressor 240 is elevated in temperature due to compression, compared to fresh intake air, and accordingly is communicated through charge air cooler 250 where it is cooled. In some embodiments, charge air cooler (CAC) bypass valve 255 allows intake air to be bypassed around charge air cooler 250, such as during startup as will be described in more detail below. There can be an additional valve (not shown) that fluidly blocks charge air cooler 250 from compressor 240 when CAC-bypass valve 255 is open. Wastegate valve 270 can be actuated by controller 100 between a closed position and a fully opened position, and in some embodiments to partially open positions therebetween, to allow at least a portion of exhaust gases to bypass turbine 230, which is referred to as wastegating herein.

[0036] During some engine operating conditions, a portion of exhaust gases from cylinders 200 can be communicated to intake manifold 280 from exhaust manifold 290 through external exhaust gas recirculation (EGR) loop 300. Controller 100 commands EGR valve 310 between a closed position and an open position, and in some embodiments partially open positions therebetween, to control the EGR mass flow rate through external EGR loop 300, for a given back pressure in exhaust manifold 290. EGR cooler 320 reduces the temperature of exhaust gasses to protect intake manifold 280 and to tower in-cylinder temperatures. In some embodiments, EGR-bypass valve 325 allows recirculated exhaust gases to bypass EGR cooler 320, such as during startup as will be described in more detail below. In the illustrated embodiment at least a portion of exhaust gases from all cylinders 200 can be recirculated to intake manifold 280. In other embodiments any other EGR architecturecan be employed, such as those that dedicate one or more cylinders, or one or more exhaust ports, to EGR. Exhaust gases not passing through external EGR loop 300 are communicated through turbine 230 and / or wastegate valve 270 to engine aftertreatment system 330. Engine aftertreatment system 330 includes a selective catalytic reduction converter.

[0037] Intake air is fluidly communicated from intake manifold 280 through respective intake ports 340 and intake valves 350 to cylinders 200. Although the illustrated embodiment shows two intake valves 350 for each cylinder 200, in other embodiments there can be only one intake valve 350 or more than two intake valves 350 for each cylinder 200. Exhaust gases are fluidly communicated from cylinders 200 through respective exhaust valves 360 and exhaust ports 370 into exhaust manifold 290. Although the illustrated embodiment shows two exhaust valves 360 for each cylinder 200, in other embodiments there can be only one exhaust valve 360 or more than two exhaust valves 360 for each cylinder 200.

[0038] Referring now to FIG. 2, each cylinder 200, having longitudinal axis 202, is defined by inner bore surface 380 in engine block 390. For each cylinder 200, there is a piston 400 that reciprocates within the cylinder 200 and that is connected with a crankshaft (now shown) through crank arm 410 that turns reciprocal motion of piston 400 into rotational motion of the crankshaft. Combustion chamber 420 is defined by engine block 390, piston 400 and cylinder head 430. More particularly, combustion chamber 420 is defined by inner bore surface 380 of engine block 390, piston crown 402 of piston 400, and piston-facing surface 432 of cylinder head 430. Intake valves 350 and exhaust valves 360 can be actuated by valve actuators 352 and 362, respectively. In some embodiments, valve actuators 352 and 362 can be variable valve actuation (VVA) system 355 that is operatively connected with and commanded by controller 100 to adjust intake valve timing (IVT) of intake valve 350 and exhaust valve timing (EVT) of exhaust valve 360. Valve actuators 352 and 362 in VVA system 355 can employ camshaft-based systems or camless systems. Camshaft based systems can be valve timing control (VTC), also known as variable valve timing (VVT) systems that change the timing of the intake and exhaust valve events without significantly altering the lift. Alternatively, camshaft-based systems can employ variable valve event control and variable valve lift (VVL) control that can provide discrete adjustments or a continuous range of lift and / or duration control between two limits, which can allow modest adjustments in phasing. Camshaft based VVL and VTC systems can be combined to enable lift and timing control. Camless systems can includehydraulically actuated or electromagnetically actuated systems. Camless systems offer more flexibility compared to camshaft-based systems in valve lift and timing but can have an increased risk of interference between the intake or exhaust valve in the lifted position and the piston. VVA system 355 can be employed for internal EGR, also known as residual EGR, which is hot, in-cylinder EGR, in comparison to external EGR loop 300, which is an external EGR system (that can be either hot or cooled depending upon whether EGR cooler 320 is employed to cool the exhaust gas). Internal EGR can be advantageous when the main fuel comprises hydrogen where external EGR loop 300 can be deteriorated due to water condensation and corrosion. In some embodiments, only external EGR loop 300 is employed. In some embodiments, only internal EGR is employed. In some embodiments, both external EGR loop 300 and internal EGR are employed. Knock sensor 450 can be employed to detect and control engine knock or preignition, particularly when internal combustion engine 10 is operated in a premixed combustion mode or a partially-premixed combustion mode. Knock sensor 450 emits signals representative of a level of knock and is operatively connected with controller 100 and can be part of a smart combustion sensor system that processes signals from the knock sensor 450 to determine combustion characteristics, which in turn can detect changes in fuel quality. Controller 100 can employ the smart combustion sensor to improve control of combustion in internal combustion engine 10 due to variations in fuel quality, particularly when the main fuel is a gaseous fuel that can have a wide variation in fuel quality such as natural gas. A preferred location for mounting knock sensor 450 is on a bearing cap (not shown) of internal combustion engine 10.

