Method and system for cylinder imbalance estimation

Abstract

Methods and systems are provided for learning cylinder-to-cylinder air variation. During a condition in which a PFDI engine is operating in port-fuel-injection-only mode, direct-injection fuel rail pressure can be reduced via direct injection prior to port fuel injection. Then, a direct injector can be temporarily opened prior to a spark event in a port-injected cylinder to use a rail pressure sensor to estimate cylinder compression pressure and infer cylinder air charge from it.

Description

TECHNICAL FIELD

[0001] The present specification generally relates to methods and systems for controlling a vehicle engine to monitor cylinder-to-cylinder imbalance of air-to-fuel ratio.

[0002] BACKGROUND / SUMMARY

[0003] Engine parameters such as air-to-fuel ratio (AFR) can be controlled to ensure improved engine performance, to effectively use exhaust catalysts, and to reduce exhaust emissions. In particular, cylinder-to-cylinder imbalance of air-to-fuel ratio can result in inefficient engine operation and increased engine-out emissions. In addition, torque imbalance can exist between engine cylinders, which can result in noise, vibration, and harshness (NVH) issues.

[0004] One way to determine AFR variation between engine cylinders is to sense engine exhaust via an oxygen sensor located downstream of an exhaust catalyst. By measuring exhaust components, it can be determined whether a given cylinder is operating with a richer or leaner air-to-fuel ratio than other cylinders. Fuel and / or charge parameters can then be adjusted based on the variations to produce an air-fuel mixture with a target air-to-fuel ratio. However, the oxygen sensor can be exposed to exhaust gas that is a combination of gases from different engine cylinders. It can therefore be difficult to accurately determine air-fuel variations between different engine cylinders. In addition, engine exhaust system geometry for engines with many cylinders can bias sensor readings more toward the output of one cylinder than others. It can therefore be even more difficult to determine air-fuel imbalance for engines with more than a few cylinders. Additional methods can include monitoring torque pulses on the crankshaft (or monitoring crankshaft acceleration at the desired AFR), and deriving a correlation between torque amplitude and combustion air-to-fuel ratio. However, in all of these methods, it can be difficult to distinguish between the air component of the error and the fuel component of the error.

[0005] Gottschalk et al. show one exemplary method for learning the air-based error in US 9,470,159. Therein, a direct fuel injector is actuated open to deliver fuel into a cylinder. While the injector is open and in use, and with a transfer function, the drop in direct injection fuel line pressure is measured to estimate the air charge in the cylinder. By comparing the air charge of each cylinder estimated in this way, the air component of the cylinder-to-cylinder AFR or torque variation can be learned.

[0006] However, the inventors have also recognized potential problems with this approach here. As one example, the estimation can be limited by the resolution range of the fuel rail pressure sensor. For example, at low engine loads, when the fuel rail pressure is low, the drop in fuel rail pressure can not be enough to be reliably measured by the sensor. As another example, the measured drop in fuel rail pressure can be affected by the position of the piston in the cylinder, specifically, based on whether the piston is at top dead center (TDC) or bottom dead center (BDC) of the compression stroke. As yet another example, it can be difficult to distinguish between a drop in fuel rail pressure due to a fuel-based error from a drop due to an air-based error.

[0007] Furthermore, exhaust gas recirculation (EGR) flow can corrupt the fuel pressure sensor output and the air flow estimation based on the fuel pressure sensor output. In particular, based on the configuration of the intake manifold and the location of intake where EGR is received, different cylinders can obtain different EGR flow, affecting the individual cylinder air charge estimation.

[0008] The inventors have recognized the above shortcomings and have developed a method for determining air-fuel ratio imbalance in engine cylinders and air-based errors that takes into account AFR variation between cylinder groups. In one example, AFR imbalance can be determined by a method for an engine, the method comprising: injecting fuel from a direct injector with a high pressure pump disabled to reduce direct injector rail pressure below a threshold pressure; then, injecting fuel into a cylinder and commanding the direct injector to selectively open for a threshold duration before a spark event in the cylinder without injecting any fuel from the direct injector. In this way, the air component of cylinder AFR variation can be accurately learned and reliably distinguished from the fuel component of AFR variation.

[0009] As one example, when operating a port fuel injection (PFI) engine in PFI only mode, the engine controller can estimate the compression pressure of the cylinders via a pressure sensor coupled to a high pressure direct injection (DI) fuel rail. The estimated compression pressure can then be used to infer the air charge of the cylinders. Specifically, the controller can disable a high pressure pump (HPP) coupled to the DI fuel rail to bleed the high pressure fuel rail to a threshold pressure (e.g., to a lower threshold) prior to injecting fuel via the port injectors, prior to injecting fuel via the direct injectors. The port fuel injection can then be enabled immediately prior to a spark being delivered to the cylinder, the DI can be commanded to open for a defined (short) duration. When the direct injectors are open, the high pressure fuel rail can be temporarily coupled to the cylinder, allowing the compression pressure in the cylinder to be estimated via the pressure sensor coupled to the high pressure fuel rail. In particular, the compression pressure can be marked as a transient spike in the fuel rail pressure. Since the compression pressure is directly related to the cylinder volume and the amount of air drawn into each cylinder, the spike in the fuel rail pressure can be correlated to the air charge in that cylinder. By continuing this operation until the air charge in each cylinder is estimated, and by repeating this operation several times for each cylinder, a stable average pressure for each cylinder can be obtained. By comparing the values for each cylinder, the air component of the cylinder to cylinder AFR variation can be learned. By performing the estimation when EGR flow is enabled, and when EGR flow is disabled, the noise impact of EGR on the air based error estimate can be quantified and compensated. Subsequently, the fuel component of the AFR variation can be estimated using the fuel rail pressure. Therein, the HPP can be actuated to raise the DI fuel rail pressure to a threshold (e.g., an upper threshold) prior to enabling direct injection of fuel into the cylinder, and the drop in fuel rail pressure after each injection pulse can be correlated to the commanded pulse width of each pulse.

[0010] In this way, the method provides improved learning of air-fuel ratio imbalances. The technical effect of measuring the cylinder compression pressure to estimate the cylinder air charge is that the air based error between cylinder groups can be more accurately learned, and the fuel based error can be more accurately distinguished. By measuring the rise in DI fuel rail pressure with only port injection refueling the cylinders, the effect of the compression pressure on the fuel rail pressure for the cylinders can be learned in the stable region of the fuel rail pressure sensor over a wider range of engine loads, including low engine loads. Thus, the method ensures that fuel efficiency is improved and emissions are reduced. Furthermore, the method can compensate for air-fuel ratio imbalances associated with EGR flow, enabling learning to be performed over a wider range of engine operating conditions without compromising EGR usage. By learning the air based error between cylinder groups, the AFR error can be better learned and compensated for.

[0011] It is to be understood that the foregoing summary of the application is intended merely to introduce a series of concepts rather than to define the scope of the claimed subject matter. The following detailed description of the application is provided as a more complete description of the various conceptsapplicable to achieving the implementations thereof. Additionally, the claimed subject matter is not limited to implementations solving any or all of the disadvantages of the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0012] Figure 1 is a diagram of an engine having cylinders.

[0013] Figure 2 shows a schematic diagram of a dual injector single fuel system coupled to Figure 1 an engine.

[0014] Figure 3 shows a high level flowchart of an exemplary method for estimating the air component of cylinder to cylinder air-fuel ratio variation.

[0015] Figure 4 shows a high level flowchart of an exemplary method for estimating the fuel component of cylinder to cylinder air-fuel ratio variation.

[0016] Figure 5 depicts the timing of intake port injector and direct injector operation relative to cylinder valve and spark events during estimation of cylinder air error.

[0017] Figure 6 depicts a predictive example of estimating cylinder to cylinder air fuel error, including determining the error of air and fuel components. DETAILED DESCRIPTION

[0018] The following description relates to systems and methods for air fuel error estimation in an engine system, such as the engine system of Figure 1 configured for intake port injection and direct injection, as shown in the fuel system of Figure 2 An engine controller can be configured to execute a control routine, such as the control routine of Figures 3-4 for detecting and distinguishing between the air component of cylinder to cylinder air-fuel ratio variation and the fuel component of the variation. The controller can adjust the timing of direct injector opening during the compression stroke of a combustion event, as shown in Figure 5 to estimate cylinder compression pressure using a fuel rail pressure sensor, and to infer cylinder air charge based on the estimated pressure. Reference is made to Figure 6 shows an example of air and fuel error estimates.

[0019] Figure 1An exemplary embodiment of a combustion chamber (or cylinder) 14 of an internal combustion engine 10 is depicted. The engine 10 can be coupled in a propulsion system configured for travel on a road, such as a vehicle 5.

[0020] The engine 10 can be controlled at least in part by a control system, including a controller 12, and by input 130 from a driver of the vehicle via an input device 132. In this example, the input device 132 includes an accelerator pedal and a pedal position sensor 134 for generating a proportional pedal position signal PP. The cylinder (also referred to herein as a "combustion chamber") 14 of the engine 10 can include a combustion chamber wall 136 in which a piston 138 is located. The piston 138 can be coupled to a crankshaft 140 such that reciprocating motion of the piston is translated into rotational motion of the crankshaft. The crankshaft 140 can be coupled to at least one drive wheel 55 of a passenger vehicle via a transmission 54. Additionally, a starter motor (not shown) can be coupled to the crankshaft 140 via a flywheel to enable starting operation of the engine 10.

[0021] In some examples, the vehicle 5 can be a hybrid vehicle having multiple sources of torque available for one or more wheels 55. In other examples, the vehicle 5 is a conventional vehicle having only an engine or an electric vehicle having only one or more electric machines. In the example shown, the vehicle 5 includes the engine 10 and an electric machine 52. The electric machine 52 can be a motor or a motor / generator. When one or more clutches 56 are engaged, the crankshaft 140 of the engine 10 and the electric machine 52 are connected to the wheels 55 via the transmission 54. In the depicted example, a first clutch 56 is disposed between the crankshaft 140 and the electric machine 52, and a second clutch 56 is disposed between the electric machine 52 and the transmission 54. The controller 12 can send signals to actuators of each clutch 56 to engage or disengage the clutch in order to connect or disconnect the crankshaft 140 from the electric machine 52 and components connected thereto, and / or to connect or disconnect the electric machine 52 from the transmission 54 and components connected thereto. The transmission 54 can be a gearbox, a planetary gear system, or another type of transmission. The powertrain can be configured in various ways, including as a parallel, series, or series-parallel hybrid vehicle.

[0022] The electric machine 52 receives electrical power from traction battery 58 to provide torque to the wheels 55. The electric machine 52 can also operate as a generator to provide electrical power to the charging battery 58, for example, during braking operations.

