DE102013113647B4 - METHOD OF CONTROLLING A DRIVE TRAIN FOR THE PRODUCTION OF AMMONIA IN A THREE-WAY CATALYST - Google Patents

Abstract

A method for controlling a powertrain that includes an internal combustion engine (10, 300, 600) having multiple cylinders (20) and an exhaust aftertreatment system (15) that includes a device (220) for selective catalytic reduction that uses ammonia as a reductant, wherein the method comprising:depleting oxygen from an ammonia production catalyst (630, 632) disposed between the engine (10) and the selective catalytic reduction device (220) and connected to the plurality of cylinders (20), comprising that for the cylinders (20) are selected with an air/fuel ratio having an operating range from stoichiometric to rich;after depleting oxygen from the ammonia production catalyst (630, 632), initiating an ammonia production cycle comprising:operating a plurality of the cylinders (20) cooperatively, wherein a proportion of the plurality of cylinders (20) operate at air/fuel ratios in a first range of stoichiometry trically operated too rich that is conducive to producing NOx and wherein a remaining proportion of the plurality of cylinders (20) at air/fuel ratios in a second region having a richer air/fuel ratio than the first region comprises, is operated, which is conducive to generate molecular hydrogen; and using the ammonia production catalyst (630, 632) to produce ammonia.

Description

This disclosure relates to control of aftertreatment of NOx emissions in internal combustion engines, and more particularly to a method of controlling a powertrain.

Emission control is an important factor in engine design and engine control. Nitrogen oxides (NOx), a well-known by-product of combustion, are produced when nitrogen and oxygen molecules present in the engine intake air separate at high combustion temperatures. Rates of NOx creation follow known relationships to the combustion process, for example, higher rates of NOx creation are associated with higher combustion temperatures and longer exposure of air molecules to the higher temperatures.

Once NOx molecules are created in the combustion chamber, they can be converted back to nitrogen and oxygen molecules in example devices known in the art within the broader category of aftertreatment devices. However, one of ordinary skill in the art will understand that aftertreatment devices are dependent in large part on operating conditions, such as device operating temperature controlled by exhaust stream temperatures and engine air/fuel ratio. In addition, aftertreatment devices include materials such as catalyst beds that are susceptible to damage or degradation as a result of usage over time and exposure to high temperatures.

Modern engine control methods use various operating strategies to optimize combustion. Some operating strategies that optimize combustion for fuel economy include lean, localized, or stratified combustion in the combustion chamber to reduce the fuel charge needed to achieve the workload required of the cylinder and increase engine efficiency, for example, by running in an unthrottled state, which reduces intake air pumping losses. While temperatures within the combustion chamber in combustion pockets can become high enough to produce significant amounts of NOx, the overall energy output of the combustion chamber, particularly the heat energy released from the engine through the exhaust stream, can be greatly reduced from normal levels. Such conditions can pose a challenge to exhaust aftertreatment strategies because, as noted above, the aftertreatment devices often require an elevated operating temperature, controlled by the exhaust stream temperature, to perform adequately to treat NOx emissions.

For example, aftertreatment devices are known that use chemical reactions to treat constituents in the exhaust stream. An example device includes a Selective Catalytic Reduction ('SCR') device. Known uses of an SCR device use ammonia derived from urea injection to treat NOx. Ammonia stored on a catalyst bed in the SCR reacts with NOx, preferably at a desired level of NO and NO ₂ , and produces favorable reactions to treat the NOx. An exemplary embodiment includes a preferred ratio of NO to NO ₂ of one to one and is known as a fast SCR reaction. It is known to operate a Diesel Oxidation Catalyst (DOC) upstream of the SCR in diesel applications to convert NO to NO ₂ for preferential treatment in the SCR. Continuous improvement of exhaust gas aftertreatment requires precise information regarding NOx emissions in the exhaust stream in order to achieve effective NOx reduction, for example dosing the right amount of urea based on monitored NOx emissions.

Additionally, other aftertreatment devices are known for treating constituents in the exhaust stream. In gasoline applications, in particular, Three Way Catalysts ('TWC') are used to treat components. Lean NOx traps ("NOx traps") use catalysts that can store some amount of NOx, and engine control technologies have been developed to combine these NOx traps or NOx adsorbers with fuel efficient engine control strategies to improve fuel efficiency and still achieve acceptable NOx emission levels. An example strategy includes using a lean NOx trap to store NOx emissions during fuel-lean operations, and then purging the stored NOx during fuel-rich, higher temperature engine operating conditions with conventional three-way catalysis to nitrogen and water. Diesel Particulate Filters (DPF) retain soot and particulate matter in diesel applications, and the retained material is periodically purged during high temperature regeneration processes.

