Torque control method and system

Abstract

Method for a machine, comprising: Indicating cylinder-to-cylinder imbalance based on each of the exhaust air-fuel ratio (LAMavg) estimated by an exhaust gas sensor (126), the exhaust manifold pressure (Pavg) estimated by a pressure sensor (124), and the individual cylinder torque (TQavg) estimated by a crankshaft torque sensor (200, 210); wherein the indicating based on the exhaust air-fuel ratio (LAMavg) is based on the exhaust air-fuel ratio (LAMavg) weighted with a first confidence factor (c1), the indicating based on the exhaust manifold pressure (Pavg), based on exhaust pressure weighted by a second confidence factor (c2), and specifying based on individual cylinder torque (TQavg) based on the individual cylinder torque (TQavg) weighted by a third confidence factor (c3); further comprising calculating a combined imbalance parameter based on the weighted exhaust air-fuel ratio (LAMavg), the weighted exhaust manifold pressure (Pavg), and the weighted individual cylinder torque (TQavg) for each engine cylinder; and by means of a control device (12) provided with computer-readable instructions stored in a non-volatile memory, Determining a first confidence factor (c1) of the estimated exhaust gas air-fuel ratio (LAMavg) based on fuel type, Determining a second confidence factor (c2) of the estimated exhaust pressure based on valve timing, Determining a third confidence factor (c3) of the estimated individual Cylinder torque (TQavg) based on exhaust temperature, indicating cylinder-to-cylinder imbalance based on two or more of the estimated exhaust air-fuel ratio (LAMavg) weighted by the first confidence factor (c1), the estimated exhaust manifold pressure (Pavg) weighted by the second confidence factor (c2) and the estimated individual cylinder torque (TQavg) weighted by the third confidence factor (c3), and applying an air-fuel ratio correction to one or more cylinders (30) based on the indication.

Description

Field of InterestThe present description relates generally to methods and systems for determining cylinder-to-cylinder torque imbalance in an internal combustion engine of a vehicle.STATE OF THE ART / SUMMARYMeeting engine emissions requires sensing the imbalance in air-fuel ratio (AFR) across all engine cylinders. Inter-cylinder AFR imbalance may occur when the AFR in one or more cylinders is different from the other cylinder due to issues such as intake manifold leakage, fuel injector errors, exhaust gas recirculation errors, and fuel flow delivery issues. In addition to degrading emissions, cylinder-to-cylinder imbalance may result in torque errors that reduce engine performance and vehicle drivability.An example approach to detecting cylinder-to-cylinder AFR imbalance is shown by Behr et al. in U.S. Pat. No. 7,802,563 B2. Herein, AFR imbalance is identified based on the response of an exhaust UEGO sensor at frequencies that are at or above an ignition frequency of the cylinders during selected operating conditions. Specifically, if the vehicle is not in a transient engine operating condition, imbalance is identified if the integration of high frequency difference signals detected by the UEGO sensor is greater than a threshold. Still other approaches to detecting AFR imbalance involve detecting AFR imbalance based on exhaust manifold pressure. However, the present inventors have recognized potential problems with these methods. As one example, when using exhaust gas sensors, such as the Behr approach, conditions may exist where cylinder-to-cylinder imbalance is not detected due to insufficient mixing of exhaust at the exhaust gas sensor. Further, the exhaust gas sensor may not be able to reliably detect cylinder-to-cylinder imbalance during a cold-start condition of the engine due to insufficient warming of the exhaust gas sensor. As another example, when using exhaust manifold pressure to detect AFR imbalance, the detection may be affected by the distance between the pressure sensor and the cylinder. As the distance increases, exhaust from other cylinders is more likely to mix with the exhaust from the cylinder being estimated. In other words, the reliability of any given approach may vary based on the operating conditions. Therefore, if cylinder fuel or air injection is adjusted in response to an indication of AFR imbalance during conditions where sensor output is not reliable, further AFR and torque issues may be generated.DE 10 2010 020 766 A1 discloses systems and methods which are useful for detecting a combustion fault in an internal combustion engine. The systems and methods include determining cylinder power density values for cylinders present on the engine during operation thereof and determining cylinder imbalance parameters for the cylinders based on the cylinder power density values. The cylinder imbalance parameters are compared to a provided diagnostic threshold.DE 10 2013 114 435 A1 discloses a method for adjusting an engine of an engine operating parameter based on the exhaust gas pressure, wherein the exhaust gas pressure is estimated based on the wastegate actuator motor currentThe object of the invention is to overcome the disadvantages known from the prior art.The object of the invention is achieved by the method according to the invention having the features of claims 1 and 12 and by a machine system having the features of claim 18.In one example, the issues described above may be at least partially addressed by a method comprising: indicating cylinder-to-cylinder imbalance based on each of the exhaust air-fuel ratio estimated by an exhaust gas sensor, the exhaust manifold pressure estimated by a pressure sensor, and the individual cylinder torque estimated by a crankshaft torque sensor. In this way, cylinder-to-cylinder imbalance may be more reliably identified over a wider range of engine operating conditions over a given drive cycle.As one example, exhaust AFR, exhaust manifold pressure, and individual cylinder torque may be estimated at different operating conditions, respectively, over a given drive cycle. Cylinder-to-cylinder imbalance may be identified by weighting each of the estimated exhaust AFR, the estimated exhaust manifold pressure, and the estimated individual cylinder torque with a confidence factor. The confidence factor may be adjusted based on the type of estimation and the operating conditions at which the type of estimation was performed. For example, the confidence factor of imbalance estimation based on the output of an exhaust gas sensor may be decreased during conditions when exhaust mixing is lower and increased during conditions when exhaust mixing is higher. The confidence factor of an AFR estimate based on the output of a pressure sensor may be decreased while the distance between the pressure sensor and the exhaust valve of the cylinder is greater than a threshold, and increased if the distance is less than the threshold. Similarly, the confidence factor of imbalance estimation based on the output of a crankshaft torque sensor may be increased during cold start and decreased during steady-state operation. As a result, a cylinder-specific imbalance estimate collected during a drive cycle at less reliable conditions may be weighted less, while a cylinder-specific imbalance estimate collected during the drive cycle at more reliable conditions may be weighted higher. The method therefore exhibits disadvantages of any individual estimation approach to be overcome which improve the overall precision and reliability of cylinder imbalance estimation.It is to be understood that the summary above is provided to introduce in simplified form a selection of concepts that are described in more detail in the detailed description. It is not intended to identify key or essential features of the claimed subject matter, the scope of which is defined uniquely by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure.Brief Description of the DrawingsFIG. 