[0023] The cylinder 14 of the engine 10 can receive intake air via a series of intake passages 142, 144, and 146. The intake passage 146 can be in communication with other cylinders of the engine 10 other than the cylinder 14. In some examples, one or more of the intake passages can include a supercharging device, such as a turbocharger or a mechanical supercharger. For example, Figure 1An engine 10 configured with a turbocharger is shown, including a compressor 174 disposed between intake passages 142 and 144, and an exhaust turbine 176 disposed along an exhaust passage 148. When the supercharging device is configured as a turbocharger, the compressor 174 can be at least partially powered by the exhaust turbine 176 via a shaft 180. However, in other examples, such as when the engine 10 is provided with a mechanical supercharger, the compressor 174 can be powered by a mechanical input from a motor or engine, and the exhaust turbine 176 can optionally be omitted.

[0024] A throttle 162 including a throttle plate 164 can be disposed in an engine intake passage for varying the flow rate and / or pressure of intake air provided to the engine cylinders. For example, the throttle 162 can be positioned downstream of the compressor 174, as shown, or alternatively upstream of the compressor 174. Figure 2

[0025] An exhaust passage 148 can receive exhaust gas from cylinders of the engine 10 other than the cylinder 14. An exhaust sensor 128 is shown coupled to the exhaust passage 148 upstream of an emission control device 178. The exhaust sensor 128 can be selected from a variety of suitable sensors for providing an indication of exhaust air-fuel ratio (AFR), such as a linear oxygen sensor or UEGO (universal or wide range exhaust gas oxygen sensor); a dual-cell oxygen sensor or EGO (as depicted); a HEGO (heated EGO sensor); or a NOx, HC, or CO sensor, for example. The emission control device 178 can be a three-way catalyst (TWC), a NOx trap, various other emission control devices, or combinations thereof.

[0026] Each cylinder of the engine 10 can include one or more intake valves and one or more exhaust valves. For example, the illustrated cylinder 14 includes at least one intake poppet valve 150 and at least one exhaust poppet valve 156 located at an upper region of the cylinder 14. In some examples, each cylinder of the engine 10 (including the cylinder 14) can include at least two intake poppet valves and at least two exhaust poppet valves located at an upper region of the cylinder. The intake valve 150 can be controlled by the controller 12 via an actuator 152. Similarly, the exhaust valve 156 can be controlled by the controller 12 via an actuator 154. The position of the intake valve 150 and the exhaust valve 156 can be determined by respective valve position sensors (not shown).

[0027] ​Under some conditions, the controller 12 can vary the signals provided to the actuators 152 and 154 to control the opening and closing of the respective intake and exhaust valves. The valve actuators can be of the electrically actuated valve actuation type, cam actuated type, or a combination thereof. The intake valve timing and the exhaust valve timing can be controlled simultaneously, or any of variable intake cam timing, variable exhaust cam timing, dual independent variable cam timing, or fixed cam timing can be used. Each cam actuation system can include one or more cams, and can utilize one or more of cam profile switching (CPS), variable cam timing (VCT), variable valve timing (VVT), and / or variable valve lift (VVL) systems, which can be operated by the controller 12 to vary the valve operation. For example, the cylinders 14 can optionally include an intake valve that is controlled via electrically actuated valve actuation and an exhaust valve that is controlled via cam actuation including CPS and / or VCT. In other examples, the intake and exhaust valves can be controlled by a common valve actuator (or actuation system) or variable valve timing actuator (or actuation system).

[0028] The cylinders 14 can have a compression ratio, which is the ratio of the volume when the piston 138 is at bottom dead center (BDC) to that at top dead center (TDC). In one example, the compression ratio is in the range of 9: 1 to 10: 1. However, in some examples using different fuels, the compression ratio can be increased. This can occur, for example, when using a higher octane fuel or a fuel with a higher latent heat of vaporization. The compression ratio can also be increased in cases where direct injection is used due to its effect on engine knock.

[0029] In some examples, each cylinder of the engine 10 can include a spark plug 192 for initiating combustion. In selected modes of operation, the ignition system 190 can provide an ignition spark to the combustion chamber 14 via the spark plug 192 in response to a spark advance signal SA from the controller 12. The timing of the signal SA can be adjusted based on engine operating conditions and driver torque demand. For example, the spark can be provided at a maximum brake torque (MBT) timing to maximize engine power and efficiency. The controller 12 can input engine operating conditions, including engine speed, engine load, and exhaust AFR, into a lookup table and output a corresponding MBT timing for the input engine operating conditions.

[0030] In some examples, each cylinder of engine 10 can be configured with one or more fuel injectors for providing fuel thereto. By way of non-limiting example, cylinder 14 is shown including two fuel injectors 166 and 170. Fuel injectors 166 and 170 can be configured to deliver fuel received from fuel system 8. Fuel system 8 can include one or more fuel tanks, fuel pumps, and a fuel rail. Fuel injector 166 is shown directly coupled to cylinder 14 for directly injecting fuel thereto in proportion to a pulse signal width FPW-1 received from controller 12 via electronic driver 168. In this manner, fuel injector 166 provides so-called direct fuel injection (hereinafter also referred to as “DI”) into cylinder 14. While Figure 1 Fuel injector 166 is shown positioned to one side of cylinder 14, but fuel injector 166 can alternatively be located at the top of the piston, such as proximate to spark plug 192. Such a location can increase mixing and combustion when operating the engine with alcohol-based fuels, as some alcohol-based fuels have lower volatility. Alternatively, the injector can be located at the top and proximate to the intake valve to increase mixing. Fuel can be delivered from a fuel tank of fuel system 8 to fuel injector 166 via a high pressure fuel pump and fuel rail. Further, the fuel tank can have a pressure sensor that provides a signal to controller 12.

[0031] Fuel injector 170 is shown arranged in intake passage 146 rather than directly coupled to cylinder 14, which configuration provides so-called port fuel injection (hereinafter also referred to as “PFI”) of fuel into the intake port upstream of cylinder 14. Fuel injector 170 can inject fuel received from fuel system 8 in proportion to a pulse signal width FPW-2 received from controller 12 via electronic driver 171. It should be noted that, instead of multiple electronic drivers (such as electronic driver 168 for fuel injector 166 and electronic driver 171 for fuel injector 170, as depicted), a single electronic driver can be used for both fuel injectors.

[0032] In alternative examples, each of fuel injectors 166 and 170 can be configured as direct fuel injectors for injecting fuel directly into cylinder 14. In yet another example, each of fuel injectors 166 and 170 can be configured as port fuel injectors for injecting fuel upstream of intake valve 150. In other examples, cylinder 14 can include only a single fuel injector configured to receive different fuels from fuel system as a fuel mixture in different relative amounts, and further configured to inject the fuel mixture directly into the cylinder as a direct fuel injector or upstream of the intake valve as a port fuel injector. Accordingly, it should be understood that the fuel systems described herein should not be limited by the particular fuel injector configurations described herein by way of example.

[0033] During a single cycle of the cylinder, fuel can be delivered to the cylinder 14 by both injectors. For example, each injector can deliver a portion of the total amount of fuel combusted in the cylinder 14. Additionally, the distribution and / or relative amount of fuel delivered by each injector can vary by operating conditions, such as engine load, knock, and exhaust gas temperature. The intake port injected fuel can be delivered during an open intake valve event, a closed intake valve event (e.g., substantially prior to the intake stroke), and during open and closed intake valve operation. Similarly, for example, the direct injected fuel can be delivered at least partially during a previous exhaust stroke, during an intake stroke, and during a compression stroke. Thus, even for a single combustion event, injected fuel can be injected from both the intake port injector and the direct injector at different times. Moreover, for a single combustion event, multiple injections of delivered fuel can be performed per cycle. The multiple injections can be performed during a compression stroke, an intake stroke, or any suitable combination thereof.

[0034] The fuel injectors 166 and 170 can have different characteristics. These different characteristics include size differences, such as one injector having a larger injection orifice than the other, for example. Other differences include, but are not limited to, different injection angles, different operating temperatures, different targets, different injection timing, different spray characteristics, different locations, etc. Moreover, depending on the distribution ratio of injected fuel among the injectors 170 and 166, different effects can be achieved.

[0035] Fuel can be delivered to the fuel injectors 166 and 170 by a high pressure fuel system (described in detail at Figure 2 , which includes a fuel tank, a fuel pump, and a fuel rail. Additionally, as Figure 2 shown, the fuel tank and the fuel rail can each have a pressure sensor that provides a signal to the controller 12.

[0036] The fuel tanks in the fuel system 8 can hold different fuel types of fuel, such as fuels having different fuel qualities and different fuel quantities. The differences can include different alcohol content, different water content, different octane ratings, different heat of vaporization, different fuel blends, and / or combinations thereof, among others. One example of fuels having different heat of vaporization includes gasoline with a lower heat of vaporization as a first fuel type and ethanol with a greater heat of vaporization as a second fuel type. In another example, the engine can use gasoline as a first fuel type and an alcohol-containing fuel blend as a second fuel type, such as E85 (approximately 85% ethanol and 15% gasoline) or M85 (approximately 85% methanol and 15% gasoline). Other possible substances include water, methanol, mixtures of alcohol and water, mixtures of water and methanol, mixtures of alcohol, among others. In yet another example, the two fuels can be alcohol blends having different alcohol quantities, where the first fuel type can be a gasoline alcohol blend with a lower concentration of alcohol, such as E10 (approximately 10% ethanol), and the second fuel type can be a gasoline alcohol blend with a higher concentration of alcohol, such as E85 (approximately 85% ethanol). In addition, the first and second fuels can also differ in other fuel qualities, such as differences in temperature, viscosity, octane rating, among others. Additionally, the fuel properties of one or both fuel tanks can change frequently, for example, on a daily basis due to refueling of the tank.