[0039] A compression ratio of internal combustion engine 10 and all other internal combustion engines disclosed herein can be a geometric compression ratio or an effective compression ratio. The geometric compression ratio is the maximum compression ratio achievable in a particular internal combustion engine and is defined as the ratio between a maximum volume of combustion chamber 420 when piston 400 is at or near bottom dead center when the intake valve 350 closes (during an intake stroke or the compression stroke) over a volume of combustion chamber 420 when piston 400 is at top dead center. The effective compression ratio is less than or equal to the geometric compression ratio, which can be achieved by adjusting intake valve closing timing using VVA (either camshaft-based systems or camless systems) and can be defined as a ratio between a volume of combustion chamber 420 when intake valve 350 closes during the intake stroke or the compression stroke (where exhaust valve 360 is already closed) over a volume of combustion chamber 420 when piston 400 is at top dead center. The effective compression ratio can also take into account blowbyof charge through piston rings (not shown), where the piston rings are disposed between piston 400 and inner bore surface 380 of cylinder 200 to seal combustion chamber 420. The geometric compression ratio and the available effective compression ratio values are programmed in controller 100 herein, whereby the controller knows the current compression ratio. The VVA system 355 can be employed to reduce the effective compression ratio to reduce the likelihood of knock when internal combustion engine 10 is operated in the premixed or the partially-premixed combustion mode.

[0040] Referring now to FIG. 3, there is shown internal combustion engine 11 that is like the previous embodiment where like parts between this and all other embodiments have like reference numerals and differences are discussed. Main-fuel injector 191 and pilot-fuel injector 192 are separate in-cylinder fuel injectors that introduce main fuel and pilot fuel, respectively directly into combustion chamber 420. Alternatively, in other embodiments, main-fuel injector 191 and pilot-fuel injector 192 can be located side-by-side within a common housing of a single, in-cylinder dual fuel injector.

[0041] In other embodiments, a compressor can be employed to pressurize the main fuel as main storage pressure PMS declines below a threshold. Referring to FIG. 4, there is shown internal combustion engine 12 with fuel system 25 including main-fuel system 27 and pilot-fuel system 18 according to an embodiment. Main-fuel system 27 includes compressor 55 that can be commanded by controller 100 to pressurize the main fuel from main-fuel supply 20. Compressor-bypass valve 44 can be actuated by controller 100 to an open position to fluidly communicate the main fuel from main-fuel supply 20 around compressor 55, and to a closed position to fluidly communicate the main fuel through the compressor. Compressor valve 46 can be employed to fluidly isolate compressor 55 from main-fuel supply 20, although in other embodiments compressor valve 46 is not required. Similarly, in other embodiments shut-off valve 40 is not required since compressor-bypass valve 44 and compressor valve 46 together can provide a similar main-fuel shut-off function.

[0042] Referring to FIG. 5, there is shown internal combustion engine 13 with fuel system 35 including main-fuel system 37 and pilot-fuel system 18 according to an embodiment. Main-fuel supply 23 stores the main fuel at cryogenic temperatures in main-fuel supply 23, which can be a double-walled, vacuum insulated storage vessel. Cryogenic pump 56 pressurizes the main fuel at cryogenic temperatures from main-fuel supply 23 to main-fuel delivery pressure PDS and heat exchanger 57 supplies heat to vaporize the main fuel into a gas state or supercritical state. In someembodiments, cryogenic pump 56 is located within main-fuel supply 23. In some embodiments, heat exchanger 57 is integrated within cryogenic pump 56. Regulator 50 down regulates main delivery pressure PMD to main rail pressure PMR. In the illustrated embodiment, pilot rail pressure PPR in pilot rail 160 is regulated from pilot delivery pressure PPD based on main rail pressure PMR (pilot follows main). In other embodiments, main rail pressure PMR in main rail 170 is regulated from main delivery pressure PMD based on pilot rail pressure PPR (main follows pilot).

