Methods and systems for an oxygen sensor

Abstract

A method for a power engine, comprising: learning a reference point for an inlet oxygen sensor at a reference inlet pressure under selected power engine idle conditions; and adjusting an EGR flow to the power engine based on an inlet oxygen concentration estimated by the sensor with respect to the learned reference point and further based on a change in the inlet pressure with respect to the reference inlet pressure, wherein adjusting the EGR flow to the power engine comprises adjusting low-pressure EGR flow along an EGR channel from an exhaust manifold downstream of an exhaust turbine to an inlet manifold upstream of an inlet compressor via an EGR valve, wherein the learned reference point is a first learned reference point, and wherein the method further comprises learning a second reference point for the inlet oxygen sensor at the reference inlet pressure under selected power engine non-fuel supply conditions.

Description

The present application relates generally to a gas component sensor included in an intake system of an internal combustion engine.Engine systems may use recirculation of exhaust gas from an engine exhaust system to an engine intake system (intake passage), a process referred to as exhaust gas recirculation (EGR), to reduce regulated emissions. An EGR system may include various sensors for measuring and / or controlling EGR. As an example, the EGR system may include an intake gas constituent sensor, such as an oxygen sensor, that may be used to measure oxygen to determine the proportion of gases burned in an intake passage of the engine. The sensor may also be used under non-EGR conditions to determine the oxygen content of fresh intake air. Additionally or optionally, the EGR system may include an exhaust oxygen sensor coupled to the exhaust manifold for estimating a combustion air-fuel ratio.Thus, when the intake oxygen sensor is used for EGR control, EGR is measured as a function of the change in oxygen due to EGR as a diluent. For determining the change in oxygen amount, a reference point corresponding to an oxygen measurement when there is no EGR is required. Such a reference point is referred to as the "zero point" of the oxygen sensor. Due to the sensitivity of the oxygen sensor to pressure and aging, there may be "zero point" deviations under various engine operating conditions. In particular, the greatest cause of a change in the zero point oxygen measurement may be due to aging and part-to-part variability. Therefore, the oxygen sensor may need periodic calibration and a correction factor needs to be learned.An example method of calibrating an exhaust gas oxygen sensor is shown by Ishiuro et al. in U.S. Pat. No. 8,417,413 B2. Therein, a correction factor is learned based on an oxygen sensor output under engine fuel cut conditions. However, the inventors have recognized that approaches used for zero point estimation in exhaust oxygen sensors may not be applied to zero point estimation of intake oxygen sensors. This is because, in addition to sensitivity to pressure and part-to-part variability due to equilibration of the sensed gas by a catalyzing sensing element of the sensor, the oxygen sensor is also sensitive to the presence of fuel or other reducing agents and oxidizing agents. As a result, the output of the intake oxygen sensor may be affected by the presence of scavenge hydrocarbons and / or crankcase ventilation gases received at the engine intake under calibration conditions. The sensor measurement may be impacted by the various sensitivities, the accuracy of the sensor may be reduced, and thus the measurement and / or control of EGR may be impacted.In document US 6 742 379 B2 a method and an apparatus for calibrating an inlet oxygen concentration sensor are described. The document DE 10 2011 003 095 A1 discloses a method for determining the oxygen concentration in a gas flow and an oxygen sensor, wherein the pressure at the measuring cell is taken into account. The document DE 10 2006 011 722 B3 describes a method for correcting the output signal of a lambda probe, wherein the air humidity is taken into account.In one example, some of the above issues are addressed by a method for an engine, comprising: learning a reference point for an intake oxygen sensor at a reference intake pressure under selected engine idle conditions; and adjusting EGR flow to the engine based on an intake oxygen concentration estimated by the sensor relative to the learned reference point and further based on a change in intake pressure from the reference intake pressure. In this way, a zero point measurement for an intake oxygen sensor may be learned more reliably, thereby increasing the accuracy of EGR control.For example, at first engine idle, an intake oxygen sensor idle adjustment may be performed following each engine start. This may allow learning of aging effects of the sensor. Moreover, when a new sensor has been installed in the vehicle, the idle adjustment may be used to compensate for part-to-part variations. During the idle adjustment, an output of the intake manifold oxygen sensor may be monitored for a duration of the engine idle condition. A relationship between the output of the sensor at a reference inlet pressure may be learned and corrected for factors such as humidity. Upon completion of an adjustment, the output of the intake oxygen sensor may be used to estimate an EGR concentration and thereby adjust an EGR flow. Specifically, the output may be adjusted with a pressure correction factor based on the actual intake pressure and the reference intake pressure, and the corrected oxygen sensor output may be used to more accurately estimate the change in intake oxygen concentration with EGR dilution. By correcting for pressure changes, the pressure effect on oxygen sensor readings can be compensated. Moreover, by performing the adjustment under idle conditions, the effect of PCV and purge HCs on oxygen sensor output is reduced.In this way, a relationship between an intake oxygen sensor and an intake pressure sensor may be learned regardless of the accuracy of one of the sensors and used to adjust EGR flow. By performing learning under idle conditions, corruption of a sensor output due to intake of PCV and purge HCs may be reduced. By performing learning during the first engine idle since an engine start, the effect of sensor aging on the sensor output may be learned. Moreover, the relationship between the oxygen sensor output and the pressure sensor output may be learned under relatively constant engine speed-load conditions. By also performing the idle adjustment each time a new oxygen sensor or pressure sensor is installed in the vehicle, the idle adjustment can be used to compensate for part-to-part variations. Overall, the accuracy of an EGR estimate may be increased, thereby allowing improved EGR control.It should be understood that the summary above is provided to introduce in simplified form a selection of concepts that are further described 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 listed above or in any other part of this disclosure. FIGS. 1-2 are schematic diagrams of an engine system. FIG. 3 is a map showing the relationship between intake pressure nd and pumping current of an intake oxygen sensor. FIG. 4 shows a flowchart for performing a zero point estimate for an intake oxygen sensor under engine idle conditions. FIG. 5 shows a flowchart for performing a zero point estimate for an intake oxygen sensor under engine non-fueling conditions. FIG. 6 shows a flowchart for identifying degradation of an EGR valve based on the zero point estimated using idle adjustment and the zero point estimated using SAS adjustment. FIG. 7 shows an example idle adjustment. FIG. 8 shows a flowchart for correcting a learned zero point based on ambient humidity. FIG. 9 shows a flowchart for EGR control using the learned intake oxygen zero point.The present description relates to methods and a system for learning a reference point or null point for an intake oxygen sensor, such as the sensor coupled to engine systems of FIGS. 1-2. The baseline may be determined based on a learned relationship between the output of the intake oxygen sensor and an output of an intake pressure sensor under selected conditions (FIG. 3 ). A controller may be configured to perform a control routine, such as the routine of FIGS. 4-5, for learning the zero point for the intake oxygen sensor during an idle adjustment or during an SAS adjustment. The learned reference point may be corrected based on ambient humidity (FIG. 8 ). The controller may also be configured to perform a routine (FIG. 6 ) for identifying EGR valve leakage based on discrepancies between the zero estimated at idle conditions and the zero estimated at SAS conditions. In response to EGR valve leakage, the EGR control may be adjusted (FIG. 9 ) to vary the feedback component of the EGR control from the oxygen sensor. An exemplary detection is shown in FIG. 7. In this way, an intake oxygen sensor reading may be corrected for aging, part-to-part variations, and effects of fuel and reductants.FIG. 1 shows a schematic illustration of an example turbo-boosted engine system 100 including a multi-cylinder engine 10 and dual turbochargers 120 and 230. As a non-limiting example, the engine system 100 may be included as part of a propulsion system for a passenger vehicle. Engine system 100 may receive intake air via intake passage 140. Intake passage 140 may include an air filter 156 and an EGR throttle 230. Engine system 100 may be a split engine system in which intake passage 140 downstream of EGR throttle 230 is branched into first and second parallel intake passages each including a turbocharger compressor. More specifically, at least a portion of the intake air is directed to compressor 122 of turbocharger 120 via a first parallel intake passage 142, and at least another portion of the intake air is directed to compressor 132 of turbocharger 130 via a second parallel intake passage 144 of intake passage 140.The first portion of the total intake air compressed by the compressor 122 may be supplied to the intake manifold 160 via the first parallel branched intake passage 146. In this way, the intake ports 142 and 146 form a first parallel branch of the engine's air intake system. Similarly, a second portion of the total intake air may be compressed by compressor 132, and may be supplied to intake manifold 160 via second parallel branched intake passage 148. In this way, the intake ports 144 and 148 form a first parallel branch of the engine air intake system. As shown in FIG. 1, intake air from intake ports 146 and 148 may be merged again via a common intake port 149 before reaching intake manifold 160, where intake air may be supplied to the engine.A first EGR throttle 230 may be positioned in the engine intake upstream of the first and second parallel intake ports 142 and 144, while a second