Urea injection in a powertrain can be challenging. Maintaining storage and replenishment of urea can be difficult. Urea is prone to freezing in common regions under normally varying climatic conditions.

DE 10 2009 054 046 A1 discloses a method of operating an internal combustion engine comprising controlling the engine to a preferred air/fuel ratio to produce an exhaust gas feedstream from the engine having a preferred concentration of nitrogen oxide, carbon monoxide and hydrogen, the nitrogen oxide, carbon monoxide and the hydrogen is converted to ammonia over a first catalyst device; and the ammonia is stored at an ammonia-using selective catalytic reduction device fluidly connected in series downstream of the first catalyst device.

Further prior art is in DE 10 2008 037 156 A1 described.

The object of the invention is to create a method for controlling a power train with which it is possible to simplify storage and refilling of urea in aftertreatment systems with SCR devices that require urea injection.

The object is solved by the subject matter of claims 1 and 10. Advantageous developments of the invention are described in the dependent claims.

A powertrain includes an internal combustion engine having multiple cylinders and an aftertreatment system that includes a selective catalytic reduction device that uses ammonia as a reductant. An ammonia generation cycle includes operating a proportion of the cylinders at an air/fuel ratio conducive to producing molecular hydrogen and operating a proportion of the cylinders at an air/fuel ratio conducive to NOx generate. An ammonia production catalyst is used between the engine and the selective catalytic reduction device to produce ammonia.

• 1 eine schematische Darstellung ist, die einen Verbrennungsmotor, ein Steuermodul und ein Abgasnachbehandlungssystem gemäß der vorliegenden Offenbarung zeigt;
• 2 schematisch ein beispielhaftes Nachbehandlungssystem gemäß der vorliegenden Offenbarung veranschaulicht, das eine Harnstoffdosierkonfiguration umfasst;
• 3 graphisch einen beispielhaften Betrieb eines Motors und die resultierende Erzeugung einer Anzahl von chemischen Verbindungen in einem Ammoniak umfassenden Abgasstrom über verschiedene Luft/ Kraftstoff-Verhältnisse gemäß der vorliegenden Offenbarung veranschaulicht;
• 4 graphisch ein zusätzliches Beispiel eines Betriebs eines Motors und einer resultierenden Erzeugung einer Anzahl von chemischen Verbindungen in einem Ammoniak umfassenden Abgasstrom über verschiedene Luft/ Kraftstoff-Verhältnisse gemäß der vorliegenden Offenbarung zeigt;
• 5 eine Tabelle von Probereaktionsgemischen, die in den ersten chemischen Reaktor eingebracht wurden, gemäß der vorliegenden Offenbarung veranschaulicht;
• 6 graphisch Ammoniakerzeugungswerte über einen Bereich von Luft/ Kraftstoff-Verhältnissen und Reaktionstemperaturen gemäß der vorliegenden Offenbarung darstellt;
• 7 graphisch Ammoniakwerte gemäß der vorliegenden Offenbarung, die von einem ersten chemischen Reaktor unter Verwendung eines Standardreaktionsgemisches und abgewandelter Reaktionsgemische erzeugt wurden, gegenüber der Temperatur darstellt;
• 8 graphisch Ammoniakwerte gemäß der vorliegenden Offenbarung, die von dem ersten chemischen Reaktor unter Verwendung des Standardreaktionsgemisches und eines abgewandelter Reaktionsgemisches erzeugt wurden, gegenüber der Temperatur darstellt;
• 9 graphisch vier unterschiedliche beispielhafte Motorsteuerstrategien und sich ergebende Motoremissionen unter einem festen Satz von Betriebsbedingungen gemäß der vorliegenden Offenbarung veranschaulicht;
• 10 schematisch eine beispielhafte bestimmte Ausführungsform eines Antriebstranges, der zur Verwendung der hier beschriebenen Verfahren konfiguriert ist, gemäß der vorliegenden Erfindung zeigt;
• 11 schematisch eine beispielhafte Anordnung von Katalysatoren in einem Nachbehandlungssystem zum Verwirklichen einer Erzeugung von Ammoniak zur Verwendung in einer SCR-Vorrichtung gemäß der vorliegenden Offenbarung veranschaulicht;
• 12 schematisch ein beispielhaftes NOx-Modellmodul gemäß der vorliegenden Offenbarung darstellt, das in einem Motorsteuermodul genutzt wird und eine Schätzung von NOx-Erzeugung ermittelt;
• 13 graphisch eine beispielhafte Massenanteilverbrennungskurve gemäß der vorliegenden Offenbarung veranschaulicht;
• 14 graphisch gemäß der vorliegenden Offenbarung einen beispielhaften Zylinderdruck während eines Verbrennungsprozesses veranschaulicht, der gegenüber dem Kurbelwinkel aufgetragen ist;
• 15 eine Anzahl von unterschiedlichen, in dem Brennraum schätzbaren Temperaturen gemäß der vorliegenden Offenbarung darstellt, die für das Beschreiben des Verbrennungsprozesses wichtig sind;
• 16 eine graphische Darstellung von beispielhaften modellierten Ergebnissen gemäß der vorliegenden Erfindung ist, die standardisierte Wirkungen einer Anzahl von Eingaben zu NOx-Emissionen unter einem vorgegebenen Satz von Bedingungen beschreiben; und
• 17 schematisch ein beispielhaftes System darstellt, das eine Schätzung von NOx-Erzeugung generiert, wobei es Modelle in einem neuronalen Netz nutzt, um Schätzungen von NOx-Erzeugung zu generieren, und ein dynamisches Modellmodul umfasst, um NOx-Erzeugungsschätzungen bezüglich der Wirkungen von dynamischen Motor- und Fahrzeugbedingungen gemäß der vorliegenden Offenbarung zu kompensieren.