[0043] Internal combustion engines 10, 11, 12 and 13 can be operated in a variety of combustion modes that each bum fuel in combustion chamber 420 in diverse ways. With reference to FIG. 6, fuel-injection mass-flow for a late-cycle, direct injection (LCDI) operating mode is illustrated where the fuel injected into combustion chamber 420 substantially bums with diffusion-flame combustion. Pilot-injection mass flow 500 represents a nominal mass flow of pilot fuel injected into combustion chamber 420 and main-injection mass flow 505 represents a nominal mass flow of main fuel injected into combustion chamber 420 in the LCDI operating mode. Pilot-injection mass flow 500 and main-injection mass flow 505 begin later during the compression stroke such that the pilot fuel and the main fuel do not have time to premix and thereby substantially bum only with diffusion combustion. In the illustrated embodiment, pilot-injection mass flow 500 begins and completes during the compression stroke, before main-injection mass flow 505, which also begins during the compression stroke and can end during the expansion stroke (however, this is not a requirement). In other embodiments, pilot-injection mass flow 500 can be overlapped with main-injection mass flow 505 and main-injection mass flow 505 can end during the compression stroke, provided in all cases the injected pilot fuel and main fuel substantially bum by diffusion-flame combustion. In the LCDI operating mode, the end of injection of the main fuel is typically later than start of ignition of the main fuel. A main injection window (as measured in crank angle degrees) during the LCDI operating mode where main-injection mass flow 505 can begin and end is between 30 crank angle degrees (CA°) before top dead center (BTDC) during the compression stroke and 40 C A° after top dead center (ATDC) during the expansion stroke. An injection pressure for the main fuel during the LCDI operating mode can be at least 250 bar and as high as the rated fill pressure of main-fuel supply 20. Exemplary injection pressures for the main fuel are substantially 300 bar and substantially 500 bar. A global fuel / air equivalence ratio in combustion chamber 420 can be less than or equal to 0.75 in the LCDI operating mode where both the main fuel and the pilot fuel are included in the determination. Combustion in the LCDI operating mode is substantially diffusion-flame combustionsuch that the global fuel / air equivalence ratio is lean to achieve good mixing quality for diffusionflame combustion. Pilot ignition window 510 illustrates the range of crank angle degrees in which ignition of pilot fuel can occur with reference to pilot-injection mass flow 500, in the illustrated embodiment. Pilot ignition window 510 during the LCDI operating mode can be between 15 CA° BTDC in the compression stroke and 15 CA° ATDC in the expansion stroke.

[0044] Referring now to FIG. 7, a mid-cycle, direct injection (MCDI) operating mode is illustrated where at least the main fuel injected into combustion chamber 420 has some time to premix such that it forms a partially-premixed main-fuel / air mixture that bums with both diffusion combustion and premixed-flame combustion, which is referred to collectively as partially -premixed combustion. In the MCDI operating mode, there are regions in combustion chamber 420 where a combustion rate is limited by a rate of diffusion and other regions where the combustion rate is controlled by the propagation of the flame. Main-injection mass flow 515 represents a nominal mass flow of main fuel injected into combustion chamber 420, and pilot-injection mass flow 520 represents a nominal mass flow of pilot fuel inj ected into combustion chamber 420 in the MCDI operating mode. Main-injection mass flow 515 begins earlier during the compression stroke compared to both pilotinjection mass flow 500 and main-injection mass flow 505 in the LCDI operating mode of FIG. 6, which allows the gaseous fuel injected due to main-injection mass flow 515 to partially premix with air. Pilot-injection mass flow 520 occurs later during the compression stroke near the time at which start of ignition of the partially-premixed main-fuel / air mixture is desired. A main injection window during the MCDI operating mode where main-injection mass flow 515 can begin and end is between 120 CA° BTDC during the compression stroke and 10 CA° BTDC during the compression stroke. End of injection of the main fuel typically occurs before start of ignition of the main fuel in the MCDI operating mode. An injection pressure for the main fuel during the MCDI operating mode can be at least 100 bar, and although the injection pressure during the MCDI operating mode can be as high as the rated fdl pressure of main-fuel supply 20 it typically isn’t higher than a lower value (or a minimum value) of the main injection pressure during the LCDI operating mode since the main fuel is injected earlier during the compression stroke in the MCDI operating mode than in the LCDI operating mode whereby cylinder pressure during injection is less. An exemplary injection pressure range for the main fuel during the MCDI operating mode is between 100 bar and 200bar. In an exemplary embodiment, when the injection pressure is 290 bar during the LCDI operating mode, the injection pressure during the MCDI operating mode can be 140 bar whereby engines 10, 11, 12, and 13 can beoperated at full engine load in the MCDI operating mode with little even negligible penalty on peak thermal efficiency. During cold start of engines 10, 11, 12, and 13, the injection pressure during the MCDI operating mode can be reduced below 100 bar, for example between 50 bar and 100 bar. As will be explained in more detail below, the MCDI operating mode can be employed during low load conditions during cold start. In an exemplary embodiment, at low load conditions, such as less than 10% of a maximum rated load, the injection pressure can be 60 bar during the MCDI operating mode when employed in cold start. A global fuel / air equivalence ratio in combustion chamber 420 can be less than or equal to 0.75 in the MCDI operating mode where both the main fuel and the pilot fuel are included in the determination. Combustion in the MCDI operating mode is partially premixed such that it operates with a lean value of the global fuel / air equivalence ratio to achieve improved combustion efficiency. Pilot ignition window 525 illustrates the range of crank angle degrees in which ignition of pilot fuel can occur with reference to pilot-injection mass flow 520, in the illustrated embodiment. Pilot ignition window 525 during the MCDI operating mode can be between 15 CA° BTDC in the compression stroke and 15 CA° ATDC in the expansion stroke.