air intake throttle 158 may be positioned in the engine intake downstream of the first and second parallel intake ports 142 and 144 and downstream of the first and second parallel branched intake ports 146 and 148, for example, in the common intake port 149.In some examples, intake manifold 160 may include an intake manifold pressure sensor 182 for estimating manifold pressure (MAP) and / or an intake manifold temperature sensor 183 for estimating manifold charge temperature (MCT) each in communication with controller 12. The intake passage 149 may include a charge air cooler (CAC) 154 and / or a throttle (such as the second throttle 158). The position of the throttle 158 may be adjusted by the control system via a throttle actuator (not shown) communicatively coupled to the controller 12. A surge prevention valve 152 may be provided to selectively bypass the compressor stages of turbochargers 120 and 130 via bypass passage 150. As an example, anti-surge valve 152 may open to allow flow through bypass passage 150 when intake air pressure downstream of the compressors reaches a threshold.The intake manifold 160 may further include an intake gas oxygen sensor 172. In one example, the oxygen sensor is a UEGO sensor. As noted herein, the intake gas oxygen sensor may be configured to provide an estimate of the oxygen content of the fresh air received in the intake manifold. Moreover, as EGR flows, a change in oxygen concentration at the sensor may be used to derive an EGR amount and used for accurate EGR flow control. In the example shown, oxygen sensor 162 is positioned upstream of throttle 158 and downstream of charge air cooler (CAC) 154. However, in other embodiments, the oxygen sensor may be positioned upstream of the CAC.A pressure sensor 174 may be positioned adjacent to the oxygen sensor for estimating an inlet pressure at which an output of the oxygen sensor is received. Since the output of the oxygen sensor is affected by the intake pressure, a reference oxygen sensor output may be learned at a reference intake pressure. In one example, the reference inlet pressure is a throttle inlet pressure (TIP), and the pressure sensor 174 is a TIP sensor. In other examples, the reference intake pressure is a manifold pressure (MAP) as sensed by the MAP sensor 182.A humidity sensor 173 may be positioned adjacent to the intake oxygen sensor and the intake pressure sensor. Specifically, as shown, humidity sensor 173, intake oxygen sensor 172, and intake pressure sensor 174 may be positioned upstream of intake throttle 158 and downstream of charge air cooler 154, respectively, in the engine intake manifold. The humidity sensor may be configured to provide an estimate of the ambient humidity. As embodied with reference to FIG. 8, a controller may estimate an ambient humidity while learning a reference point for the intake oxygen sensor at a reference intake pressure and correct the learned reference point based on the estimated ambient humidity. This allows oxygen sensor output variations due to ambient humidity variations to be learned and used for accurate estimation of EGR.The engine 10 may include a plurality of cylinders 14. In the example shown, the engine 10 includes six cylinders arranged in a V configuration. Specifically, the six cylinders are arranged on two banks 13 and 15, each bank including three cylinders. In other examples, the engine 10 may include two or more cylinders, such as 3, 4, 5, 8, 10, or more cylinders. These various cylinders may be evenly distributed and arranged in other configurations, such as a V, inline, or boxer configuration, etc. Each cylinder 14 may be configured with a fuel injector 166. In the example shown, fuel injector 166 is an in-cylinder direct injector. However, in other examples, fuel injector 166 may also be configured as an intake port injector.Intake air supplied to each cylinder 14 (also referred to herein as combustion chamber 14) via a common intake passage 149 may be used for fuel combustion, and products of combustion may then be exhausted via bank-specific parallel exhaust passages. In the example shown, a first bank 13 of cylinders of engine 10 may exhaust combustion products via a first parallel exhaust passage 17, and a second bank 15 of cylinders may exhaust combustion products via a second parallel exhaust passage 19. The first and second parallel exhaust passages 17 and 19 may each further include a turbocharger turbine. More specifically, combustion products exhausted via exhaust passage 17 may be directed through exhaust turbine 124 of turbocharger 120, whereby mechanical work may in turn be provided to compressor 122 via shaft 126 to provide compression of intake air. Alternatively, some or all of the exhaust gas(s) flowing through the exhaust passage 17 may bypass the turbine 124 via the turbine bypass passage 123 as controlled by the wastegate 128. Similarly, combustion products exhausted via exhaust passage 19 may be directed through exhaust turbine 134 of turbocharger 130, which in turn may be provided mechanical work to compressor 132 via shaft 136 to provide compression of intake air flowing through the second branch of the engine intake system. Alternatively, some or all of the exhaust gas(s) flowing through the exhaust passage 19 may bypass the turbine 134 via the turbine bypass passage 133 as controlled by the wastegate 138.In some examples, the exhaust turbines 124 and 134 may be configured as a variable geometry turbine, where the controller 12 may adjust the position of the turbine runner blades (or vanes) to vary the level of energy obtained from the exhaust stream and applied to its respective compressor. Alternatively, the exhaust turbines 124 and 134 may be configured as a variable nozzle turbine, where the controller 12 may adjust the position of the turbine nozzle to vary the level of energy obtained from the exhaust stream and applied to its respective compressor. For example, the control system may be configured to independently vary the vane or nozzle position of the exhaust turbines 124 and 134 via respective actuators.Exhaust gases in the first parallel exhaust passage 17 may be routed to atmosphere via the branched parallel exhaust passage 170, while exhaust gases in the second parallel exhaust passage 19 may be routed to atmosphere via the branched parallel exhaust passage 180. Exhaust passages 170 and 180 may include one or more exhaust aftertreatment devices, such as a catalyst and one or more exhaust gas sensors.Further, the engine 10 may include one or more exhaust gas recirculation (EGR) passages or circuits for recirculating at least a portion of the exhaust gas from the exhaust manifold to the intake manifold. These may include high pressure EGR circuits for providing high pressure EGR (HP-EGR) and low pressure EGR circuits for providing low pressure EGR (LP-EGR). In one example, in the absence of boost provided by turbochargers 120, 130, HP-EGR may be provided, while in the presence of turbocharger boost, and / or when exhaust temperature is above a threshold, LP-EGR may be provided. In still other examples, both HP-EGR and LP-EGR may be provided simultaneously.In the example shown, engine 10 may include a low pressure EGR circuit 202 for recirculating at least a portion of the exhaust gas from first branched parallel exhaust passage 170 downstream of turbine 124 to first parallel intake passage 142 upstream of compressor 122. In some embodiments, a second low pressure EGR circuit (not shown) may also be provided for recirculating at least a portion of the exhaust gas from the second branched parallel exhaust passage 180 downstream of the turbine 134 to the second parallel intake passage 144 upstream of the compressor 132. The LP-EGR circuit 202 may include an LP-EGR valve 204 for controlling an EGR flow (i.e., a recirculated amount of exhaust gas) through the circuits, and an EGR cooler 206 for lowering the temperature of an exhaust gas flowing through the EGR circuit prior to recirculation to the engine inlet. The LP EGR valve 204 may be positioned upstream or downstream of the LP EGR cooler 206. Under certain conditions, the EGR cooler 206 may also be used to warm the exhaust gas flowing through the LP EGR circuit 202 before the exhaust gas enters the compressor to prevent water drops from impinging on the compressors.The engine 10 may further include a first high pressure EGR circuit 208 for recirculating at least a portion of the exhaust gas from the first parallel exhaust passage 17 upstream of the turbine 124 to the intake manifold 160 downstream of the engine throttle 158. Similarly, the engine may include a second high pressure EGR circuit (not shown) for recirculating at least a portion of the exhaust gas from the second parallel exhaust passage 19 upstream of the turbine 134 to the intake manifold 160 downstream of the engine throttle 158. EGR flow through the HP-EGR circuits 208 may be controlled via the HP-EGR valve 210. When there are two HP-EGR circuits coupled to each branch of the air intake system, they may each use their own HP-EGR valves 201 and may be merged before introduction to the intake manifold and share the same HP-EGR valve. It should be appreciated that as an alternative to the single and dual HP EGR circuit configurations described above, HP EGR may be introduced into the intake ports 146 and / or 148, rather than into the intake manifold 160.A PCV positive crankcase ventilation (PCV) port 102 may be configured to supply blow-by gases (PCV) to the engine intake manifold along the second parallel intake passage 144. In some embodiments, flow of PCV air through the PCV opening 102 may be controlled by a dedicated PCV opening valve. Similarly, a purge port 104 may be configured to supply purge gases from a fuel system canister to the engine intake manifold along passage 144. In some embodiments, purge air through the purge port 104 may be controlled by a dedicated purge port valve. As elaborated with reference to FIG. 2, the PCV and purge ports in the supercharger air intake manifold only direct into the intake manifold under boost conditions. During non-boost conditions, purge and PCV air is directly provided to the intake manifold. In other words, during boost conditions, the purge and PCV gases are received upstream of an intake oxygen sensor 172, and therefore affect the output of the sensor during boost conditions. In other words, during boost conditions, the purge and PCV gases are received upstream of an intake oxygen sensor 172, and therefore affect the output of the sensor during boost conditions.The humidity sensor 232 and the pressure sensor 234 may be included in only one of the parallel intake ports (shown herein in the first parallel intake air port 142, but not shown in the second parallel intake port 144) downstream of the EGR throttle 230. In particular, the humidity sensor and the pressure sensor may be included in the intake passage that does not receive the PCV or purge air. The