One or more embodiments will now be described by way of example with reference to the accompanying drawings, in which:

• 1 12 is a schematic diagram showing an internal combustion engine, a control module, and an exhaust aftertreatment system according to the present disclosure;
• 2 schematically illustrates an exemplary aftertreatment system including a urea dosing configuration, according to the present disclosure;
• 3 graphically illustrates exemplary operation of an engine and the resulting generation of a number of chemical compounds in an exhaust stream comprising ammonia over various air/fuel ratios, in accordance with the present disclosure;
• 4 13 graphically depicts an additional example of operation of an engine and resultant generation of a number of chemical compounds in an exhaust stream comprising ammonia over various air/fuel ratios, in accordance with the present disclosure;
• 5 illustrates a table of sample reaction mixtures introduced into the first chemical reactor, in accordance with the present disclosure;
• 6 graphically depicts ammonia production values over a range of air/fuel ratios and reaction temperatures, in accordance with the present disclosure;
• 7 graphically depicting ammonia levels produced by a first chemical reactor using a standard reaction mixture and modified reaction mixtures versus temperature, in accordance with the present disclosure;
• 8th graphically depicting ammonia levels produced by the first chemical reactor using the standard reaction mixture and a modified reaction mixture versus temperature, in accordance with the present disclosure;
• 9 graphically illustrates four different example engine control strategies and resulting engine emissions under a fixed set of operating conditions, in accordance with the present disclosure;
• 10 12 schematically depicts an exemplary specific embodiment of a powertrain configured to use the methods described herein, in accordance with the present invention;
• 11 schematically illustrates an exemplary arrangement of catalysts in an aftertreatment system for realizing generation of ammonia for use in an SCR device according to the present disclosure;
• 12 FIG. 12 schematically depicts an exemplary NOx model module utilized in an engine control module and determining an estimate of NOx creation, in accordance with the present disclosure;
• 13 graphically illustrates an exemplary mass fraction burn curve, in accordance with the present disclosure;
• 14 graphically illustrates an exemplary cylinder pressure plotted against crank angle during a combustion process, in accordance with the present disclosure;
• 15 Figure 12 illustrates a number of different temperatures that can be estimated in the combustion chamber that are relevant to describing the combustion process, in accordance with the present disclosure;
• 16 Figure 12 is a graphical representation of exemplary modeled results describing standardized effects of a number of inputs on NOx emissions under a given set of conditions, in accordance with the present invention; and
• 17 12 schematically depicts an example system that generates an estimate of NOx creation using models in a neural network to generate estimates of NOx creation and including a dynamic model module to generate NOx creation estimates related to the effects of dynamic engine and to compensate for vehicle conditions in accordance with the present disclosure.

1 1 is a schematic diagram depicting an internal combustion engine 10 and control module 5 and exhaust aftertreatment system 15 according to the present disclosure. The exemplary engine includes a multi-cylinder direct injection internal combustion engine having reciprocating pistons 22 attached to a crankshaft 24 that are moveable within cylinders 20 that define variable volume combustion chambers 34 . Engines are known to operate under compression ignition or spark ignition. Furthermore, methods are known to either use an ignition strategy in a single engine, modifying the strategy based on factors such as engine speed and load. Engines are also known to operate in hybrid strategies, such as assisted ignition compression ignition strategies. This disclosure is intended to include, but not be limited to, these example embodiments of engine operation. The crankshaft 24 is operatively attached to a vehicle transmission and driveline system to provide traction torque thereto in response to a driver torque request (TO_REQ). The engine preferably utilizes four-stroke operation, with each engine combustion stroke comprising 720 degrees of angular rotation of the crankshaft 24 divided into four stages of 180 degrees of intake-compression-work-exhaust that describe the reciprocation of the piston 22 in the engine cylinder 20. A multi-tooth target wheel 26 is attached to and rotates with the crankshaft. The engine includes sensing devices for monitoring engine operation and actuators that control engine operation. The sensing devices and actuators are signal-connected or functionally connected to the control module 5 .