[0045] In some embodiments, alternative forced ignition techniques can be employed instead of pilot fuel, in which embodiments pilot-fuel system 18 is not required. In some embodiments, different types of forced ignition techniques can be employed where which forced ignition technique is selected to ignite the main fuel can depend upon engine operating conditions. For example, in some embodiments, a spark igniter can be employed. In some embodiments, a heated surface, such as a glow plug, can be employed. In some embodiments, at least two of the pilot fuel, the spark igniter, and the heated surface is combined. Any type of forced ignition can be employed in the LCDI operating mode and the MCDI operating mode. That is, the LCDI operating mode is primarily characterized by the injection timing and the combustion mode for the main fuel and the timing of a forced ignition event. In this regard, the pilot ignition windows 510 and 525 can be referred to as forced-ignition ignition windows, within which the forced ignition event can occur, whether by the spark igniter, the heated surface, of the pilot fuel. In some embodiments, when the forced ignition technique is the spark igniter, the forced ignition windows 510 and 525 can be advanced compared to that employed for the pilot fuel since an ignition kernel resulting from the spark igniter may require more time to develop in order to ignite the main fuel. In some embodiments, when the forced ignition technique is the heated surface, the forced ignition windows 510 and 525 can also be advanced when the ignition kernel requires more time to develop compared to the pilot fuel ignition kernel.