humidity sensor 232 may be configured to estimate a relative humidity of the intake air. In one embodiment, the humidity sensor 232 is a UEGO sensor configured to estimate the relative humidity of the intake air based on the output of the sensor at one or more voltages. Since purge air and PCV air may interfere with the humidity sensor results, the purge opening and the PCV opening are positioned in an inlet channel other than the humidity sensor. The pressure sensor 234 may be configured to estimate a pressure of the intake air. In some embodiments, a temperature sensor may also be included in the same parallel intake passage downstream of EGR throttle 230.Thus, the intake oxygen sensor 172 may be used to estimate an intake oxygen concentration and derive an EGR flow amount through the engine based on a change in intake oxygen concentration upon opening of the EGR valve 204. Specifically, a change in the output of the sensor when the EGR valve is opened is compared with a reference point at which the sensor operates without EGR (the zero point). Based on the change (for example, the decrease) in the amount of oxygen from the time of operation without EGR, an EGR current currently supplied to the engine may be calculated. Then, based on a deviation of the estimated EGR flow from the expected (or desired) EGR flow, further EGR control may be performed. As embodied with reference to FIG. 9, a controller may feed forward the opening of the EGR valve based on engine speed-load conditions while feedback adjusting the EGR valve based on an EGR current estimated by the oxygen sensor. However, the EGR estimation and control requires accurate zero point estimation. Since the output of the oxygen sensor is affected by changes in intake pressure, changes in exhaust air-fuel ratio, part-to-part variations, and reductants (such as those of PCV and purge gases), zero point estimation may be complicated. However, without accurate estimation of the zero point, reliable EGR flow control cannot be performed.To solve these problems, an estimation of the zero point of the oxygen sensor is performed under idle conditions, also referred to herein as idle adjustment, and discussed with respect to FIG. 4. By performing the adjustment under idle conditions where intake pressure variations are minimal and when PCV or purge air is not introduced to the low pressure air intake system upstream of the compressor, variations in sensor measurement due to these noise factors are reduced. Thus, purge and PCV air may flow into the engine via the intake manifold at idle. However, they do not affect intake oxygen sensor output because they are admitted directly into the intake manifold downstream of the sensor. Also, by periodically performing the idle adjustment, such as at each first idle after an engine start, the effect of sensor aging and part-to-part variability on the sensor output is corrected. Overall, a more accurate zero point can be learned.An estimate of the intake oxygen sensor zero point is also made under engine non-fueling conditions, such as during deceleration fuel shut-off (SAS), also referred to herein as an SAS adjustment, and discussed with respect to FIG. 5. By performing the adjustment under SAS conditions, sensor measurement variations due to EGR valve leakage are also reduced in addition to reduced noise factors such as those achieved during idle adjustment.Referring again to FIG. 1, the position of the intake and exhaust valves of each cylinder 14 may be regulated via hydraulically actuated lifters coupled to valve lifter rods or via direct acting mechanical cup systems in which cam lobes are used. In this example, at least the intake valves of each cylinder 14 may be controlled by cam actuation using a cam actuation system. Specifically, the valve cam actuation system 25 may include one or more cams and may use variable cam timing or variable lift for intake and / or exhaust valves. In alternative embodiments, the intake valves may be controlled by electric valve actuation. Likewise, the exhaust valves may be controlled by cam actuation systems or electric valve actuation.The engine system 100 may be at least partially controlled by a control system 15 including the controller 12 and by input from a vehicle operator via an input device (not shown). The control system 15 is shown to receive information from a plurality of sensors 16 (various examples of which are described herein) and send control signals to a plurality of actuators 81. As an example, the sensors 16 may include the humidity sensor 232, the intake air pressure sensor 234, the MAP sensor 182, the MCT sensor 183, the TIP sensor 174, and the intake air oxygen sensor 172. In some examples, the common intake passage 149 may further include a throttle inlet temperature sensor for estimating a throttle air temperature (TCT). In other examples, one or more EGR passages may include pressure, temperature, and air-fuel ratio sensors for determining EGR flow characteristics. As another example, actuators 81 may include fuel injector 166, HP-EGR valve 210, LP-EGR valve 204, throttle plates 158 and 230, and wastegates 128, 138. Other actuators, such as various additional valves and throttles, may be coupled to various locations in engine system 100. The controller 12 may receive input data from the various sensors, process the input data, and trigger the actuators in response to the processed input data based on instructions or code programmed therein according to one or more routines. Example control routines are described herein with reference to FIGS. 4-6 and 8.Referring now to FIG. 2, another embodiment 200 for the engine of FIG. 1 is shown. Thus, components previously presented in FIG. 1 are provided with like reference numerals and will not be presented again here for brevity.Embodiment 200 shows a fuel tank 218 configured to supply fuel to engine fuel injectors. A fuel pump (not shown) submerged in fuel tank 218 may be configured to pressurize fuel supplied to injectors of engine 10, such as injector 166. Fuel may be pumped from an external source into the fuel tank through a fuel filler cap (not shown). The fuel tank 218 may hold multiple fuel mixtures including fuel having different alcohol concentrations, such as different gasoline-ethanol mixtures including E10, E85, gasoline, etc., and combinations thereof. A fuel level sensor 219 positioned in the fuel tank 218 may provide an indication of the fuel level to the controller 12. As shown, fuel level sensor 219 may include a float connected to a variable resistor. Alternatively, other types of fuel level sensors may be used. One or more other sensors may be coupled to fuel tank 218, such as fuel tank pressure transducer 220 for estimating fuel tank pressure.Vapors generated in fuel tank 218 may be directed to fuel vapor canister 22 via conduit 31 before being purged to engine intake 23. These can include, for example, the fuel tank vapors generated during the day and the fuel tank vapors generated during refueling. The canister may be filled with a suitable adsorbent, such as activated carbon, for temporarily trapping fuel vapors (including vaporized hydrocarbons) generated in the fuel tank. During later engine operation, when purge conditions are met, such as when the canister is saturated, fuel vapors may then be purged from the canister into the engine intake by opening canister purge valve 112 and canister vent valve 114.The canister 22 includes a vent 27 for venting gases from the canister 22 to atmosphere when storing or trapping fuel vapors from the fuel tank 218. Vent 27 may also allow fresh air to be drawn into fuel vapor canister 22 when purging stored fuel vapors to engine intake 23 via purge lines 90 or 92 (depending on boost level) and purge valve 112. Although this example shows an unheated fresh air conveying vent 27, various modifications may be used. The vent 27 may include a canister vent valve 114 for adjusting a flow of air and vapors between the canister 22 and the atmosphere. The vent valve may be opened during fuel vapor storage operations (e.g., when refueling the fuel tank and while the engine is not running) so that air without the fuel vapor may be forced out to the atmosphere after passing through the canister. Similarly, during purges (e.g., during canister regeneration and while the engine is running), the vent valve may be opened to allow a flow of fresh air to remove the fuel vapors stored in the canister.Fuel vapors released from canister 22, for example during a purging operation, may be directed into engine intake manifold 160 via purge line 28. The flow of vapors along purge line 28 may be regulated by canister purge valve 112 coupled between the fuel vapor canister and the engine intake. The amount and rate of vapors released by the canister purge valve may be determined by the duty cycle of an associated canister purge valve solenoid (not shown). Thus, the canister purge valve solenoid duty cycle may be determined by the vehicle powertrain control module (PCM), such as controller 12, which responds to engine operating conditions including, for example, engine speed / load conditions, air-fuel ratio, canister load, etc.An optional canister check valve (not shown) may be included in the purge line 28 to prevent intake manifold pressure from directing gases in the opposite direction from the purge flow. Thus, the check valve may be required when the canister purge valve controller is not accurately clocked, or the canister purge valve itself may be forced open by a high intake manifold pressure. A manifold absolute pressure (MAP) estimate may be obtained from the MAP sensor 182 coupled to the intake manifold 160 and in communication with the controller 12. Alternatively, the MAP may be inferred from other engine operating conditions, such as mass air flow (MAF), as measured by a MAF sensor, not shown, coupled to the intake manifold.Scavenge hydrocarbons may be directed to intake manifold 160 via either boost path 92 or vacuum path 90 based on engine operating conditions. Specifically, during conditions when the turbocharger 120 is operated to provide a boosted air charge to the intake manifold, the increased pressure in the intake manifold causes the one-way valve 94 to close in the vacuum path 90 while the one-way valve 96 opens in the boost path 92. As a result, purge air is directed into the air intake passage 140 downstream of the air filter 156 and upstream of the charge air cooler 154 via the boost path 92. Herein, the purge air is introduced upstream of the intake oxygen sensor 172. In some embodiments, a venturi 98 may be positioned in the boost path such that purge air is directed to the inlet as it passes through the venturi and passage 99. This allows the flow of compressor bypass air to be advantageously utilized for improved purge flow.During conditions where the engine 10 is operated without boost, the increased vacuum in the intake manifold causes the one-way