[0037] The air-fuel ratio error can be determined based on the output of the oxygen sensor 128. In addition to the air-fuel ratio error for a given cylinder, there can also be variations in air-fuel ratio, and thus torque output, between cylinders. This can be due to differences in the air charge received by the cylinder, such as due to inherent differences in air flow, which is attributed to the configuration / design of the intake manifold, length of flow passages, valve position, and location of each cylinder on the engine block. Additionally or alternatively, the variations can be due to differences in fuel received at the cylinder, such as due to inherent differences in injector nozzle shape and size, injector position, other injector differences, fuel rail pressure pulsations, among others. As detailed with reference to Figures 3-4 As detailed, the torque variations due to air quantity can be detected and distinguished from variations in fuel quantity, enabling each error to be properly addressed. In particular, under selected conditions, the compression pressure of the cylinder can be measured using a fuel rail pressure sensor coupled to the high pressure fuel rail of the direct injector (as detailed with reference to Figure 2 As detailed, the compression pressure of the cylinder can be measured using a fuel rail pressure sensor coupled to the high pressure fuel rail of the direct injector (as detailed with reference to

[0038] The controller 12 determines the air-fuel ratio error in the cylinder based on the output of the oxygen sensor 128, as detailed with reference to FIG. 2. The controller 12 determines the air-fuel ratio error in the cylinder based on the output of the oxygen sensor 128, as detailed with reference to FIG. 2. Figure 1The microcomputer shown includes: a microprocessor unit 106, an input / output port 108, an electronic storage medium (shown in this particular example as a non-transitory read-only memory chip 110) for executable programs (e.g., executable instructions) and calibration values, random access memory 112, keep-alive memory 114, and a data bus. The controller 12 can receive various signals from sensors coupled to the engine 10, including those previously discussed, and additionally including measurements from the intake mass airflow meter (MAF) of the mass airflow sensor 122; engine coolant temperature (ECT) from the temperature sensor 116 coupled to the cooling sleeve 118; exhaust temperature from the temperature sensor 158 coupled to the exhaust passage 148; surface ignition sensing signal (PIP) from the Hall effect sensor 120 (or other type) coupled to the crankshaft 140; throttle position (TP) from the throttle position sensor; and manifold absolute pressure signal (MAP) from the MAP sensor 124. The engine speed per minute (RPM) signal can be generated from the PIP signal via controller 12. The manifold pressure signal MAP from MAP sensor 124 can be used to provide an indication of vacuum or pressure in the intake manifold. Controller 12 can infer the engine temperature based on the engine coolant temperature. Controller 12 receives signals from... Figure 1 The signals from various sensors, and adopted Figure 1 Various actuators adjust engine operation based on received signals and instructions stored in the controller's memory. For example, in response to... Figure 3 Based on the indication of air error determined at the location, the controller can adjust the engine fuel supply to maintain a target air-fuel ratio. In one example, in response to an air error where more air than required is delivered to the engine cylinder, the controller can increase the pulse width of the fuel injected into that cylinder in order to maintain the combustion air-fuel ratio at or near the stoichiometric level.

[0039] As described above, Figure 1 Only one cylinder from a multi-cylinder engine is shown. Thus, each cylinder may similarly include its own set of intake / exhaust valves, fuel injectors, spark plugs, etc. It should be understood that engine 10 may include any suitable number of cylinders, including 2, 3, 4, 5, 6, 8, 10, 12, or more cylinders. Furthermore, each of these cylinders may include a reference cylinder 14 via... Figure 1 Some or all of the various components described and depicted.

[0040] Figure 2 A dual-injector single-fuel system 200 with high-pressure and low-pressure fuel rail systems was demonstrated. The fuel system 200 can be coupled to an engine, such as... Figure 1Engine 10, and is operable to deliver fuel to the engine. Fuel system 200 can be operated by a controller to perform the operations described with reference to Figures 3-4 Some or all of the operations described are performed by the previously introduced components, denoted by like reference numerals.

[0041] Fuel system 200 can include a fuel tank 210, a low pressure or lift pump 212 that supplies fuel from fuel tank 210 to a high pressure fuel pump 214. Lift pump 212 also supplies fuel at a lower pressure to a low pressure fuel rail 260 via a fuel passage 218 (also referred to herein as a fuel line 218). Thus, low pressure fuel rail 260 is exclusively coupled to lift pump 212. Fuel rail 260 supplies fuel to port fuel injectors 262a, 262b, 262c, and 262d. High pressure fuel pump 214 supplies pressurized fuel to a high pressure fuel rail 250. Thus, high pressure fuel rail 250 is coupled to each of high pressure pump 214 and lift pump 212.

[0042] Due to aging and wear and tear, as well as learned fuel component of injector to injector air-fuel ratio variation, fuel injectors can need to be calibrated intermittently for variability. As a result of the variation, the actual amount of fuel injected into each cylinder of the engine can not be the desired amount, and the deviation can result in decreased fuel economy, increased exhaust emissions, and overall decreased engine efficiency.

[0043] High pressure fuel rail 250 supplies pressurized fuel to direct fuel injectors 252a, 252b, 252c, and 252d. Fuel rail pressure in fuel rails 250 and 260 can be monitored by pressure sensors 248 and 258, respectively. In one example, lift pump 212 can be an electronic check valve pump system that can be operated intermittently in pulse form. In another example, lift pump 212 can be a turbine (e.g., centrifugal) pump that includes an electric (e.g., DC) pump motor, whereby the pressure increase across the pump and / or the volumetric flow through the pump can be controlled by varying the electrical power provided to the pump motor, thereby increasing or decreasing the motor speed. For example, when the controller decreases the electrical power provided to lift pump 212, the volumetric flow and / or the pressure increase across the lift pump can be decreased. The volumetric flow and / or the pressure increase across the pump can be increased by increasing the electrical power provided to lift pump 212. As one example, the electrical power provided to the lift pump motor can be obtained from an alternator or other energy storage device on the vehicle (not shown), whereby the control system can control the electrical load used to power lift pump 212. Thus, by varying the voltage and / or current provided to the lift pump, the flow and pressure of fuel provided at the inlet of HP fuel pump 214 is regulated.

[0044] The boost pump 212 can be equipped with a check valve 213 such that the fuel line 218 (or optional compliant element) remains pressurized while the boost pump 212 reduces its input energy to the extent that no flow past the check valve 213 occurs. The boost pump 212 can be fluidly coupled to a filter 217, which can remove small impurities contained in the fuel that can potentially damage fuel handling components. In the case that the check valve 213 is upstream of the filter 217, the compliance of the low pressure passage 218 can be increased because the volume of the filter can be physically large. In addition, a pressure relief valve 219 can be employed to limit the fuel pressure (e.g., output from the boost pump 212) in the low pressure passage 218. The pressure relief valve 219 can include, for example, a ball and spring mechanism seated and sealed at a specified differential pressure. The pressure relief valve 219 can be configured such that the differential pressure set point at which it opens can assume various suitable values; as non-limiting examples, the set point can be 6.4 bar or 5 bar (g). In some embodiments, the fuel system 200 can include one or more (e.g., a series of) check valves fluidly coupled to the low pressure fuel pump 212 to prevent fuel from leaking back upstream of the valve.

[0045] A boost pump fuel pressure sensor 231 can be positioned along the fuel passage 218 between the boost pump 212 and the HP fuel pump 214. In this configuration, readings from the sensor 231 can be interpreted as an indication of the fuel pressure of the boost pump 212 (e.g., the outlet fuel pressure of the boost pump) and / or the inlet pressure of the high pressure fuel pump. Readings from the sensor 231 can be used to assess the operation of various components in the fuel system 200 to determine whether sufficient fuel pressure is provided to the high pressure fuel pump 214 such that the high pressure fuel pump draws in liquid fuel rather than fuel vapor, and / or to minimize the average electrical power supplied to the boost pump 212.

[0046] A high pressure fuel rail 250 can be coupled to the outlet 208 of the high pressure fuel pump 214 along a fuel passage 278. A check valve 274 and a pressure relief valve 272 (also referred to as a pump safety valve) can be positioned between the outlet 208 of the high pressure fuel pump 214 and the high pressure fuel rail 250. The pump safety valve 272 can be coupled to a bypass passage 279 of the fuel passage 278. The outlet check valve 274 opens to allow fuel to flow from the high pressure pump outlet 208 into the fuel rail only when the pressure at the outlet of the direct injection fuel pump 214 (e.g., the compression chamber outlet pressure) is higher than the fuel rail pressure. The pump safety valve 272 can limit the pressure in the fuel passage 278, downstream of the high pressure fuel pump 214, and upstream of the high pressure fuel rail 250. For example, the pump safety valve 272 can limit the pressure in the fuel passage 278 to 200 bar. When the fuel rail pressure is greater than a predetermined pressure, the pump safety valve 272 allows fuel to flow out of the DI fuel rail 250 toward the pump outlet 208.

[0047] Attached at the inlet of the LP fuel rail is a check valve 244 for controlling the flow of fuel from the lift pump to the fuel rail and from the fuel rail to the lift pump. The pressure check valve 244 opens when the fuel pump delivers a predetermined pressure to the fuel line.

[0048] Direct fuel injectors 252a-252d and port fuel injectors 262a-262d inject fuel into engine cylinders 201a, 201b, 201c, and 201d located in the engine block 201, respectively. Thus, each cylinder can receive fuel from two injectors, with the two injectors placed in different locations. For example, as discussed previously in Figure 1 , one injector can be configured as a direct injector, coupled to directly supply fuel into the combustion chamber, while the other injector is configured as a port injector, coupled to the intake manifold and delivering fuel into the intake port upstream of the intake valve. Thus, cylinder 201a receives fuel from port injector 262a and direct injector 252a, while cylinder 201b receives fuel from port injector 262b and direct injector 252b.

[0049] While each of the high pressure fuel rail 250 and the low pressure fuel rail 260 is shown distributing fuel to four fuel injectors of the respective injector groups 252a-252d and 262a-262d, it should be understood that each fuel rail 250, 260 can distribute fuel to any suitable number of fuel injectors.

[0050] Similar to Figure 1 , the controller 12 can receive fuel pressure signals from fuel pressure sensors 258 and 248 coupled to the fuel rails 260 and 250, respectively. The fuel rails 260 and 250 can also include temperature sensors for sensing the temperature of the fuel within the fuel rails, such as sensors 202 and 203 coupled to the fuel rails 260 and 250, respectively. The controller 12 can also control the operation of the intake and / or exhaust valves or throttle, engine cooling fan, spark ignition, and fuel pumps 212 and 214 to control engine operating conditions.

[0051] The fuel pumps 212 and 214 can be controlled by the controller 12, as shown in Figure 2 . The controller 12 can regulate the amount or speed of fuel fed into the fuel rails 260 and 250 by the lift pump 212 and the high pressure fuel pump 214 through respective fuel pump controllers (not shown). The controller 12 can also completely stop the supply of fuel to the fuel rails 260 and 250 by shutting off the pumps 212 and 214.

[0052] The injectors 262a-262d and 252a-252d can be operably coupled to and controlled by the controller 12. The amount of fuel injected from each injector and the injection timing can be determined by the controller 12 according to engine characteristics curves stored in the controller 12 based on engine speed and / or intake throttle angle or engine load. Each injector can be controlled via a solenoid valve coupled to the injector (not shown). In one example, the controller 12 can individually actuate each of the intake port injectors 262 via an intake port injection driver 237 and each of the direct injectors 252 via a direct injection driver 238. The controller 12, drivers 237, 238, and other suitable engine system controllers can comprise a control system. While the drivers 237, 238 are shown external to the controller 12, it should be understood that in other examples, the controller 12 can include the drivers 237, 238 or can be configured to provide the functionality of the drivers 237, 238.