1 is a schematic diagram of an example machine system.FIGS. 2A and 2B are schematic diagrams of a laser crankshaft torque sensor and a magnetic crankshaft torque sensor. FIG. 3 shows a high level flow diagram of an example method for identifying cylinder-to-cylinder imbalance. FIG. 4 shows a high level flow diagram of an alternative example method for identifying cylinder-to-cylinder imbalance. FIG. 5 shows an example method for intrusively identifying cylinder-to-cylinder imbalance. FIG. 6 shows an example method for estimating individual cylinder torque with a laser or magnetic crankshaft torque sensor.DETAILED DESCRIPTIONThe following description relates to systems and methods for identifying cylinder-to-cylinder imbalance in a vehicle. FIG. 1 is a schematic illustration of a cylinder in a multi-cylinder engine system. The engine system may include an exhaust gas sensor for sensing exhaust AFR, a pressure sensor for sensing exhaust manifold pressure, and a crankshaft torque sensor for sensing individual cylinder torque. A crankshaft torque sensor using lasers and a crankshaft torque sensor using hard disk pulse pickup technology are shown in FIGS. 2A and 2B, respectively. An engine controller may be configured to execute a control routine, such as the routine of FIGS. 3-4, to identify cylinder-to-cylinder imbalance based on the captured data. These may include passive or intrusive Unausgewogenheitsdiagnoseverfahren as shown in FIGS. 5-6. By weighting the data variously collected from the various sensors based on the operation states at which the data was acquired, the reliability of the cylinder imbalance estimation is improved. FIG. 1 illustrates a diagram showing a cylinder of a multi-cylinder engine 10 that may be included in a propulsion system of a motor vehicle. The engine 10 may be controlled at least in part by a control system including the controller 12 as well as by input from a vehicle driver 132 via an input device 130. In this example, the input device 130 includes an accelerator pedal and a pedal position sensor 134 for generating a proportional pedal position signal PP. The combustion chamber (i.e., cylinder) 30 of the engine 10 may include combustion chamber walls 32 with a piston 36 positioned therein. The piston 36 may be coupled to the crankshaft 40 such that reciprocating motion of the piston is converted to rotational motion of the crankshaft. The crankshaft 40 may be coupled to at least one drive wheel of a vehicle via an intermediate transmission system. Crankshaft 40 may also be coupled to a starter motor via a flywheel to enable starting operation of engine 10. Further, a crankshaft torque sensor may be coupled to crankshaft 40 for monitoring engine torque. In an exemplary embodiment, the torque sensor may be a laser torque sensor (as shown in FIG. 2A ) or a magnetic torque sensor (as shown in FIG. 2B ). Other torque sensors may also be used. As set forth in FIGS. 3-4, an engine controller may derive cylinder torque imbalance based on the weighted output of the torque sensor.The combustion chamber 30 may receive intake air from an intake passage 42 via the intake manifold 44 and may exhaust the combustion gases via the exhaust passage 48. The intake manifold 44 and the exhaust passage 48 may be selectively connected to the combustion chamber 30 via an intake valve 52 and an exhaust valve 54, respectively. In some embodiments, combustion chamber 30 may include two or more intake valves and / or two or more exhaust valves. In this example, intake valve 52 and exhaust valve 54 may be controlled by cam actuation via one or more cams controlled via cam profile switching (CPS) and / or variable cam control (VCT) and / or variable valve control (VVT) and / or variable valve lift (VVL) systems that may be operated by controller 12 to vary valve operation. The position of intake valve 52 and exhaust valve 54 may be determined by position sensors 55 and 57, respectively. In alternative embodiments, intake valve 52 and / or exhaust valve 54 may be controlled by electric valve actuation.For example, cylinder 30 may alternatively include an intake valve controlled via electric valve actuation and an exhaust valve controlled via cam actuation including CPS and / or VCT systems.In some embodiments, each cylinder of engine 10 may be configured with one or more fuel injectors to supply fuel thereto. As a non-limiting example, cylinder 30 is shown including a fuel injector 66 supplied with fuel from fuel system 172. Fuel injector 66 is shown coupled directly to cylinder 30 to inject fuel directly therein in proportion to the pulse width of a signal FPW received from controller 12 via an electronic driver 68. In this way, fuel injector 66 provides so-called direct injection (hereinafter also referred to as "DI" (direct injection)) of fuel into combustion cylinder 30.It should be appreciated that in an alternative embodiment, the injector 66 may be an intake port injector that supplies fuel to the intake port upstream of the cylinder 30. It is also understood that cylinder 30 may receive fuel from a plurality of injectors, such as a plurality of port injectors, a plurality of direct injectors, or a combination thereof.Continuing with FIG. 1, intake passage 42 may include a throttle 62 having a throttle plate 64. In this particular example, the position of throttle plate 64 may be varied by controller 12 via a signal provided to an electric motor or actuator present in throttle 62, a configuration commonly referred to as electronic throttle control (ETC). As such, throttle 62 may be operated to vary the intake air provided to combustion chamber 30 among the other engine cylinders. The position of throttle plate 64 may be provided to controller 12 by a throttle position signal TP. Intake passage 42 may include a mass air flow sensor 120 and a manifold air pressure sensor 122 to provide respective signals MAF and MAP to controller 12.The ignition system 88 may provide spark to the combustion chamber 30 via the spark plug 92 in response to the spark advance signal SA from the controller 12 in selected operating modes. Although spark ignition components are shown, in some embodiments, combustion chamber 30, or one or more other combustion chambers of engine 10, may be operated in a compression ignition mode with or without spark.A pressure sensor 124 may be coupled to exhaust passage 49 downstream of exhaust valve 54 and upstream of emission control device 70. The pressure sensor 124 is preferably positioned proximate the exhaust valve 54 to measure exhaust manifold pressure (EMP). In one embodiment, the pressure sensor may be a pressure transducer. As set forth in FIGS. 3-4, an engine controller may derive cylinder torque imbalance based on the weighted output of the torque sensor.An upstream exhaust gas sensor 126 is shown coupled