valve 94 to open in the vacuum path while the one-way valve 96 closes in the boost path. As a result, purge air is directed into intake manifold 160 via vacuum path 90 downstream of throttle 158. Herein, the purge air downstream of the intake oxygen sensor 172 is introduced directly into the intake manifold 160 and therefore does not affect the output of the oxygen sensor 172. In comparison, during conditions where engine 10 is operated boosted, purge air is introduced upstream of intake oxygen sensor 172, and therefore affects the output of oxygen sensor 172.PCV hydrocarbons may be directed to intake manifold 160 via either boost side PCV hose 252 or vacuum side PCV hose 254 based on engine operating conditions. Specifically, blow-by gases flow from the engine cylinders 14 past the piston rings and enter the crankcase 255. During conditions when the turbocharger 120 is operated to provide a boosted air charge to the intake manifold, the increased pressure in the intake manifold causes the one-way valve 256 in the vacuum side PCV hose 254 to close. As a result, PCV air is directed into the air intake passage 140 downstream of the air filter 156 and upstream of the charge air cooler 154 via boost side PCV hose 252. The PCV flow may be directed to the intake passage when passing through an boost side oil separator 260. The boost side oil separator may be integrated with the camshaft cover or may be an external component. Thus, under boost conditions, the PCV gases are introduced upstream of an intake oxygen sensor 172, and therefore affect the output of the oxygen sensor 172.In comparison, during conditions when the engine 10 is operating without boost, the increased vacuum in the intake manifold causes the one-way valve 256 in the vacuum side PCV hose 254 to open. As a result, PCV air is directed to intake manifold 160 via PCV hose 254 on the vacuum side downstream of throttle 158. Herein, the PCV air is introduced downstream of the intake oxygen sensor 172, and therefore does not affect the output of the oxygen sensor 172.Thus, due to the particular engine configuration, during engine idle conditions where boosted air charge is not provided, a baseline (also referred to herein as a null point) of the intake air sensor may be learned without interfering with PCV and purge air hydrocarbons.Thus, the intake air oxygen sensor may be used to determine the amount of EGR in the intake air charge as a function of the amount of oxygen content change due to the addition of EGR as a diluent. Thus, as more EGR is introduced, a sensor output corresponding to lower oxygen concentrations may be output. However, to accurately determine this change in oxygen amount, it is important to know the oxygen reading of the sensor when there is no EGR. This reference point, also known as the zero point, must be calibrated and learned. The zero point measurement has a wide range of values that may vary based on the intake pressure, sensor age, and part-to-part variation, making accurate EGR measurement difficult.FIG. 3 shows this variation of the measurement value of the inlet sensor. Specifically, map 300 shows the inlet pressure along the X axis and a pump current output through the sensor upon application of a reference voltage along the Y axis. Plots 301 a- ddisplay a first set of intake oxygen sensor outputs under a first condition without EGR. Plots 302 a- d, 303 a- d, and 304 a- ddisplay sensor outputs at gradually decreasing EGR levels, where 304 a- dis a nominal EGR percentage.As seen by comparing the output at any given inlet pressure (compare 301a with 301b, c and d, etc. for each set), there is a large amount of part-to-part variation in the output of the reference oxygen measurement by the sensor. Thus, the part-to-part variation accounts for the greatest amount of variation in the output of a given sensor. Moreover, the aging of the sensor increases the variation. Due to the variation, learning of the zero point becomes difficult, thereby disturbing the results of an EGR estimation.As elaborated with reference to FIG. 4, the variation may be reduced by performing an idle adjustment for the sensor at each engine start. Specifically, at first engine idle, a zero point of the sensor may be learned and updated since each engine start. This allows learning and compensating for part-to-part variation and sensor aging. By then using the most recently learned zero point as a reference for EGR estimation, the EGR amounts can be determined more accurately and reliably.Referring now to FIG. 4, an example routine 400 for learning an intake oxygen sensor zero point under selected engine idle conditions is shown. The method allows accurate learning of a reference point of the sensor without disturbing influence of PCV or scavenge hydrocarbons. Besides learning the relationship between the intake pressure and the oxygen sensor output, oxygen concentrations and EGR flow can be accurately measured even if there is any imprecision in one of the sensors.At 402, the routine includes estimating and / or measuring engine operating conditions. These may include, for example, engine speed, torque demand, barometric pressure, engine temperature, etc. Next, it may be determined if selected engine idle conditions are present. As set forth below at 404 and 406, the selected engine idle conditions may include a first engine idle since installation of a new intake oxygen sensor or pressure sensor, or a first engine idle since an engine start.Specifically, at 404, it may be determined whether a new intake oxygen (IAO2) sensor or a new intake pressure (P) sensor has been installed in the vehicle. For example, it may be determined whether a new sensor has been installed since the last engine shut-down and the current engine start. After installation of a new sensor, in one example, an indication that calibration of the new sensor is required may be received at a controller.If a new oxygen sensor or pressure sensor has been installed, then at 405, the routine includes resetting the previously learned adaptive values of the intake air oxygen sensor. That is, the previously learned zero point and pressure correction factor stored in a look-up table of the memory of the controller (for example, in the KAM) may be reset. The table can then be filled again with data from the learning of the current zero point and subsequent iterations of the routine.If a new oxygen or pressure sensor has not been installed or the table has been reset if a new sensor has been installed, the routine proceeds to 406 to confirm the first engine idle condition since the current engine start. If an engine idle condition is not confirmed, at 407, the look-up table in the controller's memory may not be updated further and the current zero measurements may be used. Thus, by re-learning the datum each time a new sensor is installed, differences in oxygen sensor readings due to part-to-part variations can be better accounted for. By updating and re-learning the reference point at each engine start, differences in the oxygen sensor readings due to part-to-part variations may be better accounted for.Upon confirming a first engine idle condition since the current engine start, at 408, the routine includes learning a reference point for the intake oxygen sensor at a reference intake pressure during the selected engine idle condition. Specifically, the controller may learn the oxygen sensor output at the first engine idle condition and may also record the reference intake pressure at which the oxygen sensor output has been learned. Then, the controller may update the lookup table stored in the controller's KAM with the learned oxygen sensor output. In one example, the reference intake pressure is a throttle intake pressure estimated by a TIP sensor coupled to the intake manifold at a similar location as the oxygen sensor (e.g., downstream of the charge air cooler and upstream of the intake throttle). In another example, the reference intake pressure is manifold pressure estimated by a MAP sensor coupled to the intake manifold at a similar location as the oxygen sensor.Thus, learning the reference point includes learning a relationship between a first output of the intake oxygen sensor at a first intake pressure during the first engine idle since the start, and then, using the learned relationship, the idle reference oxygen (iao2_ref) at the reference pressure (iao2_ref_press) is calculated. By determining a correction factor, (iao2_press_corr) is defined as:iao2_press_corr=a0+ a1*(iao2_ref_press - iao2_press)+ a2*(iao2_ref_press - iao2_press) 2.Then, the idle reference oxygen is calculated as: iao2_ref=iao2_o2*iao2_press_corr.By performing this learning under idling conditions, various advantages can be obtained. First, any error caused by noise factors of purge or PCV hydrocarbons is reduced. Secondly, since pressure changes at the location of the intake oxygen sensor are minimal under idle conditions, changes in oxygen sensor output due to the pressure effect (described at FIG. 3 ) also become minimal. Overall, more accurate learning of the zero point can be achieved.At 410, the intake oxygen sensor output is corrected for humidity. As elaborated with reference to FIG. 8, the output of the intake oxygen sensor estimated at the reference pressure is corrected with a correction factor based on ambient humidity. This may include correcting for no humidity (i.e., zero % humidity or dry conditions), where the output of the oxygen sensor is corrected by eliminating all moisture contribution. Alternatively, this may comprise correcting for a known standard or reference humidity level. For example, the oxygen sensor output may be corrected to a reference humidity of 1.2%.At 412, it may be determined if the idle adjustment is completed. Thus, the intake oxygen sensor readings at the given reference intake pressure may be monitored for a duration of the first engine idle since engine start, and the look-up table may be further filled with readings from the intake oxygen sensor over the duration. In one example, upon initiation of the idle adjustment at 408, a timer may be started and at 412, a determination may be made as to whether a threshold duration has expired on the timer. In one example, the idle adjustment may be completed when 15 seconds have elapsed.Upon confirming that the idle adjustment is completed, the routine includes computing a pressure correction factor at 414. The pressure correction factor is a factor that compensates for the effect of intake pressure on the output of the intake oxygen sensor. The pressure correction factor is determined from the ratio of the measured oxygen to the reference oxygen measurement (iao2_ref), where the reference oxygen measurement is the reference oxygen measurement of the intake oxygen sensor at the reference pressure. The reference pressure may be nominally 100 kPa. The pressure correction adjustment may be performed by calculating a pressure