[0053] In one example, the amount of fuel delivered via the intake port injectors and the direct injectors is determined empirically and stored in a plurality of predetermined lookup tables or functions. For example, one table can correspond to determining the intake port injection amount and one table can correspond to determining the direct injection amount. Both tables can be indexed to engine operating conditions, such as engine speed and engine load and other engine operating conditions. Additionally, the tables can output an amount of fuel to be injected to the engine cylinders via intake port fuel injection and / or direct injection in each cylinder cycle.

[0054] Accordingly, fuel can be injected to the engine via the intake port injectors and the direct injectors or via the direct injectors only or via the intake port injectors only depending on the engine operating conditions. For example, the controller 12 can determine to deliver fuel to the engine via the intake port injectors and the direct injectors or via the direct injectors only or via the intake port injectors only based on the output from the predetermined lookup tables as described above.

[0055] Various modifications or adjustments can be made to the above example system. For example, the fuel passage 218 can include one or more filters, pressure sensors, temperature sensors, and / or pressure relief valves. The fuel passage can include one or more fuel cooling systems.

[0056] In this way, Figures 1-2The components of the disclosure implement an engine system that includes an engine including a cylinder; an intake port injector coupled to the cylinder; a direct injector coupled to the cylinder; a high pressure fuel pump to deliver fuel to the direct injector via a direct injection fuel rail; a pressure sensor to estimate a direct injection fuel rail pressure; and a controller. The engine system can also include a controller configured with computer readable instructions stored on a non-transitory memory to operate the direct injector with the fuel pump disabled until the fuel pump pressure falls below a first threshold pressure and then disable the direct injector; temporarily open the direct injector during a compression stroke of the cylinder but before a spark event of the cylinder without delivering any fuel; estimate a cylinder air charge based on a change in fuel rail pressure during the temporary opening; and adjust a subsequent cylinder fueling based on the estimated cylinder air charge. In one example, the temporary opening is performed for a predetermined number of injection events of the cylinder, where the estimated cylinder air charge is an average cylinder air charge averaged over the predetermined number of injection events, and where adjusting the subsequent cylinder fueling includes adjusting the subsequent cylinder fueling via one or more of the intake port injector and the direct injector. In another example, the cylinder can be one cylinder of a plurality of engine cylinders, where the fueling and the temporary opening are performed for each of the plurality of engine cylinders in a plurality of consecutive injection events of the cylinder, and where adjusting the subsequent cylinder fueling based on the estimated cylinder air charge includes adjusting the subsequent fueling for each engine cylinder based on the estimated cylinder air charge for the corresponding cylinder relative to an average cylinder air charge estimate averaged over the plurality of engine cylinders. Further, the temporary opening can be performed when the cylinder is fueled via only the intake port injector or during a deceleration fuel cut event.

[0057] Turning now to Figure 3 , an example method 300 for learning an air component of a cylinder-to-cylinder air-fuel ratio error is shown. The method implements a reduction in cylinder-to-cylinder torque variation by compensating for the learned air error, such as using a fueling regulator. The instructions for performing the method 300 and the remaining methods included herein can be performed by a controller based on instructions stored on a memory of the controller and in conjunction with signals received from sensors of the engine system, such as the sensors described above with reference to Figure 1 . In accordance with the methods described below, the controller can employ engine actuators of the engine system to adjust engine operation.

[0058] At 302, the method includes estimating and / or measuring engine operating conditions. For example, parameters such as engine speed, engine load, operator torque demand, boost pressure, engine dilution (e.g., EGR flow) ambient conditions (such as ambient temperature, barometric pressure, ambient temperature), etc. can be determined.

[0059] At 304, the method includes determining a fuel injection specification based on the estimated engine operating conditions. Determining the fuel injection specification can include determining whether fuel is delivered via intake port injection, direct injection, or a combination thereof. In addition, the amount of fuel, injection timing, number of injections per injection event, etc. can also be determined. For example, the engine controller can determine a fuel split ratio (including a ratio of intake port injected fuel to direct injected fuel) based on the engine speed / load condition. The controller can reference an engine speed / load characteristic curve stored in the controller's memory to determine the amount of fuel to inject, the type of fuel injection (or types), and the number of injections. In the case of direct injection, the controller can further determine a ratio of intake stroke direct injected fuel to compression stroke direct injected fuel. In one example, at lower engine speeds / loads and cooler engine conditions, the fuel injection specification can include delivering all of the injected fuel via a single intake port injection in the exhaust stroke or intake stroke. As another example, at higher engine speeds / loads and warmer engine conditions, the fuel injection specification can include delivering all of the injected fuel via multiple direct injections in the intake stroke and / or compression stroke. As yet another example, at mid-load, a portion of the fuel can be delivered via intake port injection and the remaining fuel can be delivered via direct injection (single or multiple).

[0060] At 306, it can be determined whether the fuel injection specification includes intake port only fuel injection (PFI only). If so, at 310, the method includes disabling the high pressure pump coupled to the direct injector via the direct injection fuel rail. The boost pump supplying fuel from the fuel tank to the high pressure pump and also to the intake port injector via the intake port fuel rail can continue to operate. The direct injection fuel rail can be a high pressure fuel rail, while the intake port fuel rail can be a low pressure fuel rail. Further, with the high pressure pump disabled, fuel can be injected from the direct injector to lower the direct injection fuel rail pressure below a threshold pressure. For example, the controller can pulse the direct injector (e.g., a single command or a command that is intermittently repeated) to enable the fuel rail pressure to bleed off. The direct injection to bleed off the fuel rail pressure can be an intake stroke direct injection. The injected fuel is then tied to the required fuel mass to achieve the required air-fuel ratio. For example, the direct injection to bleed off the fuel rail pressure can be compensated for via the intake port injection regulator, such as by providing the remaining portion of the required fuel mass via intake port injection, to maintain the target air-fuel ratio.

[0061] At 316, it can be determined whether the fuel rail pressure (HP FRP) of the high pressure fuel rail is below a threshold pressure. The threshold pressure can be determined as a function of atmospheric pressure, and in one example, can be 100 psi. The threshold pressure can be further calibrated as a function of engine speed and load, such that air error can be reliably estimated via changes in fuel rail pressure even during low load engine operation. In one example, the threshold pressure is a lower threshold below which actuation (or opening) of the direct injector can result in no fuel flowing from the injector into the corresponding cylinder. For example, the threshold pressure can be lowered below the expected compression pressure in the cylinder during the cylinder combustion event. Since the cylinder pressure near TDC before combustion is a direct function of load, as load increases, the resulting cylinder pressure will also increase. Thus, in another example, the controller can target the same fuel rail pressure or scale the pressure based on load (cylinder pressure) to maintain the same expected offset. For example, the controller can make a logical determination regarding the threshold pressure based on a logical rule, model, or algorithm that uses engine speed and load as inputs and produces the threshold pressure as an output. If the fuel rail pressure is not below the threshold pressure, then at 318, the method includes continuing to inject fuel via direct injection with the high pressure pump (HPP) disabled until the threshold pressure is reached. After the fuel rail pressure is lowered below the threshold pressure, the direct injector can be disabled. Then, at 320, the method includes injecting fuel port into the cylinder. In one example, injecting fuel port into the cylinder includes making a port injection during an intake stroke or an exhaust stroke (immediately preceding) of the cylinder. It will be appreciated that injecting fuel port into the cylinder includes not directly injecting fuel into the cylinder and maintaining the HPP disabled.

[0062] At 322, the method includes commanding the direct injector to selectively open a threshold duration prior to a spark event in the cylinder without injecting any fuel from the direct injector. In particular, the direct injector is commanded to open during a compression stroke of the cylinder. Opening a threshold duration prior to the spark event can include opening a threshold number of degrees of crankshaft rotation prior to the spark event. Additionally, the direct injector can remain open for a defined duration, such as a defined number of degrees of crankshaft rotation. The threshold duration that the DI is commanded to open prior to the spark event or the engine position at which the DI is commanded to open can be based on engine speed. In which, the number of degrees of crankshaft rotation that the DI is commanded to open is adjusted according to a speed function. In one example, the DI is commanded to open 5 degrees prior to the spark event and remain open for a few milliseconds until enough time has elapsed so that stable pressure measurements can be made. As another example, opening a threshold duration prior to the spark event can include opening at a predefined initial engine position and closing at a predefined final engine position. In one example, the DI is commanded to open at 15 degrees BTDC and remain open for a few milliseconds until 10 degrees BTDC. Further, the timing of the DI opening can vary based on engine speed to achieve spark tracking. For example, the timing of opening the DI can be adjusted based on engine speed so that the DI opening can be completed 5 degrees prior to the spark event. In another example, a minimum pulse width can be commanded to the direct injector. In yet another example, the pulse width commanded to the DI can be adjusted based on the range and sensitivity of the fuel rail pressure sensor so that the DI is open long enough for the sensor to detect a measurable change.

[0063] For example, the controller can make a logical determination (e.g., regarding the timing of the commanded DI opening) based on a logical rule as a function of engine speed and timing of the cylinder spark event. The controller can use a model, a lookup table, or an algorithm that uses engine speed as an input and produces as an output the engine position at which the DI will be commanded to open in CAD. The controller can then generate a control signal (such as a pulse width signal) that is sent to the fuel injector actuator to open the DI at the determined engine position. As a result of opening the DI prior to the spark event in the port-fuel-injected cylinder, the compression pressure of the cylinder can be measured by the direct injector fuel rail pressure sensor. As used herein, the compression pressure of the cylinder refers to the pressure in the cylinder during the compression stroke immediately prior to the combustion process. Since the combustion pressure is directly related to the cylinder volume and the amount of air drawn into the cylinder, by temporarily coupling the cylinder to the DI fuel rail via the opening of the DI, the existing DI fuel rail pressure sensor can be used to accurately estimate the cylinder air charge. As a result of temporarily opening the DI, the pressure in the DI fuel rail increases. In one example, the fuel rail pressure can rise from 100 psi to 150 psi. Thus, any air drawn into the fuel rail from the cylinder can dissolve with the fuel in the fuel rail. At 324, the rise in fuel rail pressure (HP FRP) is estimated via the DI fuel rail pressure sensor. Various engine conditions or events can affect the fuel rail pressure measurement, and can be taken into account when calculating the fuel pressure rise attributed to each DI opening event. Thus, in some examples, the program can correlate the fuel pressure to various engine conditions sensed via various sensors. For example, the transient pressure pulsations resulting from injector opening can temporarily affect the fuel rail pressure measurement, affecting the accuracy of the calibration. As such, the sampling of the fuel pressure can be selected to reduce the transient effects of the injector firing. Additionally or alternatively, if the injector firing timing is correlated to the fuel rail pressure measurement, the transient pressure changes caused by the injector firing can be taken into account when deriving the injector calibration values. Similarly, intake and / or exhaust valve opening and closing, intake and / or exhaust pressures, crank angle position, cam position, spark firing, and engine combustion can also affect the fuel rail pressure measurement, and can be correlated to the fuel rail pressure measurement to accurately calculate the fuel rail pressure rise attributed to each cylinder event.