[0046] Referring now to FIG. 8, there is shown a flow chart view of an algorithm 600 for emission reduction performed after starting internal combustion engines 10, 11, 12, or 13. An SCR temperature TSCR of the selective catalytic reduction converter in engine aftertreatment system 330 (seen FIGS.1, 3, 4, and 5) is determined in step 610. The SCR temperature TSCR is compared to a dosing temperature TDOSING in step 620. The dosing temperature TDOSING is the minimum temperature required for urea dosing such that when the SCR temperature TSCR is less than the dosing temperature TDOSING there can be fouling due to crystallization of solid urea as well as by products of incomplete urea decomposition and / or significantly degraded conversion efficiency of NOx within the SCR converter of engine aftertreatment system 330. The dosing temperature TDOSING is dependent on the SCR converter employed and in some embodiments is within a range of 170 to 250 degrees Celsius (°C). Algorithm 600 proceeds to step 630 where regular engine operation is selected when the SCR temperature TSCR is greater than or equal to the dosing temperature TDOSING, and algorithm 600 proceeds to step 640 where an emission reduction technique is selected when the SCR temperature TSCR is less than the dosing temperature TDOSING. Regular engine operation includes employing engine operating parameters (including but not limited to the LCDI operating mode, main rail pressure PMR, and settings for EGR) calibrated for the SCR converter that is warmed up and above the dosing temperature TDOSING, and these calibrations are applied to the internal combustion engines 10, 11, 12, or 13 in step 650. Returning to step 640 when the emission reduction technique is selected, algorithm 600 determines whether the engine is operating under a high toad when the SCR temperature TSCR is less than the dosing temperature TDOSING. In some embodiments, high toad is defined as operating at or above 50% of a maximum rated load for the internal combustion engines 10, 11, 12, or 13. In some embodiments, high load is when the indicated mean effective pressure (IMEP) is greater than 15 bar. The LCDI operating mode with reduced main rail pressure PMR and late main-fuel injection timing is selected in step 660 when the internal combustion engines 10, 11, 12, or 13 are operating at high toad, and the MCDI operating mode is selected in step 670 when the internal combustion engines 10, 11, 12, or 13 are not operating at high toad. Reducing main rail pressure PMR to a tower value compared to the main rail pressure PMR during operation of the internal combustion engines 10, 11, 12, or 13 at the current load when the SCR temperature TSCR of the selective catalytic reduction converter is greater than the dosing temperature has a small to moderate negative impact on thermal efficiency (the impact is more significant under higher toad conditions at a given EGR level) but is effective in reducing NOx emissions. In some embodiments, the main rail injection pressure PMR can be reduced to between 90 bar and 240 bar. In some embodiments, whenthe load is around 10% of the maximum rated load, and a peak motoring pressure is between 50-60 bar, the lower value of the main rail pressure PMR can be as low as 90 bar, and as the load increases and / or a boost pressure increases, a minimum value of the lower value of the main rail pressure PMR increases. The peak motoring pressure is the peak cylinder pressure inside combustion chamber 420 during the compression stroke without combustion. A ratio between main rail pressure PMR over a peak cylinder pressure Ppc(combustion) in combustion chamber 420 is above a predefined value such that the main fuel is mixed well during the LCDI operating mode during cold start. The peak cylinder pressure Ppc(combustion) is the peak pressure with combustion that occurs in combustion chamber 420. In some embodiments, the PMR / Ppc(combustion) ratio is greater than 1.4 when natural gas is the main fuel and greater than 1.25 when hydrogen is the main fuel. Alternatively, a ratio between main rail pressure PMR over a peak motoring pressure PPM in combustion chamber 420 is above a predefined value such that the main fuel is mixed well during the LCDI operating mode during cold start. The peak motoring pressure PPM is the peak pressure that occurs in combustion chamber 420 without combustion, for example due to compression work of piston 400 (seen in FIG. 2) alone. In some embodiments, the PMR / PPM ratio is greater than 2.1 when natural gas is the main fuel and greater than 1.6 when hydrogen is the main fuel. The injection timing is delayed during cold start compared to normal calibration, yet the injection timing is not delayed too late such that exhaust temperature TEXHAUST rises above an exhaust temperature limit that protects components in the exhaust path. In some embodiments, the main fuel injection timing is delayed such that the exhaust temperature TEXHAUST is at the predefined temperature limit. The MCDI operating mode can achieve comparable thermal efficiency to that of the LCDI operating mode yet the engine-out NOx emissions for the MCDI operating mode is significantly lower than that of the LCDI operating mode due to the lower flame temperature in a very lean, partially premixed fuel-air mixture. The MCDI operating mode is more sensitive to knock at higher loads, which is a reason for constraining the use of the MCDI operating mode for lower load levels. With reference to FIGS. 6 and 7, pulse separation PSEP is the difference in time between a beginning of main-injection mass flow 515 and a beginning of pilot-injection mass flow 520. The pulse separation PSEP is positive when the beginning of the pilotinjection mass flow precedes the beginning of the main-injection mass flow, as shown in FIG. 6, which is referred to as positive pulse separation. The pulse separation PSEP is negative when the beginning of the pilot-injection mass flow follows the beginning of the main-injection mass flow, as shown in FIG. 7, which is referred to as negative pulse separation. There is a limit to the negative pulse separation PSEP after which the efficiency begins to suffer due to delayed ignition timing. Insome embodiments, the negative pulse separation PSEP is within a range of -1500 microseconds and -500 microseconds during step 670, particularly for very low load operation, such a less than 25% of the maximum rated load. A high EGR setting is selected in step 680 when the SCR temperature TSCR is less than the dosing temperature TDOSING regardless of the engine load. Step 680 is not required, but when employed reduces NOx emissions even further compared to employing step 660 or step 670 alone. The high EGR setting is defined herein to be an amount of recirculated exhaust gas that is at least 15% of total exhaust gas. That is, 15% of the exhaust gas from combustion chamber 420 is returned to combustion chamber 420 for the next combustion cycle, or when there are multiple combustion chambers 420, 15% of total exhaust gas is returned and divided substantially equally between all the combustion chambers 420 for respective next combustion cycles. The exhaust gas can be returned by an external EGR loop 300, internal EGR, or a combination of both. Flame temperature of the burning main fuel within combustion chamber 420 is reduced, by increasing the amount of exhaust gases that are recirculated, which decreases the formation of NOx. The emission reduction technique of steps 660, 680 and steps 670, 680 both reduce the formation of NOx during combustion. The calibrations selected in steps 660 and 680 or step 670 and 680 are applied to the internal combustion engines 10, 11, 12, or 13 in step 650. In some embodiments, instead of employing the MCDI operating mode in step 670, the LCDI operating mode with reduced main rail pressure PMR and late main-fuel injection timing can be selected when the internal combustion engines 10, 11, 12, or 13 are not operating at high load. Although the MCDI operating mode resulted in greater reduction in NOx, the LCDI operating mode with reduced main rail pressure PMR and late main-fuel injection timing also reduced NOx when the internal combustion engines 10, 11, 12, or 13 are not operating at high load.