correction factor based on the output of the intake oxygen sensor (iao2_o2) with respect to the zero point (iao2_ref) of the sensor (i.e., iao2_o2 / iao2_ref). In addition, a delta pressure may be determined based on the reference pressure, the delta pressure calculated as TIP- iao2_ref_press. Herein, TIP may be equal to boost pressure. The delta pressure is calculated as the difference between the measured boost pressure TIP and the reference pressure. The delta pressure from the reference pressure provides information about the change in the oxygen reading of iao2_ref vs. the change in pressure from the reference pressure. The reference pressure corresponds to the pressure at which iao2_ref was determined.At 418, the routine includes calculating and learning the zero point of the intake oxygen sensor. This may include, for example, performing a Recursive Least Squares (RLS) adjustment for pressure correction. This correction may be referred to as:lao2_press_corr_new = a2*dp_corr 2+ a1*dp_corr + a0, where a0, a1 and a2 are pressure correction coefficients and dp_corr is the delta pressure correction (i.e., delta pressure from the reference pressure).After learning of the zero point, an EGR flow to the engine may be adjusted based on an output of the intake oxygen sensor under EGR conditions as carried out at FIG. 9. Herein, an EGR flow to the engine is adjusted based on an intake oxygen concentration estimated by the intake oxygen sensor with respect to the learned reference point and further based on a change in the intake pressure with respect to the reference intake pressure (the reference point having been learned).At 420, the routine includes diagnosing an EGR valve based on the zero estimated during the idle adjustment relative to a zero estimated under selected engine non-fueling conditions, such as during a deceleration fuel shut-off (SAS) adjustment. An example SAS adaptation is described in FIG. 5. Thus, the zero point learned during the idle adjustment may be a first learned reference point, while the zero point learned during the SAS adjustment may be a second learned reference point (both learned at a given reference inlet pressure). As embodied at FIG. 6, the controller may indicate degradation of the EGR valve based on a difference between the learned first reference point and the learned second reference point being greater than a threshold level.Although the idle adjustment performed under idle conditions eliminates the effect of purge and PCV hydrocarbons on intake oxygen sensor output as well as the effect of pressure oscillations, the idle adjustment is susceptible to EGR leakage. Thus, if the EGR valve leaks, assuming there is no EGR recirculation, EGR may flow via the intake oxygen sensor even under idle conditions. As a result, the output from the oxygen sensor may be less than the actual value. In comparison, an adjustment performed under SAS is insensitive to the effect of a leaking EGR valve. This is because even if the valve were to leak, the leaking "EGR" would be air because no fuel is injected under these conditions. As a result, the exhaust gas leakage does not affect the output of the oxygen sensor. Thus, by comparing the zero point learned during the idle adjustment with the zero point learned during the SAS adjustment, EGR valve leakage can be identified.An example detection is shown with reference to FIG. 7. Map 700 shows an idle adjust timer at plot 702, and a change in oxygen concentration sensed by the intake oxygen sensor at plot 704.Prior to t 1, there may be no idle adjustment conditions. At t 1, a first engine idle may be confirmed since an engine start, and an idle adjust timer may be started. Plot 704 (solid line) shows a zero point of the intake oxygen sensor relative to an expected value 708. Plot 706 (dashed line) shows the intake sensor output. Before the idle adjustment, the deviation of the estimated zero point from the expected zero point is thus greater. During the adjustment, the zero point is then corrected based on the sensor measurement (plot 706), and the learned zero point gradually transitions to the expected value. At t 2, the idle adjustment is ended, and the learned zero point is used for accurate EGR control.In one example, a method for an engine includes learning a relationship between a first intake oxygen sensor output estimated at a first intake pressure during a first engine idle since engine start, and adjusting EGR flow to the engine at a second intake pressure based on a second intake oxygen sensor output estimated at the second intake pressure and the learned relationship. The adjusting includes calculating a pressure correction factor based on a difference between the first intake pressure and the second intake pressure, calculating a humidity correction factor based on a difference between ambient humidity at the second intake pressure and a reference humidity, modifying the second intake oxygen sensor output based on both the calculated pressure correction factor, the humidity correction factor, and the learned relationship, and adjusting a position of an EGR valve based on the modified second intake oxygen sensor output. The EGR valve may be coupled in a low pressure EGR passage, where learning is performed at a first engine idle following each engine restart. Herein, the first and second intake oxygen sensor outputs are generated by an intake oxygen sensor coupled upstream of an intake throttle and downstream of a charge air cooler, and the first and second intake pressures are estimated by an intake pressure sensor coupled upstream of the intake throttle and downstream of the charge air cooler. Learning is performed at a first engine idle following installation of an intake oxygen sensor and / or intake pressure sensor in the engine to correct part-to-part variations as well as sensor aging. Additionally, degradation of the EGR valve may be indicated based on the first intake oxygen sensor output estimated at the first intake pressure during the first engine idle since engine start relative to a second intake oxygen sensor output estimated at the first intake pressure during a fuel cut condition.Referring now to FIG. 5, an example routine 500 for learning an intake oxygen sensor zero point under selected engine idle conditions is shown. The method allows accurate learning of a reference point of the sensor without interfering influence from EGR valve leakage.At 502, such as 402, the routine includes estimating and / or measuring engine operating conditions. These may include, for example, engine speed, torque demand, barometric pressure, engine temperature, etc. Next, it may be determined if selected engine non-fueling conditions are present. As set forth below, the selected engine non-fueling conditions may include a deceleration fuel shut-off condition. The routine may be repeated at the first SAS event after each engine start and / or the first SAS event after installation of a new oxygen or pressure sensor.At 504, it may be determined whether a new intake oxygen (IAO2) sensor or a new intake pressure sensor has been installed in the vehicle. For example, it may be determined whether a new sensor has been installed since the last engine shut-down and the current engine start. After installation of a new sensor, in one example, an indication that calibration of the new sensor is required may be received at a controller.If a new oxygen sensor or pressure sensor has been installed, then at 505, the routine includes resetting the previously learned adaptive values of the intake air oxygen sensor. That is, the previously learned zero point and pressure correction factors stored in a look-up table of the memory of the controller (for example, in the KAM) may be reset. Then, the table may be filled again with data from learning the current zero point and subsequent iterations of the SAS adjustment routine.If a new oxygen or pressure sensor has not been installed or the table has been reset if a new sensor has been installed, the routine proceeds to 506 to confirm whether engine non-fueling conditions are present. In particular, a deceleration fuel shut-off (SAS) condition may be confirmed. If no SAS condition is asserted, at 507, the look-up table in the controller's memory cannot be further updated and the current zero measurements can be used. Thus, by re-learning the datum each time a new sensor is installed, differences in oxygen sensor readings due to part-to-part variations can be better accounted for. By updating and re-learning the reference point at each engine start, differences in the oxygen sensor readings due to part-to-part variations may be better accounted for.Upon confirming the SAS condition, at 508, the routine includes learning a reference point for the intake oxygen sensor at a reference intake pressure during the non-fueling condition. Specifically, the controller may learn the oxygen sensor output at the first engine idle condition and may also record the reference intake pressure at which the oxygen sensor output has been learned. Then, the controller may update the lookup table stored in the controller's KAM with the learned oxygen sensor output. In one example, the reference intake pressure is a throttle intake pressure estimated by a TIP sensor coupled to the intake manifold at a similar location as the oxygen sensor (e.g., downstream of the charge air cooler and upstream of the intake throttle). In another example, the reference intake pressure is manifold pressure estimated by a MAP sensor coupled to the intake manifold at a similar location as the oxygen sensor.Thus, learning the baseline includes learning a relationship between a first output of the intake oxygen sensor at a first intake pressure during the first SAS event since engine start and then using the learned relationship to determine the zero point. The learned relationship is used to determine the zero point by calculating the oxygen measurement value at the reference pressure by replacing the delta pressure with the reference pressure. By performing this learning under SAS conditions, various advantages can be achieved. First, any error caused by noise factors of purge or PCV hydrocarbons is reduced. Second, errors due to EGR valve leakage are reduced. This is because under non-fueling conditions, any checked "EGR" corresponds to intake air. Overall, more accurate learning of the zero point can be achieved.At 510, the intake oxygen sensor output is corrected for humidity. As elaborated with reference to FIG. 8, the output of the intake oxygen sensor estimated at the reference pressure is corrected with a correction factor based on ambient humidity. This may thus comprise correcting for no humidity (i.e. zero % humidity or dry conditions) where the output of the oxygen sensor is corrected by eliminating all moisture contribution. Alternatively, this may comprise correcting for a known standard or reference humidity level. For example, the oxygen