to exhaust passage 48 upstream of emission control device 70. The upstream sensor 126 may be any suitable sensor for providing an indication of exhaust gas air-fuel ratio, such as a linear wide-band oxygen sensor or universal or wide-range exhaust gas oxygen (UEGO), a two-state narrow-band oxygen sensor or EGO, a heated EGO (HEGO) sensor, a NOx, HC, or CO sensor. In one embodiment, the upstream exhaust gas sensor 126 is a UEGO configured to provide an output, such as a voltage signal proportional to the amount of oxygen present in the exhaust gas. The controller 12 uses the output to determine the exhaust air-fuel ratio. As set forth in FIGS. 3-4, an engine controller may derive cylinder torque imbalance based on the weighted output of the exhaust gas sensor.Emission control device 70 is shown arranged along exhaust passage 48 downstream of exhaust gas sensor 126. The device 70 may be a three-way catalyst (TWC) configured to reduce NOx and oxidize CO as well as unburned hydrocarbons. In some embodiments, device 70 may be a NOx trap, various other emission control devices, or combinations thereof.A second downstream exhaust gas sensor 128 is shown coupled to exhaust passage 48 downstream of emission control device 70. The downstream sensor 128 may be any suitable sensor for providing an indication of exhaust gas air-fuel ratio, such as a UEGO, EGO, HEGO, etc. In one embodiment, the downstream sensor 128 is a HEGO configured to indicate the relative enrichment or depletion of the exhaust gases after they have passed through the catalyst. The HEGO may therefore provide output in the form of a switching point or the voltage signal at the point where the exhaust gas transitions from lean to rich.Further, in the disclosed embodiments, an exhaust gas recirculation (EGR) system may route a desired portion of exhaust gas from the exhaust passage 48 to the intake passage 42 via the EGR passage 140. The amount of EGR provided to the intake passage 42 may be varied by the controller 12 via the EGR valve 142. Further, an EGR sensor 144 may be disposed in the EGR conduit and may provide an indication of pressure and / or temperature and / or exhaust concentration. Under some conditions, the EGR system may be used to regulate the temperature of the air and fuel mixture within the combustion chamber.Controller 12 is shown in FIG. 1 as a microcomputer, including microprocessor unit 102, input / output ports 104, an electronic storage medium for executable programs and calibration values, shown as read only memory chip 106 in this particular example, random access memory 108, keep alive memory 110, and a data bus. Controller 12 may receive various signals from sensors coupled to engine 10, in addition to the signals discussed above, including measurement of inducted mass air flow (MAF) from mass air flow sensor 120, exhaust manifold pressure (EMP) from pressure sensor 124, engine coolant temperature (ECT) from temperature sensor 112 coupled to cooling sleeve 114, a profile spark pickup signal (PIP) from Hall effect sensor 118 (or other type of sensor) coupled to crankshaft 40, cylinder torque from the crankshaft torque sensor coupled to crankshaft 40, throttle position (TP) from a throttle position sensor, and absolute manifold pressure (MAP) signal from sensor 122. The engine speed, RPM, signal may be generated from the PIP signal by the controller 12. The controller 12 may also use various actuators of FIG. 1 to adjust machine operation based on the received signals and instructions stored in a memory of the controller.The read-only memory storage medium 106 may be programmed with computer readable data representing non-transitory instructions executable by the processor 102 for performing the methods described below as well as other variations that are anticipated but not specifically recited.As described above, FIG. 1 shows only one cylinder of a multi-cylinder engine, and each cylinder may also have its own set of intake / exhaust valves, fuel injection valve, spark plug, etc.Referring to FIG. 2A, an example laser crankshaft torque sensor 200 is shown. The torque sensor is positioned in the torque bearing area of the crankshaft. For example, the torque sensor may be positioned between the last cylinder of the engine block and the engagement of the shaft with a gear coupled to a damper or transmission. The torque sensor may include at least two sets of lasers / detectors and at least two encoders. As an example, FIG. 2A shows a torque sensor that includes two laser / detector sets (201 and 202) and two encoders (207 and 208). The laser / detector sets are mounted on a structure that holds the crankshaft 40 and positioned at a common distance from the crankshaft 40. The lasers and detectors are arranged facing the crankshaft 40. The encoders are circumferentially applied to the crankshaft 40. Each encoder has a code (for example, a bar code) configured as multiple alternate black and white lines parallel to the crankshaft 40. In another embodiment, encoders 207 and 208 may connect to one another and form a single encoder.Each set of lasers / detectors measures the position of the crankshaft by intermittently transmitting laser pulses to the corresponding encoder and receiving laser pulses reflected from the encoder. In one example, each set of lasers / detectors includes a laser 222 and a detector 221 positioned in contact with each other. The laser 222 in the laser / detector set 201 sends laser pulses 205 to the encoder 207, and the detector 221 in the laser / detector set 201 receives reflected laser pulses 203 from the encoder 207. Distortion of the crankshaft 40 may be determined by comparing the two measured position signals from the two sets of lasers / detectors. The engine torque may be further calculated based on the estimated warp. An example method for estimating engine torque using the laser torque sensor is shown in FIG. 6.FIG. 2B shows an example magnetic crankshaft torque sensor 210. Similar to the laser torque sensor 200, the magnetic torque sensor 210 is positioned in the torque bearing area of the crankshaft. The magnetic torque sensor 210 includes at least two hard disk type pulse transducers mounted on a structure that supports the crankshaft 40 and that are positioned a common distance from the crankshaft. Each pulse transducer points to a corresponding encoder that is circumferentially attached to the crankshaft 40. As an example, FIG. 2B shows a torque sensor including two magnetic pulse receivers (211 and 213) and two encoders (212 and 214). The magnetic pulse pickups 211 and 213 can accurately measure the position of the crankshaft 40 at two locations by reading respective encoders 212 and 214. The encoders may be made of plastic with metal layer or permanent magnetic material. In one embodiment, encoders 207 and 208 may connect to one another and form a single encoder.Distortion in the crankshaft 40 can be determined by comparing the two measured position signals from the two pulse transducers. The engine torque may be further calculated based on the estimated warp. An example method for estimating engine torque based on the magnetic torque sensor is shown in FIG. 6.Referring to FIG. 3, routine 300 depicts a method for identifying cylinder-to-cylinder imbalance based on combined information of exhaust AFR, exhaust manifold pressure, and individual cylinder torque. Routine 300 