[0047] Referring now to FIG. 9, there is shown algorithm 601 for emission reduction after starting internal combustion engines 10, 11, 12, or 13, which is similar to algorithm 600 and differences are discussed. In step 620, when it is determined that the SCR temperature TSCR is less than the dosing temperature TDOSING, algorithm proceeds to step 622 where further temperature comparisons are performed. In step 622, the SCR temperature TSCR is compared to a first threshold temperature TLOWI and a time-averaged exhaust temperature TE(AVERAGED) is compared to the dosing temperature TDOSING. The first threshold temperature TLOWI is a low temperature limit below the dosing temperature TDOSING, where in some embodiments the first threshold temperature TLOWI is less than 50 °C, and in other embodiments the first threshold temperature TLOWI is within a range of20 to 50 °C. In some embodiments, the time-averaged exhaust temperature TE(AVERAGED) is the average exhaust temperature since startup. In some embodiments, the time-averaged exhaust temperature TE(AVERAGED) is the average exhaust temperature over a predefined time window that is narrower than the total time since startup. From step 622, algorithm 601 proceeds to step 625 where a low temperature operating mode is entered when the SCR temperature TSCR is less than the first threshold temperature TLOWI and the time-averaged exhaust temperature TE(AVERAGED) is less than the dosing temperature TDOSING, and algorithm 601 proceeds to step 640 when the SCR temperature TSCR is greater than or equal to the first threshold temperature TLOWI or the time-averaged exhaust temperature TE(AVERAGED) is greater than or equal to the dosing temperature TDOSING. The low temperature operating mode is employed to achieve higher exhaust temperature and faster warmup of the SCR converter in aftertreatment system 330 (seen in FIGS. 1, 3, 4, and 5) by adjusting engine operating parameters, for example by increasing the fuel / air equivalence ratio of the fuel-air mixture within combustion chamber 420 (seen in FIG. 2) of internal combustion engines 10, 11, 12, and 13 (seen in FIGS. 1, 3, 4, and 5), which also reduces NOx formation directly. If the time-averaged exhaust temperature TE(AVERAGED) is lower than the dosing temperature TDOSING for the SCR (e.g. 200 degree C) and the SCR temperature TSCR is lower than the first threshold temperature TLOWI (TLOWI < TDOSING), it can be deduced that the SCR temperature TSCR will not reach the dosing temperature TDOSING in a predefined time interval, whereby the low temperature mode in step 625 needs to be activated to speed up the warm-up. The predefined time interval can be based on an estimated amount of NOx emissions from startup to when the SCR temperature TSCR is projected to reach the dosing temperature TDOSING. In some embodiments, wastegate valve 270 is opened to reduce the boost pressure in intake manifold 280 and, accordingly, to reduce the mass of air introduced into combustion chamber 420. In some embodiments, throttle valve 260 is employed to decrease a mass flow of air from air intake 210 to intake manifold 280 and, accordingly, to decrease a mass of air in combustion chamber 420. In some embodiments, both wastegate valve 270 and throttle valve 260 are employed to increase the fuel / air equivalence ratio by decreasing the mass of air in combustion chamber 420. In some embodiments, intake valve timing can be adjusted whereby less air is admitted into combustion chamber 420 after intake valves 350 close. For example, intake valve 350 can be closed earlier during the intake stroke or closed later during the compression stroke to reduce the amount of air captured within combustion chamber 420. The LCDI and MCDI operating modes operate with a lean fuel / air equivalence ratio where there is significantly more air compared to the stoichiometric fuel / air equivalence ratio such that there is excess oxygen left over after combustionof the fuel. Since air is primarily composed of nitrogen (-78% molar fraction) and oxygen (-21% molar fraction), whenever the excess oxygen increases after combustion due to increased air mass admitted into the combustion chamber so too does the amount of nitrogen. The temperature of the excess oxygen and the other constituents of air that remain after combustion can increase more rapidly and to higher temperatures by admitting less air into combustion chamber 420 since there are less particles absorbing the thermal energy created by the combustion of fuel, thereby allowing the exhaust temperature to increase more rapidly and to higher temperatures. Higher exhaust temperatures result in faster SCR converter warmup in aftertreatment system 330 by increasing the rate of increase of the SCR temperature TSCR . The engine operating parameters associated with the low temperature operating mode are selected in step 625 and these calibrations are applied to the internal combustion engines 10, 11, 12, or 13 in step 650.