[0064] At 326, the method includes learning a cylinder air fuel error based on the rise in fuel rail pressure following the selective opening of the direct injector. In particular, the controller can learn an air charge estimate for the cylinder based on the rise in fuel rail pressure. The air charge can be determined as a function of injector flow characteristics, pulse width, and air density according to the following equation: Charge [mass] = (Flow / Duration) * Density.

[0065] The learn can continue over multiple consecutive combustion events. For example, the controller can learn an air charge estimate for each of the plurality of cylinders of the engine over multiple consecutive cylinder events while the engine is operating in PFI only mode. The controller can then average the estimates over the plurality of cylinders to obtain an average air charge estimate for the engine. Additionally, the learn can be performed for each of the plurality of cylinders over multiple combustion events in each given cylinder. The controller can repeatedly estimate the air charge for each cylinder over multiple combustion events and average the air charge estimates for the cylinder.

[0066] As noted previously, the DI fuel rail pressure can rise in each event. For example, the fuel rail pressure can gradually rise from 100 psi to 200 psi over consecutive events. At 328, it can be determined whether the fuel rail pressure is above a threshold pressure, such as an upper threshold, above which fuel can inadvertently be injected into the cylinder when the DI is commanded to open. In one example, the upper threshold pressure is 500 psi. Thus, the learn can continue until the fuel rail pressure is above the upper threshold pressure. Then, at 330, the method includes injecting fuel from the direct injector (e.g., in an intake stroke) to lower the fuel rail pressure to a lower threshold pressure while maintaining the HPP disabled, and then, at 332, resuming the learn after the fuel rail pressure is below the lower threshold pressure. For example, the controller can move to perform an air charge estimate in the next cylinder in the engine firing order. Otherwise, if the upper threshold is not reached at 328, the method moves directly to 332 and continues the learn. While the direct injection is used to lower the fuel rail pressure, the intake port fuel mass can be reduced to maintain the required fuel mass to achieve the target air-fuel ratio.

[0067] At 334, the method includes estimating an air component of the cylinder-to-cylinder air-fuel ratio (AFR) error based on the comparison of the air charge estimate for each cylinder. For example, the controller can learn the air component of the cylinder AFR error based on a deviation between the air charge estimate (e.g., the average air charge estimate) for each cylinder. As one example, the average air charge estimate for a first engine cylinder can be compared to the average air charge estimate for a second engine cylinder (such as the next cylinder fire in firing order), and the air component of the AFR error for the first and second cylinders can be derived based on the difference between them. As another example, the average air charge estimate for a first engine cylinder can be compared to the average air charge estimate for all engine cylinders, and the air component of the AFR error for the first cylinder can be derived based on the difference between them. For example, a number of samples can be taken from each cylinder. These samples can then be averaged. The overall engine air charge can then be defined by the average of all cylinders. The error for each cylinder can then be calculated based on the individual cylinder average versus the overall engine average.

[0068] At 336, the method includes adjusting the cylinder fueling based on the air component and further based on a fuel component of the AFR error. As Figure 4 As detailed at 336, the fuel component of the air error can be learned by correlating the change in fuel rail pressure after each of a series of direct fuel injections into the cylinder. By learning the air component that is distinct from the fuel component of the AFR error, each error can be compensated accordingly. In one example, the controller can increase the cylinder fueling for a cylinder because the learned air charge estimate exceeds the expected air charge estimate (or average estimate). As another example, the controller can decrease the cylinder fueling for a cylinder when the learned air charge estimate falls below the expected air charge estimate (or average estimate). In further examples, other engine torque actuators can be adjusted based on the learned air error. For example, valve timing can be adjusted based on the learned air error.

[0069] Returning to 306, if the engine is not in the PFI-only mode, then at 308, it can be determined whether the engine is in the DI-only mode. If the engine is not in the DI-only mode, that is, the engine is in the PFDI mode in which the cylinders are fueled via each of port injection and direct injection, then the method moves to 314 to delay the estimation of the air component of the AFR error. This is because the PFI-only mode provides the most stable data points for air error estimation.

[0070] If the confirmation is in DI-only mode, then at 312, a determination can be made as to whether there is a deceleration fuel shutoff (DFSO) event. During DFSO, engine fueling is instantaneously stopped while cylinder valve operation continues, resulting in the engine spinning, unfueled. DFSO can be performed during low engine load, such as in response to a soft accelerator pedal event, vehicle traveling downhill, or during coasting, to reduce engine fuel consumption. If the confirmation is not in DFSO, then the method moves to 314 to delay estimation of the air component of the AFR error. During DI-only mode, HPP and direct injectors are enabled, and cylinder fueling is provided by commanding pulse width to the direct injectors based on torque demand. If the confirmation is in DFSO, then the method moves to 310 to disable HPP, and fuel via DI to reduce fuel rail pressure. Estimation is then made as discussed during PFI-only mode, with DI commanded to selectively open prior to the cylinder spark event, and the air charge estimation value for the cylinder is inferred based on the rise in fuel rail pressure following the command.

[0071] It will be appreciated that the learning described in the method of Figure 3 may be suspended in response to torque transients that change the fuel injection regime, such as a depression of the accelerator pedal or a release of the accelerator pedal. For example, in response to a depression of the accelerator pedal, the learning can be suspended, and the controller can transition to fueling the engine via port injection and direct injection, or direct injection only. The learned air charge estimation values can be saved in the memory of the controller, after which the learning can remain suspended until engine operating conditions are favorable for resumption of learning. For example, the learning can be suspended until the engine is fueled via port injection only, at which point the procedure can continue from the last learned cylinder event, or restart from a defined starting point.

[0072] In some examples, Figure 3The method of estimating and / or measuring air error can be performed with EGR enabled, and then performed with EGR disabled to learn the effect of EGR on the air estimation. For example, based on where EGR is delivered into the engine intake, such as where and when the EGR passage is coupled to the intake passage, some engine cylinders can receive more or less EGR flow than other cylinders. Thus, by learning the effect of EGR on the charge estimation of the cylinders, the air error can be better compensated for. The compensation applied to the air error of the cylinders can be different when EGR is enabled than when EGR is disabled. For example, once the fresh air charge flow is calculated (without EGR, "cylinder air charge without EGR"), then measurements can be taken again to determine the EGR of each cylinder (with EGR enabled, "cylinder_EGR"). Since EGR replaces fresh air, the actual fresh air charge of the cylinder is calculated based on the measured air charge without EGR and the measured EGR of each cylinder. Specifically, the actual fresh air charge of the cylinder ("cylinder_fresh air") is determined as:

[0073] cylinder_fresh air = cylinder_air charge_without EGR - cylinder_EGR.

[0074] Turning now to Figure 4 , an exemplary method 400 for learning a fuel component of air-fuel ratio error between cylinders is shown. The method enables reduction of cylinder-to-cylinder torque variation by compensating for the learned air error, such as using a fuel trim regulator.

[0075] At 402, the method includes estimating and / or measuring engine operating conditions. For example, parameters such as engine speed, engine load, operator torque demand, boost pressure, engine dilution (e.g., EGR flow) around conditions (such as ambient temperature, barometric pressure, ambient temperature), etc. can be determined. At 404, it can be determined whether estimation conditions for determining a fuel component of AFR error between engine cylinders are present. In one example, the estimation conditions can be confirmed in response to the engine being in a low load operating region (such as when engine speed and / or operator torque demand is below a threshold), the engine temperature being greater than a threshold temperature that ensures that injector calibration injection events are performed when the engine temperature is relatively stable (e.g., above 80°C), and a threshold duration or distance of engine operation has elapsed since the last estimation of fuel error. If the estimation conditions are not met, then at 406, the method delays estimation of the fuel component of AFR error. This is because existing conditions do not provide stable data points for fuel error estimation. This can occur when the engine is in DI mode, PFI mode, or PFDI mode.

[0076] If the estimation condition is met, at 408, the method includes operating the HPP to raise the direct injector rail pressure above a threshold pressure. As an example, the controller can increase the fuel rail pressure by issuing additional pump strokes to the HPP, increasing the pump stroke frequency, and / or increasing the pump stroke for at least one stroke, such that the fuel pressure in the high pressure fuel rail reaches a predetermined threshold calibration pressure. In one example, the threshold calibration pressure is an upper threshold pressure, such as 200 psi. The HPP operation can be increased based on engine speed, engine load, supercharged operation, intake charge pressure, multiple calibration injections (for the engine, or for each injector), and / or other operating conditions. At 410, the fuel rail pressure can be evaluated relative to the threshold calibration pressure. If the predetermined threshold calibration pressure is not reached, at 412, the method includes continuing HPP operation until the target fuel rail pressure is reached. Otherwise, once the pressure is reached, at 414, the HPP can be disabled. Further, a fuel volume can be commanded to be injected into the first cylinder via the direct injector. The commanded volume can be based on the fuel rail pressure and fuel density. In one example, the controller determines the required fuel rail pressure and calculates the amount of fuel that needs to be removed from the rail to achieve the target pressure. The fuel mass is converted to volume based on the fuel density. The volume is then converted to a flow duration (i.e., pulse width) based on the injector flow characteristics. The controller can command the pulse width to the direct injector based on the target volume to be delivered. As detailed below, the controller can run a series of fuel injections in a predetermined sequence (e.g., injector #1, injector #2, injector #3, injector #4, or in the firing order specified for the engine), and repeat the sequence for a predetermined number of times (e.g., 3 engine cycles, with each injector operating at least once during each engine cycle).