further includes adjusting engine operation in response to the identified cylinder-to-cylinder imbalance.Instructions for executing the method 300 as well as the remainder of the methods included herein may be executed by a controller 12 based on instructions stored in a memory of the controller and associated with signals received from sensors of the machine system, such as the sensors described above with reference to FIGS. 1 and 2A-B. The controller may use engine actuators of the engine system to adjust engine operation according to the methods described below.At 301, routine 300 reads a diagnostic code stored in the memory of controller 12. A diagnostic code associated with cylinder imbalance may be fetched from memory of the controller, for example. In an alternative example, the controller may retrieve a cylinder imbalance history from memory, with details stored relating to the tendency of each cylinder to torque errors, as well as directionality (positive or negative) of the torque error. Then, at 302, method 300 estimates or measures engine operating conditions. The estimated conditions may include, for example, engine speed, torque demand, boost pressure, MAP, engine temperature, combustion AFR, exhaust catalyst temperature, ambient conditions, etc.At 303, method 300 determines whether an intrusive cylinder-to-cylinder imbalance diagnostic should be performed based on the estimated engine operating conditions. As an example, the intrusive imbalance diagnostic may be performed during engine idle conditions. As another example, the intrusive imbalance diagnostic may be performed during steady state engine conditions when operating at medium to low engine load. If the controller determines that conditions are present for executing the intrusive method, routine 300 proceeds to 304 where cylinder-to-cylinder imbalance is determined by actively adjusting the combustion AFR. The intrusive imbalance procedure is explained in detail in Figure 5. If the answer at 303 is NO, routine 300 proceeds to 305 and cylinder-to-cylinder imbalance is passively diagnosed.At 305, passive cylinder imbalance diagnostic includes estimating exhaust AFR, exhaust manifold pressure, and cylinder torque for each engine cylinder. Specifically, for each cylinder i, the exhaust air-fuel ratio LAM ¡ is estimated by an exhaust gas sensor (such as by the exhaust gas sensor 126 of FIG. 1 ), the exhaust manifold pressure P i is estimated by a pressure sensor (such as by the pressure sensor 124 of FIG. 1 ), and the individual cylinder torque TQ i is estimated by a crankshaft torque sensor (such as by a laser torque sensor or a magnetic crankshaft torque sensor of FIGS. 2A and 2B ).At 306, based on the estimated operating conditions, confidence factors are determined for each AFR imbalance estimate performed at step 305. In particular, a first confidence factor c 1 may be determined for the AFR estimate, a second confidence factor c 2 may be determined for the exhaust manifold pressure estimate, and a third confidence factor c 3 may be determined for the individual cylinder torque estimate. The confidence factor of a given estimate therefore reflects the reliability or precision of the cylinder-to-cylinder imbalance estimate based on the given estimate. The confidence factor may then be set to a maximum value of 1.0 (indicating maximum confidence), or it may be set to the lowest value of zero if the estimate is not available or not reliable. The confidence factors may be further set to any number between zero and one based on the operating state at which the estimation is performed. A higher confidence factor value indicates that the imbalance estimate is more reliable, while a lower confidence factor value indicates that the imbalance estimate is less reliable. The confidence factor for a given estimation method may therefore vary based on the operating conditions under which the estimation was performed. Further, the confidence factor may be different for each cylinder. The confidence factors may also be adjusted while the sum of the confidence factors is maintained at a constant value so that measurements between different drive cycles may be compared.As one example, the first confidence factor c 1 for the AFR estimate may be decreased when the mixing of exhaust at the exhaust gas sensor is below a threshold. In one embodiment, sufficient mixing may be estimated by observing whether an AFR change in a particular cylinder is reflected by the reading of the exhaust gas sensor, for example during depletion or fuel cut. The exhaust gas sensor may not have the same sensitivity for all cylinders because the exhaust gas sensor may be located at a location in the exhaust flow that always receives exhaust from some cylinders but no exhaust from the other cylinders. The first confidence factor may therefore be different for each cylinder. As another example, the first confidence factor c 1 may be decreased during engine cold-start conditions because AFR from the exhaust gas sensor may not be accurately estimated due to insufficient warming of the exhaust gas sensor. As another example, the first confidence factor may be decreased when the exhaust temperature is less than a threshold. As yet another example, the confidence factor for the exhaust AFR measured within a predetermined amount of time after the engine cold start (or a predetermined number of combustion events since a first combustion event of the cold start). In another example, the first confidence factor may be decreased if the exhaust AFR is estimated during fuel type adjustment (such as when transitioning from gasoline to ethanol fuel, or when transitioning from supplying fuel from a first fuel tank to a second fuel tank), or if the fuel type is unknown. Since exhaust AFR cannot be calculated robust without information on the exhaust type, less confidence is given to the first confidence factor c1 if the fuel composition cannot be estimated reliably. In yet another example, the first confidence factor may be increased if the machine is in rich operation.In another example, the second confidence factor for estimating exhaust manifold pressure may be decreased as the distance between the pressure sensor and the exhaust valve of the cylinder increases. The farther the pressure sensor is from the cylinder, the greater the chance that exhaust from other cylinders will mix with the exhaust from the cylinder being estimated. In another example, the second confidence factor may be decreased if the variation in valve timing is within a threshold.As another example, the third confidence factor may be increased if the engine is in lean operation. As another example, the third confidence factor may be increased when the first and second confidence factors are decreased. After determining the confidence factors, routine 300 proceeds to 307.At 307, routine 300 estimates the average exhaust AFR (LAMaavg), the average exhaust pressure (Pave), and the average cylinder torque (TQave) for all cylinders. As one example, the estimate may be based on the combustion ignition event timing of each cylinder. As another example, the average may be calculated by averaging the estimate acquired at 305. The averaged AFR for an n-cylinder engine may be calculated, for example, as follows:At 308, a combined imbalance parameter for each cylinder is calculated based on the confidence factors and