The engine preferably comprises a direct injection four-stroke internal combustion engine having a variable volume combustion chamber defined by the piston reciprocating in the cylinder between top dead center and bottom dead center and a cylinder head including an intake valve and an exhaust valve. contains. The piston reciprocates in repetitive cycles, each cycle including intake, compression, power, and exhaust strokes.

The engine preferably has an air/fuel operating regime that is predominantly lean of stoichiometry. One of ordinary skill in the art understands that aspects of the invention are applicable to other engine configurations that operate primarily lean of stoichiometry, e.g., lean-burn spark-ignition engines. During normal operation of the compression ignition engine, a combustion event occurs during each engine cycle when a charge of fuel is injected into the combustion chamber to form the cylinder charge with the intake air. The charge is then burned by the action of its compression or with the initiation of an ignition spark from a spark plug during the compression stroke.

The engine is designed to operate over a wide range of temperatures, cylinder charge (air, fuel, and EGR), and injection events. The procedures described herein are ins particularly suitable for operation with direct injection engines that work over-stoichiometrically. The methods set forth herein are applicable to multiple engine configurations, including spark-ignition engines, compression-ignition engines, including those configured to use Homogeneous Charge Compression Ignition (HCCI) strategies. The methods are applicable to systems that use multiple fuel injection events per cylinder per engine cycle, e.g. a system that uses pilot injection for fuel reforming, main injection event for engine performance and optionally post combustion fuel injection, retard combustion fuel injection event for aftertreatment control, all affecting the same cylinder pressure.

Sensing devices are installed on or near the engine to monitor physical properties and generate signals that can be correlated to engine and environmental parameters. The sensing devices include a crankshaft rotation sensor that includes a crankshaft angle encoder 44 for monitoring crankshaft speed (RPM) through sensing edges on the teeth of the multi-tooth target wheel 26 . The crankshaft sensor is known and can include a Hall sensor, an inductive sensor or a magnetoresistive sensor, for example. The signal output from the crankshaft angle sensor 44 (rpm) is input to the control module 5 . There is a combustion pressure sensor 30 comprising a pressure sensing device designed to monitor in-cylinder pressure (COMB_PR). Combustion pressure sensor 30 preferably comprises a non-intrusive device comprising a force transducer having an annular cross-section adapted to be installed at an opening for a glow plug 28 in the cylinder head. The combustion pressure sensor 30 is installed with the glow plug 28, with combustion pressure being mechanically transmitted to the sensor 30 through the glow plug. The output signal com_pr of the sensing element of the sensor 30 is proportional to the cylinder pressure. The sensing element of the sensor 30 comprises a piezoceramic or other device configurable as such. Other sensing devices preferably include a manifold pressure sensor for monitoring manifold pressure (MAP) and ambient barometric pressure (BARO), a mass air flow meter for monitoring intake air mass flow (MAF) and intake air temperature (T _IN ), and a coolant sensor 35 (COOLANT). The system may include an exhaust gas sensor (not shown) for monitoring conditions of one or more exhaust gas parameters, eg, temperature, air/fuel ratio, and constituents. One of ordinary skill in the art understands that there may be other sensing devices and methods for control and diagnostic purposes. Driver input in the form of driver torque request TO_REQ may typically be obtained through an accelerator pedal and a brake pedal, among other devices. The engine is preferably equipped with other sensors (not shown) for monitoring operation and for system control purposes. Each of the sensing devices is signal connected to the control module 5 to provide signal information which is converted by the control module into information representative of the particular parameter being monitored. It is understood that this configuration is illustrative, not restrictive, implying that the various sensing devices are interchangeable with functionally equivalent devices and algorithms and still fall within the scope of the invention.

The actuators are mounted on the engine and are controlled by the control module 5 in response to driver inputs to achieve various performance goals. Actuators include an electronically controlled throttle device that controls opening of a throttle valve to a commanded input (ETC), and a plurality of fuel injectors 12 for injecting fuel directly into each of the combustion chambers in response to a commanded input (INJ_PW), all in response to the Driver torque request (TO_REQ) are controlled. There is an exhaust gas recirculation valve 32 and a cooler (not shown) that controls the flow of externally recirculated exhaust gas to the engine intake in response to a control signal (EGR) from the control module. The glow plug 28 comprises a known device installed in each of the combustion chambers that is adapted for use with the combustion pressure sensor 30 .