sensor output may be corrected to a reference humidity of 1.2%.At 512, it may be determined if the SAS adjustment is completed. Thus, the intake oxygen sensor readings at the given reference intake pressure may be monitored for a duration of the SAS, and the look-up table may be further filled with readings from the intake oxygen sensor over the duration. Upon initiation of the SAS at 508, in one example, a timer may be started and at 512, a determination may be made as to whether a threshold duration has expired on the timer. In one example, the SAS adjustment may be completed when 4 seconds have elapsed.Upon confirming that the SAS adjustment is completed, the routine includes computing a pressure correction factor at 514. The pressure correction factor is a factor that compensates for the effect of intake pressure on the output of the intake oxygen sensor. The pressure correction adjustment may be performed by calculating a pressure correction factor based on the output of the intake oxygen sensor (iao2_o2) with respect to the zero point (iao2_ref) of the sensor (i.e., iao2_o2 / iao2_ref). In addition, a delta pressure may be determined based on the reference pressure, the delta pressure calculated as TIP- iao2_ref_press. Herein, TIP may be equal to boost pressure. At idle condition, the reference intake oxygen and pressure are determined. The pressure correction factor at a given pressure condition is calculated as the ratio of the intake oxygen sensor measurement value to the reference oxygen concentration (i.e., iao2_o2 / iao2_ref). This correction factor is learned as a relationship between the delta pressure and the reference pressure. This normalizes the pressure input into the relationship to the reference pressure.At 518, the routine includes calculating and learning the zero point of the intake oxygen sensor. This may include, for example, performing a Recursive Least Squares (RLS) adjustment for pressure correction. This correction may be referred to as:lao2_press_corr_new = a2*dp_corr 2+ a1*dp_corr + a0, where a0, a1 and a2 are pressure correction coefficients and dp_corr is the delta pressure correction.After learning of the zero point, an EGR flow to the engine may be adjusted based on an output of the intake oxygen sensor under EGR conditions as carried out at FIG. 9. Herein, an EGR flow to the engine is adjusted based on an intake oxygen concentration estimated by the intake oxygen sensor with respect to the learned reference point and further based on a change in the intake pressure with respect to the reference intake pressure.At 520, the routine includes diagnosing an EGR valve based on the zero estimated during SAS adjustment relative to a zero estimated during idle adjustment. An example SAS adaptation is described in FIG. 4. Thus, the zero point learned during the idle adjustment may be a first learned reference point, while the zero point learned during the SAS adjustment may be a second learned reference point (both learned at a given reference inlet pressure). As embodied at FIG. 6, the controller may indicate degradation of the EGR valve based on a difference between the learned first reference point and the learned second reference point being greater than a threshold level.Referring now to FIG. 8, an example routine 800 for correcting a nominal output of an intake oxygen sensor upon learning of the zero point based on an ambient humidity estimate is shown. The routine allows oxygen displaced by the moisture to be accounted for.At 802, the routine includes confirming that learning of the null point is enabled. In particular, it may be confirmed that either the idle adjustment or the SAS adjustment of the intake oxygen sensor is performed, as discussed above with reference to FIGS. 4-5.Upon confirmation, the routine includes learning a reference point for the intake oxygen sensor at a reference intake pressure at 804. This includes learning a nominal amount of oxygen based on an output of the intake oxygen sensor at the reference intake pressure under selected engine idle conditions or selected engine non-fueling conditions. Thus, the reference inlet pressure is a throttle inlet pressure or an intake manifold pressure. The selected engine idle conditions include a first engine idle since an engine start, a first engine idle since installation of the intake oxygen sensor, or installation of an intake pressure sensor configured to estimate the reference intake pressure. The selected non-fueling conditions include a fuel cut condition.At 806, an intake oxygen concentration is estimated based on the sensor output. At 808, ambient humidity is estimated via, for example, an intake manifold humidity sensor (such as sensor 173 of FIG. 1 ). At 810, the routine includes calculating an amount of oxygen displaced by the estimated ambient humidity. Thus, the change in oxygen concentration due to humidity may be defined according to the following equation: where O 2 MeasuredConis the measured oxygen concentration with volume % water (proportion) amount of water in air (i.e., humidity).At 812, it may be determined whether to correct the nominal oxygen concentration based on the ambient humidity to reflect dry conditions or standard humidity conditions. In one example, under a first condition (at 814), the reference point may be calibrated to dry conditions (zero humidity) with the effect of all humidity eliminated from the oxygen sensor output. In another example, the reference point may be calibrated to standard humidity conditions under a second condition (at 816), wherein the effect of humidity on the oxygen sensor output is corrected for predefined humidity conditions. An example of a standard moisture condition may be a moisture of 8 g / kg or 1.28%.If a dry condition calibration is selected, then at 814, the routine includes correcting the learned reference point by adding the calculated oxygen amount to the learned nominal oxygen amount. This corrects the reference point for dry air conditions (i.e., zero humidity) and eliminates the effect of all humidity on the oxygen sensor output. Then, the routine proceeds to 820 to update the zero point value in the adaptive value table. Specifically, the corrected zero point with respect to the intake pressure reference is learned and stored in the memory of the controller.If a calibration to the default humidity condition is selected, then at 816 the routine includes correcting the learned reference point by adding the calculated oxygen amount to the learned nominal oxygen amount, as at 814. Then, at 818, after correcting the reference point to dry air, the routine further includes calibrating the reference point to a standard humidity level. In one example, the standard humidity level is 1.2% humidity. Then, the routine proceeds to 520 to update the zero point value in the adaptive value table.Thus, the moisture corrected null point is then used to estimate EGR and adjust EGR flow. For example, the controller may subsequently (i.e., after learning and under engine non-idling conditions) adjust EGR flow to the engine based on an intake oxygen concentration estimated by the sensor with respect to the corrected reference point and further based on a change in intake pressure from the reference intake pressure.In one example, the intake oxygen sensor measurement may be 19.5% oxygen and the ambient humidity estimated by the humidity sensor may be 30 g / kg air. The moisture measurement may be converted to mole percent water according to the calculation 100*(30 / 1000)*29 / 18=4.83%, where 29 is the molecular weight of air and 18 is the molecular weight of water. The 4.83% water displaces an amount of oxygen corresponding to 4.83*21 / 100=1.01% oxygen, 21 being the measurement of the oxygen of the dry air. The corrected dry air measurement of the intake oxygen sensor is then calculated as 19.5% (intake air sensor measurement)+1.01% (humidity correction factor)=20.5%.Alternatively, the dry air oxygen measurement learned above is further adjusted to a standard humidity level oxygen measurement. The information from the moisture sensor is used to calculate the measurement value of the oxygen of the dry air, which is then adjusted with the amount of oxygen that would be displaceable by a quantity of adjustable moisture. Referring to the above example, if the calibrateable amount of moisture comprised 10 g / kg of air, the oxygen corresponding to that amount of moisture would be only 0.34%. The nominal inlet oxygen sensor reading would then be set to 20.5% (dry air reading) - 0.34% (displaced oxygen for calibrated humidity level) = 20.16%.As another example, an engine system may include an engine including an intake manifold, a turbocharger including an exhaust turbine and an intake compressor, a charge air cooler coupled downstream of the compressor, and an intake oxygen sensor coupled to the intake manifold downstream of the charge air cooler and upstream of an intake throttle. Alternatively, the intake oxygen sensor may be positioned upstream of the CAC when the total LP-EGR concentration delivered to the engine is well mixed. Further, the engine system may include a pressure sensor coupled to the intake manifold downstream of the charge air cooler and upstream of the intake throttle, and a humidity sensor coupled to the intake manifold downstream of the charge air cooler and upstream of the intake throttle. The engine may include an EGR system including an EGR passage and an EGR valve for recirculating residual exhaust gases from downstream of the turbine to upstream of the compressor. An engine controller may be configured with computer readable instructions to: during a first engine idle since an engine start, learn an oxygen sensor output and a humidity sensor output at a reference intake pressure, and adjust the oxygen sensor output based on the humidity sensor output. During subsequent engine non-idling conditions, the controller may then be configured to adjust an opening of the EGR valve based on an intake oxygen concentration estimated by the intake oxygen sensor relative to the reference oxygen sensor output and further based on an intake pressure relative to the reference intake pressure. Herein, adjusting the oxygen sensor output based on the humidity sensor output under a first condition at idle includes estimating a first amount of oxygen displaced by total humidity based on the humidity sensor output and adjusting (i.e., increasing) the reference oxygen sensor output for either dry or standard humidity conditions. In a second condition, such as non-idling conditions, the oxygen sensor may, in comparison, reliably predict the oxygen concentration and adjust the EGR valve previously corrected for part-to-part variations, change over time, and variable humidity levels.In this way, a controller may correct a first nominal output of an intake oxygen sensor learned under selected engine idle conditions at a reference intake pressure based on a measured