the weighted imbalance estimates. As an example, the combined imbalance parameter for the i-th cylinder may be calculated as follows:At 309, one or more cylinders with imbalanced AFR are identified based on a comparison of the combined imbalance parameters calculated for each cylinder. As an example, an imbalanced cylinder may be identified if the combined imbalance parameter for a given cylinder has a greater deviation from the averaged combined imbalance parameter for all cylinders. As another example, one or more imbalanced cylinders may be identified if the deviation of the combined imbalance parameter for any of the cylinders from an average of the combined imbalance parameter for all cylinders is greater than a predetermined threshold. The amount of deviation may correspond to the amount of imbalance, and the sign of the deviation may correspond to the direction of imbalance. For example, if the combined imbalance parameter for a cylinder is less than the average of the combined imbalance parameters (a negative deviation), the cylinder may be considered a rich outlier. As another example, if the combined imbalance parameter for a cylinder is higher than the average of the combined imbalance parameters (a positive deviation), the cylinder may be considered a lean outlier.At 310, the diagnostic code for the imbalanced cylinders identified at 309 is updated. In one embodiment, the diagnostic code for the imbalanced cylinder may be changed based on the combined imbalanced parameter determined at step 308. The diagnostic code may be updated, for example, based on the difference between the current combined imbalance parameter and the diagnostic code read at 301. In another embodiment, the diagnostic code may be updated based on the deviation of the combined imbalance parameter of the imbalanced cylinder from the average combined imbalance parameter of all cylinders. In addition, imbalance history of the engine cylinders may be updated.At 311, routine 300 includes applying an AFR correction to the one or more cylinders indicated as imbalanced. For example, an AFR correction may be applied to identified cylinders based on the identified amount and direction of air-fuel imbalance in the identified cylinder. For example, controller 12 may adjust the amount of fuel supplied to cylinders identified as potentially imbalanced. The controller 12 may then continue monitoring air-fuel imbalance in an attempt to correct the air / fuel imbalance in the identified cylinders. Specifically, the fuel injector pulse width of the fuel injector actuator of the imbalanced cylinder fuel injector is adjusted to provide the corrected fuel injection amount.FIG. 4 shows an alternative method 400 for identifying cylinder-to-cylinder imbalance. Unlike the method shown in FIG. 3, routine 400 performs different imbalance estimates at different operating conditions. In this way, method 400 may reliably determine imbalance over a wide range of operating conditions without interrupting engine operation. In addition, the duration required to identify imbalanced cylinders may be reduced.Similar to steps 301 through 304 of routine 300 shown in FIG. 3, routine 400 reads the stored diagnostic code at 401, and estimates engine operating conditions at 402. If an intrusive imbalance estimate is determined at 403, method 400 proceeds to 404 where the engine is actively depleted to identify lean or rich outliers (as set forth at FIG. 5 ). Otherwise, routine 400 determines the type of imbalance estimation to be performed based on operating conditions. Specifically, routine 400 proceeds to 405 in a first operating state, to 409 in a second operating state, and to 413 in a third operating state. In one example, the first, second, and third states are mutually exclusive.At 405, when the engine is in the first operating state, an AFR for each cylinder is determined by the exhaust gas sensor at 406. The first operating state may be a steady state medium load state or a steady state idle state. Further, the first operating state may be when the type of fuel injected into the cylinder is known. During the first state, for example, the percentage of ethanol in the injected fuel may be known. Additionally, the first operating state may include exhaust gas being sufficiently mixed at the exhaust gas sensor. The first operating state may also include the exhaust gas sensor being sufficiently warmed up. The first operating state may further include rich engine operation (where the engine is operated richer than stoichiometric). The first operating state may therefore comprise any of the mentioned operating states or any of their combinations.At 407, an average exhaust AFR during the first operating state is estimated. As one example, the estimate may be based on the combustion ignition event timing of each cylinder. As another example, the average may be calculated by averaging the AFR estimate for each cylinder.At 408, the first confidence factor is determined for the first operating state. Similar to step 306 in FIG. 3, the confidence factor may be adjusted to reflect the reliability of the estimation performed at the given operating state (here, the first operating state). The confidence factor may then be set to the highest value of one to assign the highest confidence to the estimate, and may be set to the lowest value of zero if the estimate is not available. Further, the confidence factor may be set to a number between zero and one based on the operating state at which the estimation is performed.The first confidence factor may be increased at a lower combustion AFR and decreased at a higher combustion AFR. As another example, the confidence factor may be increased with better mixing of exhaust at the exhaust gas sensor, and decreased with less sufficient mixing of the exhaust at the exhaust gas sensor.If the vehicle is in the second operating state at 409, exhaust pressure is estimated by a pressure sensor positioned at 410 on the exhaust manifold. The second state may be a medium load steady state or an idle steady state. Further, the second operating state may be when the variation of the valve timing is within a threshold. Further, the second operating state may be present if the distance between the pressure sensor and the exhaust valve of the cylinder is less than a threshold. The second operating state may therefore include any of the above-mentioned operating states or any of their combinations.At 411, an average exhaust pressure during the second operating state is estimated. As one example, the estimate may be based on the combustion ignition event timing of each cylinder. In another example, the average may be calculated by averaging the exhaust pressure estimate for each cylinder.At 412, the second confidence factor for the second operating state is determined. The second confidence factor may be increased with less variation in valve timing and decreased with greater variation in valve timing. The confidence factor may be further adjusted lower if the distance between the pressure sensor and the exhaust valve of the cylinder is greater than a threshold and increase if the distance is less than the threshold.If the vehicle is in the third operating state at 413, individual cylinder torque is estimated by a torque sensor coupled to the crankshaft at 414. The third state may be a cold start state. For example, the cold start