Fuel injector 12 is a component of a fuel injection system that includes a plurality of high pressure fuel injectors, each configured to directly inject a fuel charge, comprising a mass of fuel, into one of the combustion chambers in response to the INJ_PW instruction signal from the control module. Each of the fuel injectors 12 is supplied with pressurized fuel from a fuel rail system (not shown) and has operational characteristics that include a minimum pulse width and associated controllable minimum and maximum fuel flow rates.

The engine may be equipped with a controllable valve train operable to control the opening and closing of intake and exhaust valves of each of the cylinders, including one or more of valve timing, phasing (ie, timing relative to crank angle and piston position), and magnitude of lift of valve opening. An example system includes variable cam phasing applicable to compression-ignition engines, spark-ignition engines, and homogeneous-charge compression-ignition engines.

The control module 5 is preferably a general purpose digital computer, generally comprising a microprocessor or main processor, storage media comprising non-volatile memory including read only memory (ROM) and electrically programmable read only memory (EPROM), random access memory (RAM), a high speed clock generator, a includes analog-to-digital (A/D) circuitry and digital-to-analog (D/A) circuitry, as well as input/output (I/O) circuitry and devices, and appropriate signal conditioning and buffering circuitry. The control module has a set of control algorithms, comprising resident program instructions and calibrations stored in non-volatile memory and executed to provide the appropriate functions for each computer. The algorithms may be executed during predetermined iterations of the loop such that each algorithm is executed at least once per iteration of the loop. Algorithms are executed by the central processing unit and are used to monitor inputs from the aforementioned sensing devices and execute control and diagnostic routines to control the operation of the actuators using predetermined calibrations. Loop cycles are typically performed at regular intervals, for example each 3.125, 6.25, 12.5, 25 and 100 milliseconds during ongoing engine and vehicle operation. Alternatively, algorithms can be executed in response to an event occurring.

The control module 5 executes algorithmic code stored therein to control the aforementioned actuators to control engine operation, including throttle position, fuel injection mass and timing, EGR valve position to control the flow of recirculated exhaust gases, glow plug operation, and control of intake and/or or to control exhaust valve timing, phasing and lift on systems so equipped. The control module is configured to receive input signals from the driver (e.g., an accelerator pedal position and a brake pedal position) to determine the driver torque request TO_REQ and from sensors indicative of engine speed (RPM) and intake air temperature (T _IN ), as well as coolant temperature and other ambient conditions. to recieve.

1 describes an exemplary gasoline engine. However, it is understood that NOx treatment and aftertreatment systems are used in other engine configurations, including diesel engines, and the disclosure is not intended to be limited to the specific example engine embodiment described herein.

2 schematically illustrates an exemplary aftertreatment system including a urea dosing configuration, in accordance with the present disclosure. The aftertreatment system 200 includes a control module 205, a DOC 210, an SCR 220, an upstream NOx sensor 230, a downstream NOx sensor 240, a temperature sensor 250, and a urea dosing module 260. As is known in the art, takes over a DOC 210 a set of catalytic functions required for aftertreatment of an exhaust stream. One of the functions of the DOC 210 is to convert NO, a form of NOx that is not easily treated in an SCR, to NO ₂ , a form of NOx that is easily treated in an SCR. The SCR 220 uses urea as a reactant to reduce NOx to other molecules. The upstream NOx sensor 230 detects and quantifies NOx in the exhaust stream entering the aftertreatment system 200 . While the upstream NOx sensor 230 is illustrated as an exemplary means of quantifying NOx flowing into the aftertreatment system, it should be noted that NOx entering the system may be quantified by other means for use in evaluating conversion efficiency in an SCR , for example by a NOx sensor placed between the DOC 210 and the SCR 220 or by a virtual NOx sensor that models engine performance and conditions in the exhaust stream to estimate the presence of NOx entering the aftertreatment system. This disclosure discusses a sensor input describing NOx entering the aftertreatment system according to the exemplary embodiment, but it should be understood that depending on an upstream placement of the sensor, the input could describe the NOx level entering a portion of the aftertreatment system. The SCR 220 uses ammonia, derived from, for example, injected urea, to convert NOx into other molecules by methods known in the art. A temperature sensor 250 is shown positioned in an area positio is ned to detect exhaust stream temperatures in the aftertreatment system 200 . A urea dosing module 260 is depicted at a location upstream of the SCR 220 . The urea can be injected directly into the exhaust stream entering the SCR. However, a preferred method utilizing a mixer device 270 is illustrated. The urea dosing module 260 injects urea onto the mixer device 270 and then the urea is carried by the exhaust flow in a substantially even distribution onto the catalyst surfaces on the interior of the SCR 220 . The downstream NOx sensor 240 detects and quantifies NOx in the exhaust stream exiting the aftertreatment system 200 . The control module 205 includes programming required to process inputs associated with the aftertreatment system and may include programming to use the methods described herein.