[0077] At 416, after the injection in the first cylinder, the method includes estimating the drop in high pressure fuel rail pressure after each injection event. Specifically, in each injection event, the DI fuel rail pressure can drop when fuel is delivered into the cylinder with HPP disabled. For example, in a consecutive event, the fuel rail pressure can gradually drop from 200 psi to 100 psi. The controller can calculate the fuel pressure drop (APij) due to each injection of the ith injector (e.g., if each injector injects 3 times during a calibration injection cycle and the calibration injection cycle runs 3 times during a calibration event, then j = 1, 2, 3...9). APij corresponds to the drop in pressure in the DI fuel rail due to an injection by the ith injector during the jth injection. Various engine conditions or events can affect the fuel rail pressure measurement and can be taken into account when calculating the fuel pressure drop (APij) attributed to each injection. Thus, in some examples, the procedure can associate the fuel pressure with various engine conditions sensed via various sensors. For example, the transient pressure pulsation resulting from injector firing can temporarily affect the fuel rail pressure measurement, affecting the calibration accuracy. As such, the sampling of the fuel pressure can be selected to reduce the transient effects of the injector firing. Additionally or alternatively, if the injector firing timing is associated with the fuel rail pressure measurement, then the temporary pressure drop resulting from the injector firing can be taken into account when determining the injector calibration values. Similarly, intake and / or exhaust valve opening and closing, intake and / or exhaust pressure, crank angle position, cam position, spark firing, and engine combustion can also affect the fuel rail pressure measurement and can be associated with the fuel rail pressure measurement to accurately calculate the fuel rail pressure drop attributed to each injection.

[0078] At 418, the method includes estimating the volume actually injected into the cylinder at each injection event based on the estimated drop in fuel rail pressure at that injection event. For example, the controller can use equation (1) to calculate the amount of fuel actually injected Qij in each injection, as follows:

[0079] Qij = APij / C (1)

[0080] where C is a predetermined constant factor for converting the amount of fuel pressure drop to the amount of injected fuel. Additionally, the controller can use equation (2) to determine the average amount of fuel actually injected by injector i (Qi), as follows:

[0081]

[0082] where j is the number of injections for injector i (e.g., if each injector injects 3 times during a calibration injection cycle and the calibration injection cycle runs 3 times during a calibration event, then j = 1, 2, 3...9).

[0083] At 420, a cylinder fueling error is determined based on a difference between a commanded volume (based on a pulse width to a direct injector) and an actual volume received in the cylinder (based on a drop in a corresponding fuel rail pressure). At 422, after determining the fueling error for the first cylinder, the controller moves to perform a fuel error estimation in the next cylinder in the firing order (or predetermined calibration sequence).

[0084] At 424, the method includes estimating a fuel component of the cylinder-to-cylinder AFR error based on a comparison of the fueling estimation values (or fueling errors) for each cylinder. In one example, the controller can use equation (3) to calculate a correction factor for each fuel injector i (e.g., for a four cylinder engine, i = 1, 2, 3, or 4) as follows:

[0085] ki = Qc / Qi (3)

[0086] The controller can update the correction factor for injector i with the newly calculated ki. For example, the newly calculated ki will replace the old ki stored in the keep alive memory (KAM) of the control unit that can currently be used to calibrate injector i. In further examples, the controller can learn the fuel component of the cylinder-to-cylinder AFR error based on a deviation between the fuel error estimation values for each cylinder. As one example, an average fuel error estimation value for a first engine cylinder can be compared to an average fuel error estimation value for a second engine cylinder (such as the next cylinder to fire in the firing order), and the fuel component of the AFR error for the first and second cylinders can be derived based on a difference between them. As another example, an average fuel error for a first engine cylinder can be compared to an average fuel error for all engine cylinders, and the fuel component of the AFR error for the first cylinder can be derived based on a difference between them.

[0087] At 426, the method includes retrieving an air component of the cylinder-to-cylinder AFR error from a memory of the controller. The air error can be determined during the PFI only mode based on a rise in the fuel rail pressure after the opening of the DI prior to the cylinder spark event, as Figure 3 is detailed.

[0088] At 428, the method includes adjusting cylinder fueling based on the air component and further based on the fuel component of the AFR error. By knowing the air component that is different from the fuel component of the AFR error, each error can be compensated accordingly. As one example, when the learned fuel error is increasing, the controller can increase the cylinder fueling to the cylinder. As another example, when the learned fuel error is decreasing, the controller can decrease the cylinder fueling to the cylinder. In further examples, other engine torque actuators can be adjusted based on the learned fuel error. For example, spark timing can be adjusted based on the learned air error. As another example, fuel rail pressure can be adjusted based on the learned fuel error. In some examples, each of the air and fuel errors can be adjusted via a fueling adjuster. In other examples, the air error can be compensated via a different adjuster (e.g., a different torque actuator) than the fuel error compensation. For example, torque can be adjusted using spark. As another example, EGR flow can be adjusted to change the overall percentage error (e.g., by decreasing EGR flow rate from 10% to 5%). This would still allow some EGR benefits without pushing the cylinder over the OBD threshold and out of balance.

[0089] In this way, with the high pressure fuel pump disabled, the engine controller can learn an air component of a cylinder torque change based on a first change in direct injector fuel rail pressure when selectively opening the direct injector for a threshold duration before commanding a spark event in a cylinder that is fueled via intake port injection only. Then, the controller can learn a fuel component of the cylinder torque change based on a second change in direct injector fuel rail pressure when commanding the direct injector to open in a cylinder that is fueled via direct injection only. In one example, the first change in direct injector fuel rail pressure includes an increase in fuel rail pressure and the second change in direct injector fuel rail pressure includes a decrease in fuel rail pressure. While learning the air component, the direct injector can be commanded to selectively open after the direct injector fuel rail pressure has decreased below a first threshold pressure. In contrast, while learning the fuel component, the direct injector can be commanded to open after the direct injector fuel rail pressure has increased above a second threshold pressure. During learning of the air component and during learning of the fuel component, the high pressure fuel pump coupled to the direct injector is disabled. Further, during learning of the air component, the engine can be fueled via intake port injection only and during learning of the fuel component, the engine can be fueled via direct injection only. In a PFDI engine, where the engine can be fueled via intake port injection and direct injection, the controller can fuel the engine via intake port injection only during learning. If the engine is a DI engine, the engine will be fueled via direct injection even during learning.

[0090] Turning now toFigure 5 FIG. 5 shows an exemplary characteristic curve 500 of valve timing and piston position for a given engine cylinder relative to engine position, and depicts the timing of direct injector opening for air error estimation. During selected conditions, such as when a cylinder is refueled via port injection only, the engine controller can command the direct injector to open to temporarily couple the cylinder to the DI fuel rail (and its pressure sensor) without injecting fuel into the cylinder. The air charge estimation error for the cylinder can then be inferred based on the change in fuel rail pressure.

[0091] Characteristic curve 500 exhibits engine position in degrees of crank angle (CAD) along the x-axis. Curve 508 depicts piston position (along the y-axis) relative to their position from top dead center (TDC) and / or bottom dead center (BDC), and further relative to their position within the four strokes (intake, compression, power, and exhaust) of an engine cycle. As indicated by sinusoidal curve 508, the pistons move gradually downward from TDC, dropping to a lowest point at BDC at the end of the power stroke. The pistons then return to the top at TDC at the end of the exhaust stroke. The pistons then move downward again toward BDC during the intake stroke, returning to the initial top position at TDC at the end of the compression stroke.

[0092] Curves 502 and 504 depict valve timing for the exhaust valve (dashed curve 502) and the intake valve (solid curve 504) during normal engine operation. As shown, the exhaust valve can open just as the pistons are dropping to a lowest point at the end of the power stroke. The exhaust valve can then close as the pistons complete the exhaust stroke, remaining open at least until the subsequent intake stroke has begun. In the same manner, the intake valve can open at or before the start of the intake stroke, and can remain open at least until the subsequent compression stroke has begun.

[0093] Due to the timing difference between exhaust valve closing and intake valve opening, both the intake and exhaust valves can be open for a short duration before the end of the exhaust stroke and after the start of the intake stroke. The period of time during which both valves are open is referred to as positive intake and exhaust valve overlap 506 (or simply positive valve overlap), represented by the hatched area of intersection of curves 502 and 504. In one example, positive intake and exhaust valve overlap 506 can be a default cam position that the engine assumes during cold engine start.

[0094] The third plot of characteristic curves 500 (from the top) depicts exemplary timing of fuel injector opening and closing during a cylinder event. Operation of the intake port injector is shown as a hatched block, while operation of the direct injector is shown as a striped block. The fourth plot from the top of characteristic curves 500, plot 510, depicts fuel rail pressure of a high pressure fuel rail coupled to the direct injector.

[0095] In the depicted cylinder event, the cylinder is operated with the HPP coupled to the direct injector disabled, resulting in a fuel rail pressure (HP FRP) below a threshold. The engine controller is configured to provide the total amount of fuel to the cylinder via intake port injection in the exhaust stroke at CAD1. Then, in the compression stroke, the direct injector is commanded to open for a short duration at CAD2, prior to a spark event 514 in the cylinder. In the depicted example, a minimum pulse width is commanded to the direct injector. Since the DI is commanded to open when the fuel rail pressure is low, no fuel is directly injected into the cylinder. As a result of the DI opening, the combustion chamber of the cylinder is temporarily coupled to the DI fuel rail, and the compression pressure of the cylinder is sensed via the DI fuel rail pressure sensor. In particular, a spike 512 in the fuel rail pressure is observed. Since the compression pressure is a function of the cylinder volume and the air charge, an air charge estimate for the given cylinder can be inferred based on the sensed fuel rail pressure spike 512. By subsequently comparing the air charge estimate for the given cylinder with estimates for other engine cylinders, an air component of cylinder-to-cylinder AFR error can be determined and compensated for.

[0096] Reference is made to Figure 6 Exemplary engine air and fuel component error estimates are depicted. Characteristic curves 600 depict high pressure fuel pump operation at plot 602, high pressure (DI) fuel rail pressure at plot 604, pulse width commanded to the intake port injector of the corresponding cylinder at plot 606, and pulse width commanded to the direct injector of the corresponding cylinder at plot 608. All plots are depicted along the x-axis over time. Cylinder events (1-4) are labeled based on firing order (1-3-4-2 in the depicted example). Cylinder spark events are depicted with asterisks. The position of the asterisks relative to the pulse width commanded to at least the DI indicates relative firing timing.

[0097] Before ti, the engine is operating with each of the boost pump (not shown) and HPP. At this time, the engine is refueled via each of port injection and direct injection. The split of the delivered fuel includes a higher ratio of PFI fuel to DI fuel as shown by the difference in commanded pulse width. At ti, the operator torque demand (e.g., accelerator pedal event) drops, in response to which the engine is refueled via direct injection only. Thus, at ti, HPP is disabled. The air error estimation condition is deemed satisfied. Between ti and t2, the fuel rail pressure (FRP) is reduced to a lower threshold Thr L to enable air error estimation. The FRP is reduced by repeatedly injecting fuel via DI, the controller commanding a short (e.g., minimum) pulse width to DI.