condition may be determined when the exhaust temperature is less than a threshold. Further, the third state may include the exhaust gas not being sufficiently mixed at the exhaust gas sensor. Additionally, the third state may be lean engine operation. The third operating state may therefore include any of the above-mentioned operating states or any of their combinations.At 415, an average engine torque during the third operating state is estimated. As one example, the estimate may be based on the combustion ignition event timing of each cylinder. As another example, the average may be calculated by averaging the torque estimate for each cylinder.At 416, a third confidence factor for the third operating state is determined. The third confidence factor may be decreased with better mixing of the exhaust at the exhaust gas sensor, and increased with less sufficient mixing of the exhaust at the exhaust gas sensor. The third confidence factor may be increased at leaner combustion AFR and decreased at richer combustion AFR.At 417, the engine controller determines whether there is any change in operating conditions and thus determines whether continued imbalance diagnostic is required. In one example, imbalance diagnostic may be discontinued if operating conditions deviate by more than a threshold amount or at a higher than the threshold frequency. If a change in operating condition is detected and continued imbalance diagnostic is not required, routine 400 proceeds to 418 to determine current operating conditions. Based on current operating conditions, routine 400 proceeds to 405, 409, or 413 for further estimation. On the other hand, if the controller determines at 417 that sufficient data has been acquired and that imbalance may be reliably determined based on the acquired data, routine 400 proceeds to 419.At 419, one or more cylinders with imbalanced AFR are determined. Similar to steps 308 and 309 in FIG. 3, a combined imbalance parameter for each cylinder is calculated according to Equation 2, and then the imbalance cylinder is identified based on the deviation of the combined imbalance parameter for any of the cylinders from the average of the combined imbalance parameters of all cylinders. The amount of deviation may correspond to the amount of imbalance, and the sign of the deviation may correspond to the direction of imbalance. Routine 400 then proceeds to step 420, where the diagnostic code is updated similar to step 310 in FIG. 3.At 421, the corresponding AFR correction is applied based on the determined imbalance. Similar to step 311 in FIG. 3, the AFR correction may be applied to the identified cylinder based on the identified amount and direction of air-fuel imbalance in the identified cylinder. Further, the applied AFR correction may be different based on operating conditions at which imbalance is identified. For example, during a first operating state, responsive to imbalance, the AFR of an imbalanced cylinder may be adjusted via only fuel settings. During a second state, the imbalanced cylinder AFR may be adjusted via intake air settings only. During a third state, the imbalanced cylinder AFR may be adjusted via both fuel and intake air settings.Referring to FIG. 5, routine 500 identifies cylinders with imbalanced AFR by intrusively depleting each or all cylinders and monitoring the response of each cylinder. By depleting each of the cylinders, lean outliers may be identified. Rich outliers can be identified by depleting all cylinders.At 510, routine 500 determines whether a lean outlier should be identified. By way of example, the lean outlier may be identified when engine operation is sufficiently affected by depleting any (but not all) of the cylinders significantly. If the answer at 510 is YES, routine 500 proceeds to 511 where the AFR is sequentially depleted for each of the cylinders. If the answer at 510 is NO, routine 500 proceeds to 520 where the diagnostic code remains unchanged.At 511, the method includes sequentially increasing the AFR for each of the cylinders. The order of depletion may be based on cylinder firing order or cylinder position along an engine block. In response to increased AFR of the enriched cylinder, the variations in exhaust AFR, exhaust pressure, and individual cylinder torque of the enriched cylinder are estimated at 512.At 513, confidence factors are determined for each type of estimate performed at 512. For example, a first confidence factor is determined for the variation in exhaust AFR, a second confidence factor is determined for the variation in exhaust pressure, and a third confidence factor value is determined for the variation in cylinder torque. The confidence factors may be determined based on engine operating conditions similar to step 306 in FIG. 3. For example, the third confidence factor may be set higher than the first confidence factor because cylinder torque is more sensitive to AFR imbalance during lean operation.At 514, a lean outlier is identified based on the variation estimated at step 512 and the confidence factor determined at step 513. Each of the variations estimated at step 512 is weighted with corresponding confidence factors determined at step 513, for example. For each cylinder, a summation of the weighted exhaust AFR, the weighted exhaust pressure, and the weighted cylinder torque is calculated. Then, a deviation of the summation for each cylinder is calculated from the average of the summation for all cylinders. As an example, a cylinder with the largest deviation is identified as a lean outlier. As another example, one or more cylinders having a deviation greater than a predetermined threshold are determined to be lean outliers. The lean outlier diagnostic code is updated similar to step 310 in FIG. 3. Then, routine 500 proceeds to 515.At 515, routine 500 determines whether a rich outlier should be identified. By way of example, the rich outlier may be identified when engine operation is not significantly affected by depleting all cylinders. If the answer at 515 is YES, routine 500 proceeds to 516 where the AFR is depleted for all cylinders. If the answer at 515 is NO, routine 500 proceeds to 521 where the diagnostic code remains unchanged.At 516, the method includes depleting all cylinders by increasing combustion AFR. In response to increased AFR, the variation in exhaust AFR, exhaust pressure, and individual cylinder torque for each cylinder is estimated at 517.At 518, similar to step 512, confidence factors are determined for each estimation type performed at 517. The confidence factors may be determined based on engine operating conditions similar to step 306 in FIG. 3. For example, the first confidence factor may be set higher than the third confidence factor because the exhaust AFR is more sensitive to AFR imbalance during rich operation.At 519, similar to step 514, a rich outlier is identified based on the variation estimated at step 517 and the confidence factors determined at step 518. Each of the variations estimated at step 517 is weighted with corresponding confidence factors determined at step 518, for example. For each cylinder, a summation of the weighted exhaust AFR, the weighted exhaust pressure, and the weighted cylinder torque is calculated. Then, a deviation of the summation for each cylinder is calculated from the average of the summation for all cylinders. As an example, a cylinder with the