Ammonia as a reductant can be introduced into the aftertreatment system by injecting urea as described above. However, storing and maintaining sufficient urea levels in a mobile or consumer powertrain can be problematic. One of ordinary skill in the art will understand that ammonia is a known byproduct of the combustion and aftertreatment process. Known methods optimize the combustion process and the use of after-treatment devices to reduce the occurrence of ammonia in order not to create another substance that needs to be converted. A method is disclosed to instead selectively mitigate the operation of the combustion cycle and use aftertreatment devices conducive to generating ammonia periodically in an ammonia generation cycle and storing that ammonia for subsequent NOx conversion.

Ammonia may be generated in a catalyst device, such as a TWC device. Such ammonia (NH ₃ ) generation results from an exemplary conversion process described by the following equation. NO+CO+ _1.5H2 -> _NH3 + _CO2 [1]

One of ordinary skill in the art will recognize that this conversion requires molecular oxygen to be removed from the catalyst before NO reacts with the molecular hydrogen. Excess oxygen is often present when the internal combustion engine is operated in lean operating modes where the air/fuel ratio (AFR) is lean of stoichiometry or with excess air. As such, using a selectable ammonia generation cycle requires controlling the AFR to a value that is determined to deplete oxygen in the exhaust stream. Furthermore, choosing an AFR in the stoichiometric and rich operating regions further facilitates the production of ammonia, for example by producing NO and H ₂ in appropriate amounts. In the example equation above, an ideal ratio of 1.5 to one is evident. However, based on the environment provided by the catalyst and other reactions occurring in the aftertreatment device, a different actual ratio may effect optimal production of ammonia. It has been determined that an example test level utilizing a particular example catalyst performs optimally at a ratio of between three and five hydrogen molecules to one NO molecule. Choosing a catalyst that allows for lower hydrogen to NO ratios is preferred because hydrogen demand is directly related to an amount of fuel that must be consumed to enable ammonia production. Calibration according to test results or modeling according to methods sufficient to accurately estimate operation of the combustion cycle and aftertreatment processes and conversions can be used to choose an AFR that is helpful in controlling an ammonia production cycle. One of ordinary skill in the art will recognize that CO presence must also be taken into account in order to promote the reaction described above.

The timing of an ammonia generation cycle can be controlled or selected according to a number of factors affecting ammonia utilization in the SCR device, including estimated ammonia storage at the catalyst, estimated or detected ammonia slip, estimated or detected NOx breakthrough past the SCR device, and Engine operation conducive to operation in an ammonia generation cycle. Monitoring of these factors can be accomplished by monitoring a number of inputs including engine operation, exhaust gas characteristics, and NOx conversion efficiency in the SCR device. Periods of engine acceleration have been found to typically include higher levels of NOx and hydrogen production, and AFR closer to stoichiometric. Such periods of time that are conducive to ammonia production can be used to induce intrusive operation of an ammonia production cycle under less favorable engine conditions drive to minimize. The length of operation of an ammonia production cycle will vary depending on the ammonia production required, the details of the system employed, and the particular operation of the engine.

Molecular hydrogen production required for ammonia production can occur through the combustion process in the engine. Combustion in an AFR-rich environment where molecular oxygen is scarce tends to produce elevated levels of molecular hydrogen. Hydrogen production may occur as a result of a single injection combustion cycle, with hydrogen production resulting from a primary combustion process that provides work to the engine.

3 1 graphically illustrates exemplary operation of an engine and the resulting formation of a number of chemical compounds in an exhaust stream comprising ammonia over various air/fuel ratios during a single injection combustion cycle, in accordance with the present disclosure. The exemplary test results represent the operation of an engine on a dynamometer using lean-burn spark-ignition with direct injection and running at a speed of 2000 rpm and a load of 2 bar. As described above, changing the AFR changes the chemical composition of the exhaust stream. It is known that stoichiometric operation in gasoline engines occurs at an AFR of about 14.7 to one. AFR values greater than 14.7 describe lean operation or excess air operation. AFR values below 14.7 describe rich or excess fuel operation. In the example dataset from 3 engine out NOx is shown to decrease with decreasing AFR, and engine out H ₂ is shown to increase with decreasing AFR. The resulting presence of NH ₃ exiting the TWC is shown to first increase, peak at an exemplary value of about 14.2, and then decrease with decreasing AFR. As a result, in the exemplary configuration that includes the particular catalyst used in generating the in 3 Using the data set shown, an ammonia production cycle is best operated at an AFR equal to 14.2. However, as described above, other configurations, and particularly other catalysts, may change the ratio of hydrogen and NOx to best promote ammonia production. This allows the selected AFR to deviate from the value of 14.2 given in the example above.