[0098] At t2, once the FRP is reduced, air estimation is initiated in the next firing cylinder (here, cylinder 2) by port injecting fuel during the exhaust stroke, then commanding DI to open before the cylinder's spark event. The opening of DI during the compression stroke results in no fuel being directly injected, but results in a spike in the fuel rail pressure depicted at 610 for one cylinder. Similarly, in the successive cylinder events between t2 and t3, air charge estimation is made multiple times for each of cylinders 1-4 based on the rise in FRP after DI is opened during the cylinder's compression stroke, the cylinder being refueled via port injection only.

[0099] At t3, the FRP reaches an upper threshold Thr U from which further rise in FRP cannot be reliably estimated. Thus, learning is suspended, and between t3 and t4 (as between ti to t2), the fuel rail pressure (FRP) is reduced to a lower threshold Thr L by repeatedly injecting fuel via DI, the controller commanding a short (e.g., minimum) pulse width to DI. At t4, once the FRP has been reduced to Thr L, learning is resumed. Learning includes learning an air charge estimate for each cylinder based on the corresponding FRP (e.g., 610) rise for that cylinder event. The air charge estimates for each cylinder are then compared to each other to identify cylinders that are leaner or richer than expected.

[0100] At t5, the operator torque demand (e.g., an accelerator pedal event) rises, in response to which the engine is refueled via port injection and direct injection. Thus, at t5, HPP is enabled. The split ratio of the fuel delivered includes a higher ratio of DI fuel to PFI fuel, as shown by the difference in commanded pulse width. Shortly before t6, the operator torque demand (e.g., another accelerator pedal event) rises further, in response to which the engine is refueled via direct injection only. Prior to t6, the combustion event in cylinder 4 occurs with direct injection only of fuel.

[0101] At t6, the fuel error estimation condition is deemed satisfied. Since the FRP is already at or above the upper threshold pressure Thr U, no further pump operation is needed, and HPP is disabled. In addition, by direct injection of a predetermined amount of fuel during the intake stroke and measuring the resulting fuel rail pressure drop (depicted at 612 for one cylinder), a fuel estimation is initiated in the next firing cylinder (here, cylinder 2). Similarly, in the successive cylinder events between t6 and t7, fuel is estimated for each of cylinders 1-4 multiple times based on the drop in FRP after DI fuel injection into the cylinder (which is refueled via direct injection only). The fueling estimation for each cylinder is learned including based on the corresponding FRP (e.g., 612) drop for that cylinder event. The fuel estimation for each cylinder is then compared to the fuel volume based on the commanded pulse width to identify cylinders that are leaner than expected or richer than expected.

[0102] At t7, the operator torque demand (e.g., an accelerator pedal event) increases, in response to which the engine is refueled via direct injection only. Thus, at t7, HPP is enabled, and learning is disabled. After t7, the fueling to each cylinder is adjusted based on the learned air and fuel error components of the cylinder-to-cylinder AFR variation for each cylinder. For example, the fueling in cylinder 1 is increased by extending the pulse width (compared to the unadjusted pulse width shown in dashed line). As another example, the fueling in cylinder 4 is decreased by reducing the pulse width (compared to the unadjusted pulse width shown in dashed line).

[0103] In this way, cylinder-to-cylinder variability can be reduced by knowing and distinguishing the air component of the AFR error from the fuel component of the AFR error. By adjusting subsequent engine fueling based on the air and fuel components, a single actuator can be used to compensate for torque variations between cylinders. By inferring the air error from the cylinder compression pressure, cylinder air charge can be accurately estimated while relying on existing sensors and without introducing noise effects of EGR. By commanding the DI to open prior to the spark event without fuel being directly injected into the cylinder, reducing the potential for damage to the results, while the high pressure pump is disabled. By estimating the rise in the DI fuel rail pressure during the period when the cylinder is being fueled with only port injection, a more reliable and stable data point can be used to infer the air charge. By knowing the air error and compensating for it, cylinder torque variations can be reduced, improving engine emissions and NVH.

[0104] An example method includes injecting fuel from a direct injector with the high pressure pump disabled to lower a direct injection fuel rail pressure below a threshold pressure; then injecting a port fuel injection into a cylinder and commanding the direct injector to selectively open for a threshold duration prior to a spark event in the cylinder without injecting any fuel from the direct injector. In the foregoing example, additionally or optionally, the method further includes learning a cylinder air-fuel ratio error based on a rise in the fuel rail pressure after the selective opening. In any or all of the foregoing examples, additionally or optionally, learning the cylinder air-fuel ratio error includes learning an air charge estimate for the cylinder based on the rise in the fuel rail pressure. In any or all of the foregoing examples, additionally or optionally, the method further includes adjusting a cylinder fueling in response to the learned cylinder air-fuel ratio error, the cylinder fueling increasing when the learned air charge estimate exceeds an expected air charge estimate, the cylinder fueling decreasing when the learned air charge estimate exceeds the expected air charge estimate. In any or all of the foregoing examples, additionally or optionally, the cylinder is one of a plurality of engine cylinders, the method further including learning an air charge estimate for each of the plurality of engine cylinders in a plurality of consecutive cylinder events. In any or all of the foregoing examples, additionally or optionally, learning the cylinder air-fuel ratio error further includes learning the cylinder air-fuel ratio error based on a deviation between the air charge estimates for the plurality of engine cylinders. In any or all of the foregoing examples, additionally or optionally, the learning is performed for each of the plurality of cylinders in a plurality of combustion events in each cylinder, and wherein the air charge estimate for a given cylinder is an average air charge estimate averaged over the plurality of combustion events in the given cylinder. In any or all of the foregoing examples, additionally or optionally, the threshold pressure is a function of atmospheric pressure, and wherein the threshold duration is based on engine speed and load. In any or all of the foregoing examples, additionally or optionally, the threshold pressure is a lower threshold pressure, the method further including learning the cylinder air-fuel ratio error until the fuel rail pressure is above an upper threshold pressure, the upper threshold pressure being higher than the lower threshold pressure, then injecting fuel from the direct injector with the high pressure pump disabled to lower the fuel rail pressure to the lower threshold pressure, and resuming the learning after the fuel rail pressure is below the lower threshold pressure. In any or all of the foregoing examples, additionally or optionally, injecting a port fuel injection into the cylinder includes making a port injection during an exhaust stroke or an intake stroke of the cylinder, and wherein the direct injector is commanded to selectively open during a compression stroke of the cylinder.In any or all of the foregoing examples, additionally or optionally, the method further includes disabling the direct injector after the direct injection fuel rail pressure is reduced below the threshold pressure, and wherein injecting fuel into the cylinder via intake port injection includes not directly injecting fuel into the cylinder and maintaining the direct injector disabled.

[0105] Another example method for an engine includes, with a high pressure fuel pump disabled, learning an air component of a cylinder torque change based on a first change in a direct injection fuel rail pressure when the direct injector is selectively opened for a threshold duration before a spark event in a cylinder that is fueled via intake port injection only is commanded, and learning a fuel component of the cylinder torque change based on a second change in the direct injection fuel rail pressure when the direct injector is selectively opened for the threshold duration before the spark event in the cylinder that is fueled via intake port injection only is commanded. In the foregoing example, additionally or optionally, the first change in the direct injection fuel rail pressure includes a rise in the fuel rail pressure, and wherein the second change in the direct injection fuel rail pressure includes a fall in the fuel rail pressure. In any or all of the foregoing examples, additionally or optionally, during the learning of the air component, the direct injector is commanded to selectively open after the direct injection fuel rail pressure has fallen below a first threshold pressure, and wherein during the learning of the fuel component, the direct injector is commanded to open after the direct injection fuel rail pressure has risen above a second threshold pressure. In any or all of the foregoing examples, additionally or optionally, during the learning of the air component and the learning of the fuel component, a high pressure fuel pump coupled to the direct injector is disabled. In any or all of the foregoing examples, additionally or optionally, during the learning of the air component, the engine is fueled via intake port injection only, and wherein during the learning of the fuel component, the engine is fueled via direct injection only.

[0106] An example engine system includes: an engine including a cylinder; a port fuel injector coupled to the cylinder; a direct injector coupled to the cylinder; a high pressure fuel pump delivering fuel to the direct injector via a direct injection fuel rail; a pressure sensor for estimating direct injection fuel rail pressure; and a controller having computer readable instructions stored on a non-transitory memory for: operating the direct injector with the fuel pump disabled until the fuel rail pressure falls below a first threshold pressure, then disabling the direct injector; temporarily opening the direct injector without delivering any fuel during a compression stroke of the cylinder but before a spark event of the cylinder; estimating a cylinder air charge based on a change in the fuel rail pressure during the temporary opening; and adjusting a subsequent cylinder fueling based on the estimated cylinder air charge. In the foregoing example, additionally or optionally, the temporary opening is performed for a predetermined number of injection events of the cylinder, wherein the estimated cylinder air charge is an average cylinder air charge averaged over the predetermined number of injection events, and wherein adjusting a subsequent cylinder fueling includes adjusting a subsequent cylinder fueling via one or more of the port fuel injector and the direct injector. In any or all of the foregoing examples, additionally or optionally, the cylinder is one of a plurality of engine cylinders, wherein the fueling and temporary opening are performed for each of the plurality of engine cylinders in a plurality of consecutive injection events of the cylinder, and wherein adjusting a subsequent cylinder fueling based on the estimated cylinder air charge includes adjusting a subsequent fueling for each engine cylinder based on the estimated cylinder air charge for the corresponding cylinder relative to an average cylinder air charge estimate averaged over the plurality of engine cylinders. In any or all of the foregoing examples, additionally or optionally, the temporary opening is performed when the cylinder is fueled via only the port fuel injector or during a deceleration fuel cut event.

[0107] In another representation, the engine system can be coupled in a hybrid electric vehicle.

[0108] It should be noted that the example control and estimation procedures included herein can be used with a variety of engine and / or vehicle system configurations. The control methods and procedures disclosed herein can be stored as executable instructions in non-transitory memory and can be executed by control systems including controllers in combination with various sensors, actuators, and other engine hardware. The particular procedures described herein can represent one or more of any number of processing strategies of the example embodiments described herein. As such, the various acts, operations, and / or functions illustrated can be performed in the manner shown, contemporaneously, or in any order, omitted, or in some cases, combined. Likewise, the term "function" or "procedure" as used herein does not necessarily mean that all requirements of that function or procedure are performed in one act or operation. In addition, one or more of the acts, operations and / or functions illustrated can be repeated any number of times, unless the context clearly dictates otherwise. Further, the described acts, operations and / or functions can graphically represent code to be programmed into non-transitory memory of a computer readable storage medium of an engine control system, where the described acts are performed by execution of instructions in the system including various engine hardware components in combination with an electronic controller.