largest deviation is identified as a rich outlier. As another example, one or more cylinders having a deviation greater than a predetermined threshold are determined to be rich outliers. The rich outlier diagnostic code is updated similar to step 310 in FIG. 3.Finally, at 522, similar to step 311 in FIG. 3, the AFR for the one or more imbalanced cylinders is corrected.Referring to FIG. 6, routine 600 shows an example method for determining engine torque responsive to each cylinder firing event using a laser torque sensor or a magnetic torque sensor. Both the laser torque sensor 200 and the magnetic torque sensor 210 shown in FIGS. 2A and 2B may measure crankshaft position at two locations along the crankshaft. The difference between the measured crankshaft positions at the two locations corresponds to a phase delay between the two position signals. The difference between the measured crankshaft positions may therefore be referred to herein as a phase delay value.At 601, routine 600 determines whether zero torque calibration has been performed. If calibration has been performed, routine 600 proceeds to 604 where a stored phase delay value is read. The phase delay may be stored in the memory of the controller, and fetched from the memory, for example. If calibration has not been performed, routine 600 proceeds to 602 for calibration. Further, calibration may be performed if a threshold duration has elapsed since the last calibration.At 602, routine 600 determines whether zero torque is applied to the crankshaft. As one example, zero torque may be applied to the crankshaft during an engine off condition. As another example, zero torque may be applied to the crankshaft during engine idling. If the crankshaft is below zero torque, the method includes reading and storing (in the controller memory) the current phase delay value from the torque sensor at step 605. If the torque applied to the crankshaft is not equal to zero, routine 600 indicates that the cylinder torque measurement is not available.After fetching the zero torque phase retard value at step 604 or 605, routine 600 proceeds to 606 to determine if a cylinder firing event occurs. In response to the cylinder firing event, routine 600 continues to 608 to estimate cylinder torque. If no firing event has occurred, routine 600 continues to monitor the cylinder firing events at 607.At 608, a phase delay value responsive to cylinder firing is determined based on a time delay and speed of the crankshaft. In particular, the time delay between the two locations along the crankshaft is measured by the torque sensor. As an example, at time t 0 the torque sensor records an encoder pattern at a first location. Subsequently, at time t 1, the same encoder pattern is recorded at the second location. Then, the time delay dt is determined as dt = t 1- t 0. The post-cylinder ignition phase retard value (phi) is calculated as: where r is the rotational speed of the crankshaft.At 609, the crankshaft distortion is estimated by subtracting the phase delay value for zero torque from the phase delay value calculated at step 608.Finally, at 610, engine torque after cylinder firing is estimated based on crankshaft torsion at 609 and further based on crankshaft temperature. As one example, the crankshaft temperature may be inferred based on cylinder head temperature. As another example, if the torque sensor is a laser torque sensor, the crankshaft temperature may be measured by operating the laser sensor at infrared wavelength. The cylinder torque values estimated using the laser or magnetic torque sensor may then be applied to the Zylinderunausgewogenheitsschätzung as discussed above with reference to FIGS. 3-4.As such, multiple air-fuel ratio imbalance estimates may be performed in response to varying operating conditions over an engine drive cycle. Then, a combined imbalance parameter is calculated by weighting each estimate with a confidence factor determined based on the operating state at which the estimate is performed. The combined imbalance parameter may therefore provide a more reliable identification of cylinder imbalance compared to any single imbalance estimate. The disadvantages of each estimation can therefore be overcome. Zylinderunausgewogenheitsdiagnosen may also be performed at a wider range of engine operating conditions while decreasing overall diagnostic time. By identifying cylinder imbalance in a more rapid and reliable manner, torque disturbances due to cylinder imbalance may be countered in a more timely manner. This therefore improves engine fuel saving and engine performance as well as vehicle drivability.It should be appreciated that the example control and estimation routines included herein may be used with various machines and / or vehicle system configurations. The control methods and routines disclosed herein may be stored as executable instructions in non-transitory memory and may be executed by the control system including the controller in combination with the various sensors, actuators, and other machine hardware. The particular routines described herein may represent one or more of a number of processing strategies such as event-driven, interrupt-driven, multi-tasking, multi-threading, and the like. Thus, various illustrated acts, operations, and / or functions may be performed in the order illustrated, in parallel, or in some cases omitted. Likewise, the order of processing is not necessarily required to achieve the features and advantages of the example embodiments described herein, but is provided for ease of illustration and description. One or more of the illustrated acts, operations, and / or functions may be repeatedly performed depending on the particular strategy being used. Furthermore, the described acts, operations, and / or functions may graphically represent code to be programmed into non-transitory memory of the computer readable storage medium in the machine control system, where the described acts are performed by executing the instructions in a system including the various machine hardware components in combination with the electronic controller.It should be understood that the configurations and routines disclosed herein are exemplary in nature and that these specific embodiments are not to be interpreted in a limiting sense as numerous variations are possible. The above technology can be applied to, for example, V-6, I-4, I-6, V-12, Boxer-4, and other types of machines. The subject matter of the present disclosure includes all novel and non-obvious combinations and sub-combinations of the various systems and configurations, and other features, functions, and / or characteristics disclosed herein.The following claims particularly point out certain combinations and sub-combinations which are considered novel and not obvious. These claims may refer to "a" element or "a first" element or the equivalent thereof. Such claims should be understood to include one or more such elements, neither requiring nor excluding two or more such elements. Other combinations and sub-combinations of the disclosed features, functions, elements, and / or characteristics may be claimed by altering the present claims or by providing novel claims in this or a related application. Such claims, whether broader, narrower, equal or different than the original claims, are also considered to be within the scope of the present disclosure.