4 12 graphically illustrates another example of operation of an engine and the resulting formation of a number of chemical compounds in an exhaust stream comprising ammonia over various air/fuel ratios during a single injection combustion cycle, in accordance with the present disclosure. The exemplary test results represent the operation of an engine on a dynamometer using lean-burn spark-ignition with direct injection and running at a speed of 1500 rpm and a load of 1 bar. As above in connection with 3 described 4 represents ammonia production over a range of AFR values. Ammonia production again peaks at a certain AFR value and is controlled in part by the presence of molecular hydrogen and NOx. In the sample test results from 4 peak ammonia production occurs at an AFR of about 14.2. As described above, this value depends on the properties of the catalyst used.

5 - 8th graphically illustrate test results utilizing a single injection to form ammonia and depict reactants introduced to a first chemical reactor comprising a first TWC brick and a second TWC brick configured to form TWC devices to simulate in a vehicle exhaust stream. 5 Figure 12 shows a table of sample reaction mixtures introduced into the first chemical reactor according to the present disclosure. Each sample reaction mixture includes values of constituent gases determined based on engine models simulating exhaust gas compositions at selected engine air/fuel ratios. The ideal average ratio ('Ideal Average A/F') is the target engine air/fuel ratio that would produce exhaust gas compositions that correlate with the sample reaction mixtures based on the engine models. The calculated average air/fuel ratio ('Calculated Average A/F') is the modeled air/fuel ratio achieved based on the actual reactant measurements. Calculated Average Lambda ('Calculated Average Lambda') is the lambda value for the calculated average air/fuel ratio. The amounts of oxygen ('% O ₂ '), carbon monoxide ('% CO'), hydrogen ('% H ₂ '), carbon dioxide ('% CO ₂ '), water ('% H ₂ O') hydrocarbons (' ppm HC') and nitrogen oxide ('%NO') contained in each contained in the sample reaction mixture were measured. Furthermore, each sample reaction mixture includes a sulfur dioxide ('SO ₂ ') value of 2.7 ppm.

6 FIG. 12 graphically depicts ammonia production levels over a range of air/fuel ratios and reaction temperatures, in accordance with the present disclosure. The graph depicts ammonia levels ('NH ₃ (ppm)') produced by the first chemical reactor at target air/fuel ratios (' A/F ratio (+/- 0.25 A/F')) and at reaction temperatures of 300°C, 400°C, 500°C and 600°C. At each reaction temperature, the highest levels of ammonia were produced at the target air/fuel ratio of 14.2 and generally decreases as the air/fuel ratio increases. Furthermore, ammonia values decrease with increasing reaction temperature from 300°C to 600°C at the desired air/fuel ratio of 14.2.

7 graphically depicts ammonia ('NH ₃ (ppm)') levels produced by the first chemical reactor using a standard reaction mixture (STD = m.H ₂ O, m.H ₂ , m.HC, m.CO m.O ₂ ) and of a modified reaction mixture versus temperature ('Temperature °C') in accordance with the present disclosure 14, 2 in the table of 5 The modified reaction mixtures contain a sample reaction mixture comprising constituent amounts of the standard reaction mixture but without water ('o. H ₂ O'), a sample reaction mixture comprising constituent amounts of the standard reaction mixture but with elevated levels of carbon monoxide in place of hydrogen ( 'o. H ₂ (CO adjust)'), and a sample reaction mixture comprising constituent amounts of the standard mixture but with increased levels of oxygen instead of hydrogen (('o. H ₂ (O ₂ adjust)').

8th graphically represents ammonia values ('NH ₃ (ppm)') produced by the first chemical reactor using the standard reaction mixture ('STD = m.H ₂ O, m.H ₂ , m.HC, m.CO m.O ₂ ' ) and a modified reaction mixture versus temperature ('Temperature C') according to the present disclosure. The standard reaction mixture comprises water, hydrogen, hydrocarbon, carbon monoxide and oxygen in the amounts required for the sample reaction mixture with a target air/fuel ratio of 14, 2 in the table of 5 The modified reaction mixtures contain a sample reaction mixture comprising component amounts of the standard reaction mixture, but with oxygen in place of one-half the amount of hydrocarbon ('m. ½ HC (adjust O ₂ )'). 8th Figure 12 further represents ammonia produced by a second chemical reactor utilizing the default reaction mixture ('1st Brick Only'), where the second reactor comprises only a first TWC Brick with no additional TWC Bricks.