[0109] It should be understood that the configurations and procedures disclosed herein are exemplary in nature, and that these specific embodiments are not to be taken as limiting, as numerous variations are possible. For example, the above-described techniques can be applied to V-6, I-4, I-6, V-12, opposed 4 cylinder, and other engine types. The subject matter of the present disclosure includes all novel and nonobvious combinations and subcombinations of the various systems and configurations disclosed herein, as well as other features, functions, and / or properties noted herein.

[0110] The claims particularly point out certain combinations and subcombinations that are regarded as novel and nonobvious. These claims can refer to "an" element or to "a first" element or to "one" element and equivalent therefor. Such claims should be understood to include one or more such elements, neither requiring nor excluding two or more elements. Other combinations and subcombinations of the disclosed features, functions, elements, and / or properties can be claimed through amendment of the present claims or through the presentation of additional claims in this or a related application. Such altered claims, whether they are broader, narrower, equal in scope with, or otherwise potentially different from the original claims, are also regarded as included within the subject matter of the present disclosure.

[0111] According to the present invention, there is provided a method having: injecting fuel from a direct injector with a high pressure pump disabled to reduce a direct injection fuel rail pressure below a threshold pressure; then, injecting fuel into a cylinder and commanding the direct injector to selectively open for a threshold duration prior to a spark event in the cylinder without injecting any fuel from the direct injector.

[0112] According to one embodiment, the above invention is further characterized by learning a cylinder air-fuel ratio error based on a rise in the fuel rail pressure after the selective opening.

[0113] According to one embodiment, the above invention is further characterized by learning an air charge estimate for the cylinder based on the rise in the fuel rail pressure.

[0114] According to one embodiment, the above invention is further characterized by adjusting a cylinder fueling in response to the learned cylinder air-fuel ratio error, the cylinder fueling being increased when the learned air charge estimate exceeds an expected air charge estimate, the cylinder fueling being decreased when the learned air charge estimate exceeds the expected air charge estimate.

[0115] According to one embodiment, the cylinder is one of a plurality of engine cylinders, the method further comprising: learning an air charge estimate for each of the plurality of engine cylinders in a plurality of consecutive cylinder events.

[0116] According to one embodiment, the above invention is further characterized by learning the cylinder air-fuel ratio error further comprising: learning the cylinder air-fuel ratio error based on a deviation between the air charge estimates for the plurality of engine cylinders.

[0117] According to one embodiment, the above invention is further characterized by performing the learning for each of the plurality of cylinders in a plurality of combustion events in each cylinder, and wherein the air charge estimate for a given cylinder is an average air charge estimate averaged over the plurality of combustion events in the given cylinder.

[0118] According to one embodiment, the threshold pressure is a function of atmospheric pressure, and wherein the threshold duration is based on engine speed and load.

[0119] According to one embodiment, the threshold pressure is a lower threshold pressure, the method further comprising: learning the cylinder air-fuel ratio error until the fuel rail pressure is above an upper threshold pressure, the upper threshold pressure being higher than the lower threshold pressure, then injecting fuel from the direct injector to lower the fuel rail pressure to the lower threshold pressure with the high pressure pump disabled, and resuming the learning after the fuel rail pressure is below the lower threshold pressure.

[0120] According to one embodiment, the above invention is further characterized by injecting fuel into the cylinder comprises: injecting during an exhaust stroke or an intake stroke of the cylinder, and wherein the direct injector is commanded to selectively open during a compression stroke of the cylinder.

[0121] According to one embodiment, the above invention is further characterized by disabling the direct injectors after the direct injection fuel rail pressure is reduced below a threshold pressure, and wherein injecting fuel into the cylinder via intake port injection includes not directly injecting fuel into the cylinder and maintaining the direct injectors disabled.

[0122] According to the invention, there is provided a method for an engine having: learning an air component of a cylinder torque change based on a first change in a direct injection fuel rail pressure when the high pressure fuel pump is disabled and the direct injectors are selectively opened for a threshold duration before a spark event in a cylinder that is fueled via intake port injection only is commanded; and learning a fuel component of the cylinder torque change based on a second change in the direct injection fuel rail pressure when the direct injectors are selectively opened for a threshold duration before a spark event in a cylinder that is fueled via intake port injection only is commanded.

[0123] According to one embodiment, the first change in the direct injection fuel rail pressure includes a rise in the fuel rail pressure, and wherein the second change in the direct injection fuel rail pressure includes a fall in the fuel rail pressure.

[0124] According to one embodiment, during the learning of the air component, the direct injectors are commanded to selectively open after the direct injection fuel rail pressure has fallen below a first threshold pressure, and wherein during the learning of the fuel component, the direct injectors are commanded to open after the direct injection fuel rail pressure has risen above a second threshold pressure.

[0125] According to one embodiment, during the learning of the air component and the learning of the fuel component, a high pressure fuel pump coupled to the direct injectors is disabled.

[0126] According to one embodiment, during the learning of the air component, the engine is fueled via intake port injection only, and wherein during the learning of the fuel component, the engine is fueled via direct injection only.

[0127] According to the present invention, there is provided an engine system having: an engine including a cylinder; an intake port injector coupled to the cylinder; a direct injector coupled to the cylinder; a high pressure fuel pump delivering fuel to the direct injector via a direct injection fuel rail; a pressure sensor for estimating a direct injection fuel rail pressure; and a controller having computer readable instructions stored on a non-transitory memory for: operating the direct injector with the fuel pump disabled until the fuel rail pressure falls below a first threshold pressure, then disabling the direct injector; temporarily opening the direct injector without delivering any fuel during a compression stroke of the cylinder, but prior to a spark event of the cylinder; estimating a cylinder air charge based on a change in the fuel rail pressure during the temporary opening; and adjusting a subsequent cylinder fueling based on the estimated cylinder air charge.

[0128] According to one embodiment, the temporary opening is performed for a predetermined number of injection events of the cylinder, wherein the estimated cylinder air charge is an average cylinder air charge averaged over the predetermined number of injection events, and wherein adjusting a subsequent cylinder fueling includes adjusting a subsequent cylinder fueling via one or more of the intake port injector and the direct injector.

[0129] According to one embodiment, the cylinder is one of a plurality of engine cylinders, wherein the fueling and temporary opening are performed for each of the plurality of engine cylinders in a plurality of consecutive injection events of the cylinder, and wherein adjusting a subsequent cylinder fueling based on the estimated cylinder air charge includes adjusting a subsequent fueling for each engine cylinder based on the estimated cylinder air charge for the corresponding cylinder relative to an average cylinder air charge estimate averaged over the plurality of engine cylinders.

[0130] According to one embodiment, the temporary opening is performed when the cylinder is fueled via only the intake port injector or during a deceleration fuel cut event.

Claims

1. A method for an engine, comprising: injecting fuel from a direct injector without the high pressure pump to lower a direct injection fuel rail pressure below a threshold pressure; then injecting fuel into a cylinder and commanding the direct injector to selectively open for a threshold duration before a spark event in the cylinder without injecting any fuel from the direct injector; and learning an air charge estimate for the cylinder based on a rise in the fuel rail pressure.

2. The method of claim 1, further comprising: learning an air component of a cylinder air-fuel ratio error based on the air charge estimate for the cylinder.

3. The method of claim 2, further comprising: adjusting a cylinder fueling in response to the learned air component of the cylinder air-fuel ratio error, the cylinder fueling increasing when the learned air charge estimate exceeds an expected air charge estimate, the cylinder fueling decreasing when the learned air charge estimate falls below the expected air charge estimate.

4. The method of claim 2, wherein the cylinder is one of a plurality of engine cylinders, the method further comprising: learning an air charge estimate for each of a plurality of engine cylinders in a plurality of consecutive cylinder events.

5. The method of claim 4, wherein learning the air component of the cylinder air-fuel ratio error further comprises: learning an air component of a cylinder air-fuel ratio error based on a deviation between the air charge estimates for the plurality of engine cylinders.

6. The method of claim 5, wherein the learning is performed for each of the plurality of cylinders in a plurality of combustion events in each cylinder, and wherein the air charge estimate for a given cylinder is an average air charge estimate averaged over the plurality of combustion events in the given cylinder.

7. The method of claim 1, wherein the threshold pressure is a function of atmospheric pressure, and wherein the threshold duration is based on engine speed and load.

8. The method of claim 2, the threshold pressure is a lower threshold pressure, the method further comprising: learning the air component of the cylinder air-fuel ratio error until the fuel rail pressure is above an upper threshold pressure, the upper threshold pressure being higher than the lower threshold pressure, then injecting fuel from the direct injector without the high pressure pump to lower the fuel rail pressure to the lower threshold pressure, and resuming the learning after the fuel rail pressure is below the lower threshold pressure.

9. The method of claim 1, wherein injecting fuel into the cylinder comprises: injecting during an exhaust stroke or an intake stroke of the cylinder, and wherein the direct injector is commanded to selectively open during a compression stroke of the cylinder.

10. The method of claim 1, further comprising: deactivating the direct injector after lowering the direct injection fuel rail pressure below a threshold pressure, and wherein injecting a port fuel into the cylinder includes not directly injecting fuel into the cylinder and maintaining the high pressure pump deactivated.

11. An engine system, comprising: an engine including a cylinder; a port injector coupled to the cylinder; a direct injector coupled to the cylinder; a high pressure fuel pump delivering fuel to the direct injector via a direct injection fuel rail; a pressure sensor for estimating a direct injection fuel rail pressure; and a controller having computer readable instructions stored on a non-transitory memory, the instructions for: operating the direct injector without the fuel pump until the fuel rail pressure falls below a first threshold pressure, and then deactivating the direct injector; temporarily opening the direct injector without delivering any fuel during a compression stroke of the cylinder, but prior to a spark event of the cylinder; estimating a cylinder air charge based on a change in the fuel rail pressure during the temporary opening; and adjusting a subsequent cylinder fueling based on the estimated cylinder air charge.

12. The system of claim 11, wherein the temporary opening is performed for a predetermined number of injection events of the cylinder, wherein the estimated cylinder air charge is an average cylinder air charge averaged over the predetermined number of injection events, and wherein adjusting a subsequent cylinder fueling includes: adjusting a subsequent cylinder fueling via one or more of the intake port injector and the direct injector.

13. The system of claim 11, wherein the cylinder is one of a plurality of engine cylinders, wherein the fueling and the temporary opening are performed for each of the plurality of engine cylinders in a plurality of consecutive injection events of the cylinder, and wherein adjusting a subsequent cylinder fueling based on the estimated cylinder air charge comprises: adjusting a subsequent fueling of each engine cylinder based on the estimated cylinder air charge of the corresponding cylinder relative to an average cylinder air charge estimate averaged over the plurality of engine cylinders.

14. The system of claim 11, wherein the temporary opening is performed when the cylinder is being fueled via only the intake port injector or during a deceleration fuel cut event.

Images