ambient humidity. The engine idle conditions selected include a first engine idle since engine start, a first engine idle following installation of the intake oxygen sensor, and a first engine idle following installation of an intake pressure sensor. Then, the controller may adjust EGR flow to the engine based on a second output of the sensor estimated at the second intake pressure relative to the corrected first output. The EGR flow may be further adjusted based on the second intake pressure relative to the reference intake pressure.The correction performed by the controller may comprise calculating an amount of oxygen displaced by the estimated ambient humidity and increasing the first output to include the amount of displaced oxygen, the increased first output indicative of an oxygen content for dry air. In this way, the effect of all moisture on the oxygen sensor output is eliminated. Alternatively, the correction may comprise adjusting the increased first output based on an amount of displaceable oxygen by a calibrated humidity level, the adjusted output indicative of an air oxygen content for calibrated humidity. In this way, the oxygen sensor output is calibrated to a standard humidity level.The controller may adjust EGR flow by estimating a delivered EGR flow based on a difference between the second output and the corrected first output and adjusting a position of an EGR valve based on a difference between the delivered EGR flow and a desired EGR flow, the desired EGR flow based on engine speed-load conditions.Referring now to FIG. 6, an example routine 600 is shown for diagnosing an EGR valve coupled to a low pressure EGR system based on intake oxygen sensor references learned during idle adjustment and SAS adjustment. The method allows identifying and compensating for EGR valve leakage.At 602, the routine includes retrieving a first reference point learned during a first idle adjustment (ref_idle), such as the idle adjustment of FIG. 4. At 604, the routine includes retrieving a second reference point learned during a SAS adjustment (ref_DFSO), such as the SAS adjustment of FIG. 5. At 606, the two reference points may be compared and it may be determined whether there are any discrepancies between them. In particular, it may be determined whether the first reference point is in a threshold range of the second reference point or whether they differ by more than a threshold amount. Then, the controller may indicate EGR valve leakage based on the first reference point of the intake oxygen sensor learned under engine idle conditions with respect to the second reference point of the oxygen sensor learned under engine non-fueling conditions. Specifically, at 610, EGR valve leakage is indicated based on a difference between the first reference point and the second reference point being greater than a threshold. The controller may indicate EGR valve degradation by setting a diagnostic code. On the other hand, at 608, no EGR valve leakage is indicated if the difference is less than the threshold.As discussed at FIG. 9, the EGR control responsive to an output of the intake oxygen sensor may be adjusted based on the indication of EGR valve leakage. Specifically, responsive to the indication that there is no EGR valve leakage, the EGR valve may be adjusted to be feedforward based on engine speed-load conditions and adjusted to be feedback based on an output of the intake manifold sensor with respect to the first or second reference point. On the other hand, in response to the indication of EGR valve leakage, the controller may continue feedforward adjustment of the EGR valve based on the engine speed-load conditions, but may end feedback adjustment of the EGR valve based on the output of the intake manifold sensor with respect to the first or second reference point.As used herein, the indication of EGR valve degradation includes indicating leakage of an EGR valve coupled to a low pressure EGR passage configured to recirculate residual exhaust gases from an exhaust manifold downstream of a turbine to an intake manifold upstream of a compressor. The intake oxygen sensor may be coupled to the engine intake manifold upstream of an intake throttle and either upstream or downstream of a charge air cooler, the cooler coupled downstream of the compressor. Herein, each of the first and second reference points is learned at a reference intake pressure, the reference intake pressure estimated by an intake pressure sensor coupled to the engine intake manifold upstream of the intake throttle and downstream of the charge air cooler.Referring now to FIG. 9, routine 900 shows an example method for performing EGR control using the output of an intake manifold oxygen sensor relative to a zero point of the sensor learned during an idle adjustment and / or an SAS adjustment. Further, the method adjusts the feedforward feedback components of the EGR control based on an indication of EGR valve degradation.At 902, the output of an intake manifold oxygen sensor is received. An inlet pressure at which the output was received is also recorded since the output is affected by the inlet pressure. At 904, a pressure correction of the output based on the inlet pressure at which the sensor output was sensed is performed with respect to a reference inlet pressure. Further, at 904, a difference between the pressure corrected oxygen output and the zero point of the oxygen sensor is learned. Thus, as an amount of EGR flow increases, exhaust dilution of the intake air increases, thereby reducing the amount of oxygen available in the intake air and thereby reducing the output of the intake sensor. The EGR dilution may be reflected as a drop in oxygen concentration sensed by the intake oxygen sensor.Thus, at 906, a change in oxygen concentration may be determined based on the determined difference between the oxygen sensor output with respect to the zero point. At 908, an amount of EGR dilution of the intake air is determined based on the change in oxygen concentration. At 910, an EGR flow is controlled based on the determined EGR dilution and the desired EGR. As used herein, the EGR flow may be a low pressure EGR flow along an EGR passage from an exhaust manifold downstream of an exhaust turbine to an intake manifold upstream of an intake compressor via an EGR valve. For example, EGR may be provided at a fixed rate or variable rate based on engine operating conditions. For example, at all engine speed-load conditions from a medium load down to a minimum load, low pressure EGR may be supplied at a fixed rate with respect to intake airflow (i.e., at a fixed EGR percentage). On the other hand, at engine speed-load conditions above a medium load, low pressure EGR may be supplied at a variable rate with respect to intake airflow (i.e., at a variable EGR percentage).Controlling EGR flow includes, at 911, feedforward adjusting the EGR valve based on engine operating conditions, such as speed-load conditions. For example, during conditions of higher engine speed load, an opening of the EGR valve may be increased, while during conditions of lower engine speed load, the opening of the EGR valve may be decreased. Additionally, at 912, controlling includes feedback adjusting the EGR valve based on the calculated EGR flow relative to a desired EGR flow. For example, if the actual current estimated by the intake oxygen sensor exceeds the desired or expected current, the EGR valve opening may be decreased. As another example, if the actual current estimated by the intake oxygen sensor is below the desired or expected current, EGR valve opening may be increased.At 914, it may be determined if there is an indication of EGR valve leakage. As embodied in FIG. 6, an EGR valve leakage may be identified based on deviations between an oxygen sensor zero point learned using the idle adjustment and a zero point learned using the SAS adjustment. If no EGR valve leakage is identified, the routine may end. Otherwise, at 816, in response to the indication of EGR valve leakage, the controller may end the feedback adjustment of the EGR valve based on the output of the intake oxygen sensor and may temporarily switch to using only feedforward control of the EGR valve. In other embodiments, EGR may be temporarily disabled in response to the indication of EGR valve leakage, or a diagnostic flag may be set.In other words, in response to the indication that there is no EGR valve leakage, the EGR valve is adjusted feedforward based on engine speed-load conditions and adjusted feedback based on an output of the intake manifold sensor with respect to the first and / or second reference point learned during SAS adjustment. In contrast, in response to the indication of EGR valve leakage, the EGR valve is only feedforward adjusted based on engine speed-load conditions, while feedback adjustment of the EGR valve based on the output of the intake manifold sensor with respect to the first and / or second datum is ended. This allows EGR control to be improved when EGR valve leakage is known.In one example, an engine system includes an engine including an intake manifold, a turbocharger including an exhaust turbine and an intake compressor, a charge air cooler coupled downstream of the compressor, and an intake oxygen sensor coupled to the intake manifold downstream of the charge air cooler and upstream of an intake throttle. A pressure sensor may be coupled to the intake manifold downstream of the charge air cooler and upstream of the intake throttle. Further, the engine may include an EGR system including an EGR passage and an EGR valve for recirculating residual exhaust gases from downstream of the turbine to upstream of the compressor. A controller of the engine system may be configured with computer readable instructions to: during a first engine idle since an engine start, learn a reference point for the oxygen sensor at a reference intake pressure and adjust the EGR valve based on an intake oxygen concentration estimated by the sensor relative to the learned reference point and further based on an intake pressure relative to the reference intake pressure. Additionally or optionally, the controller may learn a reference point for the oxygen sensor at the reference intake pressure and adjust an opening of the EGR valve based on an intake oxygen concentration estimated by the sensor with respect to the learned reference point and further based on an intake pressure with respect to the reference intake pressure under a fuel cut condition. Further, the engine system may include a humidity sensor for estimating an ambient humidity, wherein the controller then resets opening of the EGR valve based on an ambient humidity relative to a reference humidity level. The controller may further determine degradation of the EGR valve based on differences between the reference points learned under the idle condition with respect to the SAS condition.In this way, a relationship between an intake oxygen