[0048] Referring now to FIGS. 10 through 18, engine test data was obtained to quantify the improvements in reducing emissions when operating algorithm 601 while starting internal combustion engines 10, 11, 12, or 13. The tests were performed on a 13 liter, 6 cylinder engine where the main fuel was hydrogen unless stated otherwise. With reference to FIGS. 10 and 11, the effect of operating in the LCDI operating mode with low main rail pressure PMR and in the MCDI operating mode with different pulse separation timing PSEP is illustrated for an A10 operating condition where letter ‘A’ represents an engine speed of 1200 revolutions per minute (RPM) and the number ‘10’ represents an engine load that is 10% of the maximum rated load. Bars 700 and 705 illustrate the normalized brake thermal efficiency and normalized engine-out NOx, respectively for a baseline in the LCDI operating mode with normal main rail pressure PMR (for example, 295 bar) when the algorithm 601 is not used. Bars 710 and 715 illustrate the normalized brake thermal efficiency and normalized engine-out NOx, respectively when algorithm 601 is used and when operating in the LCDI operating mode with reduced main rail pressure PMR (for example, 120 bar) and late main-fuel injection timing (for example, a start of injection around 1.9 CAD BTDC), where the brake thermal efficiency is slightly reduced compared to the baseline, but the NOx emissions is more significantly reduced. Bars 720 and 725 illustrate the normalized brake thermal efficiency and normalized engine-out NOx, respectively when algorithm 601 is used and when operating in the MCDI operating mode with a pulse separation PSEP of 800 microseconds and with start of injection timing for the main-injection mass flow 515 (seen in FIG. 7) beginning at 21 CAD BTDC, where the brake thermal efficiency is effectively unchanged but the NOx emissions are significantly reduced compared to thebaseline. Bars 730 and 735 illustrate the normalized brake thermal efficiency and normalized engine-out NOx, respectively when algorithm 601 is used and when operating in the MCDI operating mode with a pulse separation PSEP of 800 microseconds, similar to bars 720 and 725, respectively, but where the start of injection timing for the main-injection mass flow 515 timing is shifted slightly earlier to begin at 23 CAD BTDC resulting in a slight improvement in the brake thermal efficiency compared to the baseline but where the NOx emissions increased compared to bar 725. Bars 740 and 745 illustrate the normalized brake thermal efficiency and normalized engine-out NOx, respectively when algorithm 601 is used and when operating in the MCDI operating mode with a pulse separation PSEP of 1200 microseconds and with start of injection timing for the main-injection mass flow 515 beginning at 26 CAD BTDC, where the NOx emissions is reduced compared to bars 725 and 735 but where noticeably the brake thermal efficiency is significantly reduced compared to bars 720 and 730. Increasing the pulse separation PSEP beyond a limit (which is dependent upon other operating parameters such as rail pressure, injection timing, compression ratio among other engine parameters) causes over leaning of the fuel to occur in the outer boundary of main-fuel penetration in the combustion chamber such that incomplete combustion occurs and unbumed fuel emissions increase. Bars 750 and 755 illustrate the normalized brake thermal efficiency and normalized engine-out NOx, respectively when algorithm 601 is used and when operating in the MCDI operating mode with a pulse separation PSEP of 800 microseconds and a start of injection timing for the main-injection mass flow 515 beginning at 3 CAD BTDC, and operating with natural gas, where there is a noticeable reduction in the brake thermal efficiency but a more significant reduction in the NOx emissions compared to the baseline case.

[0049] With reference to FIGS. 12 and 13, the effect on the normalized brake thermal efficiency and normalized NOx emissions, respectively when operating in the LCDI operating mode with different values for the main rail pressure PMR is illustrated for different operating conditions. The operating conditions include A10, A25, A50, and C50 where the letters ‘A’ and ‘C’ refer to 1200 RPM and 1600 RPM engine speed, respectively and where the numbers ‘10’, ‘25’, and ‘50’ refer to the percentage load of the maximum rated load. Lowering the main rail pressure PMR has a mildly to moderately negative impact on the brake thermal efficiency. The impact is more significant under higher load conditions at a given EGR level. Lowering the main rail pressure PMR is effective in reducing NOx emissions, where there is an inverse relationship between NOx emissions and the main rail pressure.

[0050] With reference to FIGS. 14, 15, and 16, the effect of wastegating and throttling on the normalized inlet manifold pressure (IMP), the normalized brake thermal efficiency, and the normalized exhaust temperature, respectively is illustrated. The engine was operated under the A10 and A25 operating conditions for each of the baseline, a wastegating operating mode, and a wastegating and throttling operating mode. As previously discussed, the baseline represents the LCDI operating mode with normal main rail pressure PMR (for example, 295 bar) when the algorithm 601 is not used. Wastegating represents employing wastegate valve 270 to bypass exhaust around turbine 230. Throttling represents employing throtte valve 260 to reduce the mass flow of air into combustion chamber 420. Wastegating the exhaust and throttling the intake reduces the mass of intake air and increases the exhaust temperature, which helps reduce the warm-up period. Inlet manifold pressure (IMP) is decreased with opening of wastegate 270 and closing (at least partially) throttle valve 260 can result in significant increase in exhaust temperature. There is a negative impact on brake thermal efficiency when IMP is reduced by wastegating and throttling.

[0051] With reference to FIGS. 17 and 18, the effect of uncooled and cooled EGR on the normalized brake thermal efficiency and normalized NOx emissions is illustrated. The impact on the normalized BTE is similar at A10, but cooled EGR has less impact on brake thermal efficiency than uncooled EGR at A25. At the low load, the fraction of NOx emission reduction with uncooled EGR is similar to that of cooled EGR. Either uncooled or cooled EGR system when available on the engine can be used with the cold start algorithms 600 and 601 disclosed above.