Claims

A method for an engine, comprising: indicating cylinder-to-cylinder imbalance based on each of the exhaust air-fuel ratio (LAMavg) estimated by an exhaust gas sensor (126), the exhaust manifold pressure (Pavg) estimated by a pressure sensor (124), and the individual cylinder torque (TQavg) estimated by a crankshaft torque sensor (200, 210); wherein the indicating based on the exhaust air-fuel ratio (LAMavg) based on the exhaust air-fuel ratio (LAMavg) weighted with a first confidence factor (c1), the indicating based on the exhaust manifold pressure (Pavg) based on the exhaust pressure weighted with a second confidence factor (c2), and the indicating based on individual cylinder torque (TQavg) based on the individual cylinder torque (TQavg) weighted with a third confidence factor (c3); Further comprising computing a combined imbalance parameter based on the weighted exhaust air-fuel ratio (LAMavg), the weighted exhaust manifold pressure (Pavg), and the weighted individual cylinder torque (TQavg) for each engine cylinder; and by means of a controller (12) having computer readable instructions stored in a non-transitory memory, determining a first confidence factor (c1) of the estimated exhaust air-fuel ratio (LAMavg) based on fuel type, determining a second confidence factor (c2) of the estimated exhaust pressure based on valve control, determining a third confidence factor (c3) of the estimated individual cylinder torque (TQavg) based on exhaust gas temperature, indicating cylinder-to-cylinder imbalance based on two or more of the estimated exhaust air-fuel ratio (LAMavg) weighted with the first confidence factor (c1), the estimated exhaust manifold pressure (Pavg) weighted with the second confidence factor (c2), and the estimated individual cylinder torque (TQavg) weighted with the third confidence factor (c3), and applying an air-fuel ratio correction to one or more cylinders (30) based on the indication.The method of claim 1, further comprising, in response to the indication, applying an air-fuel ratio correction to an imbalanced cylinder (30), the air-fuel ratio correction having a corrected fuel injection amount, wherein a fuel injector pulse width of a fuel injector actuator is adjusted to provide the corrected fuel injection amount.The method of claim 1, further comprising setting a diagnostic code if a deviation of the combined imbalance parameter for any of the engine cylinders from an average of the combined imbalance parameter for all engine cylinders is greater than a threshold.The method of claim 1, wherein one or more of the first, second and third confidence factors (c1, c2, c3) are adjusted based on engine operating conditions.The method of claim 4, wherein the adjusting comprises decreasing the first confidence factor (c1) during conditions where exhaust mixing at the exhaust gas sensor (126) is below a threshold.The method of claim 4, wherein the adjusting comprises decreasing the first confidence factor (c1) during engine warm-up after a cold start.The method of claim 4, wherein the adjusting comprises decreasing the first confidence factor (c1) during the fuel type adjustment.The method of claim 4, wherein the adjusting comprises decreasing the first confidence factor (c1) during rich operation.The method of claim 4, wherein the adjusting comprises decreasing the second confidence factor (c2) as the distance between the pressure sensor (124) and the cylinder (30) is increased.The method of claim 4, wherein the adjusting comprises decreasing the third confidence factor (c3) during lean operation.The method of claim 1, wherein the individual cylinder torque (TQavg) is estimated by a laser torque sensor (200) or a magnetic torque sensor (210) coupled to the crankshaft (40).A method for an engine, comprising: indicating cylinder-to-cylinder imbalance based on the exhaust air-fuel ratio (LAMavg) estimated by an exhaust gas sensor (126) during a first state, the exhaust manifold pressure (Pavg) estimated by a pressure sensor (124) during a second state, and the individual cylinder torque (TQavg) estimated by a crankshaft torque sensor (210) during a third state; and wherein the indicating based on the exhaust air-fuel ratio (LAMavg) based on the exhaust air-fuel ratio (LAMavg) weighted with a first confidence factor (c1), the indicating based on the exhaust manifold pressure (Pavg) based on the exhaust pressure weighted with a second confidence factor (c2), and the indicating based on individual cylinder torque (TQavg) based on the individual cylinder torque (TQavg) weighted with a third confidence factor (c3); Further comprising computing a combined imbalance parameter based on the weighted exhaust air-fuel ratio (LAMavg), the weighted exhaust manifold pressure (Pavg), and the weighted individual cylinder torque (TQavg) for each engine cylinder; and by means of a controller (12) having computer readable instructions stored in a non-transitory memory, determining a first confidence factor (c1) of the estimated exhaust air-fuel ratio (LAMavg) based on fuel type, determining a second confidence factor (c2) of the estimated exhaust pressure based on valve control, determining a third confidence factor (c3) of the estimated individual cylinder torque (TQavg) based on exhaust gas temperature, indicating cylinder-to-cylinder imbalance based on two or more of the estimated exhaust air-fuel ratio (LAMavg) weighted with the first confidence factor (c1), the estimated exhaust manifold pressure (Pavg) weighted with the second confidence factor (c2), and the estimated individual cylinder torque (TQavg) weighted with the third confidence factor (c3), and applying an air-fuel ratio correction to one or more cylinders (30) based on the indication.The method of claim 12, wherein the indicating is further based on the exhaust air-fuel ratio (LAMavg) weighted with a first confidence factor (c1), the exhaust pressure weighted with a second confidence factor (c2), and the individual cylinder torque (TQavg) weighted with a third confidence factor (c3).The method of claim 12, wherein the first state comprises steady state engine operation of known fuel type.The method of claim 12, wherein the second state comprises steady state engine operation and the variation of the valve timing is within a threshold.The method of claim 12, wherein the third state comprises engine cold start state.The method of claim 12, further comprising: during the first state, responsive to imbalance, adjusting an air-fuel ratio of an imbalanced cylinder (30) via fuel adjustments, during the second state, responsive to imbalance, adjusting the air-fuel ratio of an imbalanced cylinder (30) via intake air adjustments, and during the third state, responsive to imbalance, adjusting the air-fuel ratio of the imbalanced cylinder (30) via both fuel and intake air adjustments.An engine system comprising: an engine having a plurality of cylinders (30); an exhaust gas sensor (126) for estimating an exhaust air-fuel ratio (LAMavg); a pressure sensor (124) for estimating an exhaust manifold pressure (Pavg); a crankshaft torque sensor (210) coupled to a crankshaft (40) for estimating individual cylinder torque (TQavg); and a controller (12) configured with computer readable instructions stored in non-transitory memory to: determine a first confidence factor (c1) of the estimated exhaust air-fuel ratio (LAMavg) based on fuel type; determine a second confidence factor (c2) of the estimated exhaust pressure based on valve control, a third confidence factor (c3) of the estimated individual cylinder torque (TQavg) based on exhaust temperature, to indicate cylinder-to-cylinder imbalance based on two or more of the estimated exhaust air-fuel ratio (LAMavg) weighted with the first confidence factor (c1), the estimated exhaust manifold pressure (Pavg) weighted with the second confidence factor (c2), and the estimated individual cylinder torque (TQavg) weighted with the third confidence factor (c3), and to apply an air-fuel ratio correction to one or more cylinders (30) based on the indication.

Images