sensor and an intake pressure sensor may be learned under various humidity conditions, and an EGR flow may be learned based on a change in output of the oxygen sensor regardless of the accuracy of either the oxygen sensor or the pressure sensor. By adjusting the output of an oxygen sensor based on an ambient humidity estimated by an intake humidity sensor, the displacement of the intake oxygen by humidity may be accurately estimated and accounted for, thereby improving the reliability of the zero point measurement of the oxygen sensor. By performing learning under idle conditions, noise factors due to intake of PCV and purge HCs, intake pressure changes, sensor aging, and part-to-part variations are reduced. By performing learning even under engine non-fueling conditions such as SAS, noise factors due to EGR valve leakage are reduced. By increasing the accuracy of the zero point measurement value of the oxygen sensor, EGR can be estimated more reliably, thereby improving EGR control.It should be appreciated that the example control and estimation routines included herein may be used with various engine and / or vehicle system configurations. The specific routines described herein may represent one or more of any number of processing strategies such as event-driven, interrupt-driven, multi-tasking, multi-threading, and the like. Thus, various acts, operations, and / or functions illustrated 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 merely 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. Further, 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 engine control system.It should be understood that the configurations and routines described 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-3, I-4, I-6, V-12, Boxer-4, and other engine types. 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 are specifically directed to 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 inclusion of 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 presenting novel claims in this or a related application. Such claims, whether broader, narrower, equal, or different in scope to the original claims, are also considered to be included within the subject matter of the present disclosure.EXPLANATION OF CHARACTERSFIG. 4 Zero point at idle 402 estimate and / or measure 404 New IAO2 or new P sensor has been installed? 405 Reset learned adaptive table values 406 First idle since engine start? 407 Do not update 408 Oxygen sensor output at idle for a given sensor age. Record reference pressure at which oxygen sensor output was learned. 410 Adjust oxygen sensor output for humidity (FIG. 8 ) 412 Idle Adjustment Completed? 414 Calculate pressure correction factor based on reference pressure 416 Perform pressure correction for oxygen sensor output 418 Calculate zero point of IAO2sensor and learn (idle adjustment) 420 EGR valve diagnostic based on learned idle adjustment zero point of IAO2sensor and SAS learned zero point (FIG. 5 ) of IAO2sensor (FIG. 6 ). FIG. 5 Estimate and / or measure 504 New IAO2 or new P sensor installed? 505 Reset learned adaptive table values 506 SAS conditions? 507 Not update adaptive table values 508 Learn oxygen sensor output at idle. Record reference pressure at which oxygen sensor output was learned. 510 Correct oxygen sensor output for humidity (FIG. 8 ) 512 SAS adjustment completed? 514 Calculate pressure correction factor based on reference pressure 516 Perform pressure correction for oxygen sensor output 518 Calculate zero point of IAO2 sensor and learn (idle adjustment) 520 EGR valve diagnostic based on learned idle adjustment zero point from IAO2 sensor (FIG. 4 ) and SAS learned zero point from IAO2 sensor (FIG. 6 ). FIGS. 6 602 Fetch 604 During idle adjustment learned reference zero (ref_idle) 604 During SAS adjustment learned reference zero (ref_DFSO) fetch 606 Ref_DFSO within Ref_idle range? 608 No LP-EGR valve leakage determines 610 LP-EGR valve leakage. Set diagnostic code. FIG. 8 802 illustrates learning of zero point enabled? 804 oxygen sensor output and reference intake pressure learn 806 intake oxygen concentration based on learned sensor output estimate 808 ambient humidity estimate 810 amount of oxygen displaced by estimated ambient humidity calculate 812 dry condition correction or default humidity desired? 814 zero point by adding the amount of displaced oxygen to estimated intake oxygen concentration correct 816 adding the amount of displaced oxygen to estimated intake oxygen concentration 818 estimated intake oxygen concentration based on oxygen amount displaceable by calibrated amount of moisture further set 820 zero point value table update FIG. 9 902 receiving output of intake oxygen sensor. Intake pressure records 904 difference between received output and oxygen sensor zero when corrected for pressure difference from reference pressure, determine 906 change in oxygen concentration based on determined difference calculate 908 EGR based on change in oxygen concentration calculate 910 EGR flow based on calculated EGR adjust 911 EGR valve feedforward based on engine operating conditions adjust 912 EGR valve feedback based on calculated EGR relative to target EGR adjust 914 EGR valve leakage? 916 stop feedback adjustment of EGR flow and use only feedforward control of EGR flow

Claims

A method for an engine, comprising: learning a reference point for an intake oxygen sensor at a reference intake pressure under selected engine idle conditions; and adjusting an EGR flow to the engine based on an intake oxygen concentration estimated by the sensor relative to the learned reference point and further based on a change in intake pressure relative to the reference intake pressure, wherein adjusting the EGR flow to the engine comprises adjusting low pressure EGR flow along an EGR passage from an exhaust manifold downstream of an exhaust turbine to an intake manifold upstream of an intake compressor via an EGR valve, wherein the learned reference point is a first learned reference point, the method further comprising learning a second reference point for the intake oxygen sensor at the reference intake pressure under selected engine non-fuel delivery conditions.The method of claim 1, wherein the reference inlet pressure is a throttle inlet pressure.The method of claim 1, wherein the reference intake pressure is an intake manifold pressure.The method of claim 1, wherein the selected engine idle conditions comprise a first engine idle since engine start.The method of claim 1, wherein the selected engine idle conditions comprise a first engine idle since installation of the intake oxygen sensor.The method of claim 1, wherein the selected engine idle conditions comprise a first engine idle since installation of an intake pressure sensor configured to estimate the reference intake pressure.The method of claim 1, wherein learning the reference point comprises correcting an output of the intake oxygen sensor estimated at the reference pressure with a correction factor based on ambient humidity.The method of claim 7, wherein learning the reference point further comprises performing a recursive least squares adjustment for pressure correction.The method of claim 1, further comprising indicating degradation of the EGR valve based on a difference between the learned first reference point and the learned second reference point being greater than a threshold level.The method of claim 1, wherein the estimated intake oxygen concentration is tuned to an output of the intake oxygen sensor at the reference intake pressure by dividing the estimated intake oxygen concentration with a pressure correction factor at a current pressure measurement.A method for an engine, comprising: learning a relationship between a first intake oxygen sensor output estimated at a first intake pressure during a first engine idle since engine start; and the first intake pressure; and adjusting EGR flow to the engine at a second intake pressure based on a second intake oxygen sensor output estimated at the second intake pressure and the learned relationship, wherein the first and second intake oxygen sensor outputs are generated by an intake oxygen sensor coupled upstream of an intake throttle and downstream of an intercooler, and wherein the first and second intake pressures are estimated by an intake pressure sensor coupled upstream of the intake throttle and downstream of the intercooler.The method of claim 11, wherein the adjusting comprises calculating a pressure correction factor based on a difference between the first intake pressure and the second intake pressure; calculating a humidity correction factor based on a difference between ambient humidity at the second intake pressure and a reference humidity; modifying the second intake oxygen sensor output based on both the calculated pressure correction factor, the humidity correction factor, and the learned relationship; and adjusting a position of an EGR valve based on the modified second intake oxygen sensor output.The method of claim 12, wherein the EGR valve is coupled in a low pressure EGR passage, and wherein learning is performed at a first engine idle following each engine restart.The method of claim 11, wherein the learning is performed at a first engine idle following installation of the intake oxygen sensor and / or intake pressure sensor in the engine.The method of claim 12, further comprising indicating degradation of the EGR valve based on the first intake oxygen sensor output estimated at the first intake pressure during the first engine idle since the engine start relative to a second intake oxygen sensor output estimated at the first intake pressure during an engine skip fire condition.An engine system, comprising: an engine including an intake manifold; a turbocharger including an exhaust turbine and an intake compressor; an intercooler coupled downstream of the compressor; an intake oxygen sensor coupled to the intake manifold downstream of the intercooler and upstream of an intake throttle; a pressure sensor coupled to the intake manifold downstream of the intercooler and upstream of the intake throttle; an EGR system including an EGR passage and an EGR valve for recirculating residual exhaust gases from downstream of the turbine to upstream of the compressor; and a controller with computer readable instructions for: during a first engine idle since an engine start, learning a reference point for the oxygen sensor at a reference intake pressure; adjusting an opening of the EGR valve based on an intake oxygen concentration estimated by the sensor with respect to the learned reference point and further based on an intake pressure with respect to the reference intake pressure.The system of claim 16, further comprising a humidity sensor for estimating an ambient humidity, wherein the controller comprises further instructions for further adjusting the opening of the EGR valve based on an ambient humidity relative to a reference humidity.

Images