[0052] While particular elements, embodiments and applications of the present invention have been shown and described, it will be understood that the invention is not limited thereto since modifications can be made by those skilled in the art without departing from the scope of the present disclosure, particularly in light of the foregoing teachings.

Claims

We claim:

1. A method of reducing emissions in an internal combustion engine, the method comprising:determining whether a temperature of an aftertreatment device is less than a dosing temperature;when the temperature of the aftertreatment device is less than the dosing temperature, determining whether a load on the internal combustion engine is above a load threshold;when the load is below the load threshold,1) injecting fuel with an early timing calibration whereby the fuel bums in a partially-premixed combustion mode; or2) decreasing a fuel injection pressure to a lower value compared to the fuel injection pressure during operation of the engine at the load when the temperature of the aftertreatment device is greater than the dosing temperature; and injecting the fuel with a delayed injection timing at the lower value of the fuel injection pressure, whereby the fuel bums in a diffusion-flame combustion mode;when the load is at or above the load threshold,decreasing the fuel injection pressure to the lower value compared to the fuel injection pressure during operation of the engine at the load when the temperature of the aftertreatment device is greater than the dosing temperature; andinjecting the fuel with a delayed injection timing at the lower value of the fuel injection pressure, whereby the fuel bums in a diffusion-flame combustion mode;wherein oxides of nitrogen (NOx) emissions are reduced.

2. The method as claimed in claim 1, wherein the load threshold is 50% of a maximum rated load.

3. The method as claimed in claim 1 or 2, wherein the fuel is a main fuel ignited by a pilot fuel.

4. The method as claimed in any one of claims 1 to 3, wherein the main fuel is injected between 120 crank angle degrees before top dead center during a compression stroke and 10 crank angle degrees before top dead center during the compression stroke when burned in the partially-premixed combustion mode, and the main fuel is injected between 30 crank angle degrees before top dead center during a compression stroke and 40 crank angle degrees after top dead center during a power stroke when burned in the diffusion-flame combustion mode.

5. The method as claimed in claim 1, wherein the injection timing is delayed such that an exhaust temperature is at an exhaust temperature limit.

6. The method as claimed in any one of claims 3-5, wherein a main fuel comprises hydrogen and a pilot fuel comprises diesel, dimethyl ether, or kerosene.

7. The method as claimed in claim 1, further comprisingdetermining whether a temperature of the aftertreatment device is less than a first low temperature threshold;determining whether a time-averaged exhaust temperature is less than the dosing temperature;when the temperature of the aftertreatment device is less than the first low temperature threshold and the time-averaged exhaust temperature is less than the dosing temperature,decreasing an amount of air within a combustion chamber in the internal combustion engine after an intake valve is closed for respective engine load and engine speed conditions such that a fuel / air equivalence ratio is increased;wherein an exhaust temperature increases causing an increase in a rate of rise of the temperature of the aftertreatment device.

8. The method as claimed in claim 7, further comprising actuating a throttle valve to decrease the amount of air within the combustion chamber.

9. The method as claimed in claim 7, further comprising actuating a wastegate valve to decrease the amount of air within the combustion chamber by decreasing a boost pressure.

10. The method as claimed in any one of claims 1 to 9, further comprising returning at least a portion of exhaust to a combustion chamber of the internal combustion engine for a subsequent engine cycle.

11. The method as claimed in claim 10, wherein the exhaust is returned by an external exhaust gas recirculation loop.

12. The method as claimed in claim 10, wherein the exhaust is returned by internal exhaust gas recirculation.

13. The method as claimed in claim 10, wherein an amount of recirculated exhaust gas is at least 15% of total exhaust gas.

14. The method as claimed in claim 1, further comprisingwhen the temperature of the aftertreatment device increases to or above the dosing temperature,increasing the fuel injection pressure; andinjecting the fuel without the delayed timing calibration at the increased fuel injection pressure;wherein the fuel bums in the diffusion-flame combustion mode.

15. The method as claimed in claim 14, further comprising injecting a diesel exhaust fluid into an exhaust upstream of the aftertreatment device wherein oxides of nitrogen (NOx) emissions are reduced.

16. The method as claimed in claim 15, wherein the diesel exhaust fluid is urea.

17. The method as claimed in claim 15 or 16, wherein the aftertreatment device comprises a catalytic converter.

18. The method as claimed in claim 17, wherein the catalytic converter is a selective catalytic reduction converter.

19. The method as claimed in claim 1, wherein a minimum value of the lower value of the fuel injection pressure increases with at least one of the load and a boost pressure.

20. The method as claimed in claim 1, wherein the fuel is ignited by at least one of a pilot fuel, a spark igniter, and a heated surface.

Images