Both hydrogen production and NOx production in a single injection combustion cycle can be modulated in different ways. 9 1 graphically illustrates four different example engine control strategies and resulting engine emissions under a fixed set of operating conditions, in accordance with the present disclosure. All tests were conducted in a single engine configuration operating at 1000 rpm and an engine load of 3 bar. A first engine control strategy, established as a baseline data set, includes operation with a standard valve strategy (95/-95 (IMOP/EMOP)), 31% EGR, and an AFR of 22:1. A second engine control strategy, defined as a high valve overlap (HVO) data set, includes operation with a modified valve strategy (95/-80 (IMOP/ EMOP)) that includes a period in which both an intake valve and an exhaust valve are open, a condition known in the art as internal EGR, and an AFR of 14:1. Exemplary high valve overlap strategies include substantially symmetric opening and closing of intake and exhaust valves about a top dead center crank angle. A third engine control strategy, defined as a Late Intake Valve Close (LIVC) dataset, includes operation with a modified valve strategy (140/-80 (IMOP/ EMOP)) that involves holding an intake valve open over longer duration than the standard valve strategy and an AFR of 14:1. A fourth engine control strategy, specified as 14:1 m.EGR, involves operation with a standard valve strategy (95/-95 (IMOP/EMOP)), 24% EGR and an AFR of 14:1. As is evident in the data, adjustment of AFR and other operating conditions can raise molecular hydrogen to high levels above levels present in the baseline data set. In addition, adjustments to valve strategies and EGR rates have an effect on NOx levels. But as in the data sets and when examining 3 and 4 As is evident, increased hydrogen production from single injection at lower AFR levels involves a limitation in NOx production, and NO levels may not be at the levels required to support the reaction described in Equation 1.

Engines using direct injection are known to include methods for injecting precise amounts of fuel into the combustion chamber at selected times of the combustion cycle through a direct injection fuel injection system. Those skilled in the art will recognize that direct injection coupled to an appropriate control module allows for control over combustion characteristics within a cylinder from combustion cycle to combustion cycle and control over combustion characteristics from cylinder to cylinder.

As described above, an exhaust flow comprising a mixture of molecular hydrogen and NOx may be used to generate ammonia through an ammonia generation catalyst. As above in connection with the 3 and 4 patent, hydrogen and NOx can both be produced in one combustion stroke, and manipulation of combustion properties, such as AFR, can affect how much of each substance is produced. However, using AFR as a control within a single combustion event has a limited ability to produce these substances since high AFR values increase NOx production and low AFR values increase molecular hydrogen production. A method is disclosed for producing molecular hydrogen and NOx for use in ammonia production by discretely controlling a plurality of cylinders, modulating AFR in at least one cylinder to produce molecular hydrogen and modulating AFR in at least one cylinder is used to generate NOx. By controlling operation on a cylinder-by-cylinder basis, fuel penalties associated with driving all cylinders to a rich AFR setting may be avoided.

Cylinders in a vehicle can be arranged in a number of patterns. For example, a common four-cylinder arrangement has an "in-line four" configuration in which all four cylinders use a single exhaust manifold to direct exhaust from the engine into the aftertreatment system. A common eight-cylinder configuration includes a "V-eight" design in which two banks of cylinders each use an exhaust manifold. Six-cylinder designs are known to have both "in-line six" and "V-six" configurations. Catalyst designs are known to depend on engine configuration, and placement of catalysts in the exhaust system is known to depend on proximity to the engine and resulting temperature and exhaust flow composition required for the catalyst. For example, a TWC used in one embodiment to contain an ammonia generation catalyst required for the present disclosure must be located relatively close to the engine to support the needs of the catalyst. Because of this requirement, V designs that use two exhaust manifolds often use two TWCs, one for each exhaust manifold. Since, in order to produce ammonia in the TWC, the component substances for the reaction used to produce ammonia must be present in the TWC, the above method using different cylinders to optimally produce molecular hydrogen and NOx must be in feed the same catalyst device. Therefore, in configurations such as a V configuration, multiple cylinders that are coordinated to produce hydrogen and NOx must feed into the same catalyst to effectively produce ammonia.

Cylinders modulated to help produce hydrogen and NOx can be operated in pairs, with one cylinder modulated to produce the required hydrogen and the other cylinder modulated to produce the required NOx, such as is described by example Equation 1. Additional cylinders in the same bank as the pair can be operated in a substance-neutral configuration that does not interfere with the resulting mixture of substances in the exhaust flow. Alternatively, the additional cylinder or cylinders can be selectively deactivated while the pair is producing the necessary substances for ammonia production. As described above, higher load conditions may support increased hydrogen and NOx production. Deactivation of one or more cylinders results in a greater load on the remaining cylinders, aiding in the production of hydrogen and NOx. Alternatively, a plurality of cylinders feeding into the same catalyst can be used together to produce the substances required to produce ammonia. For example, in a V-six cylinder configuration where three cylinders feed into a single TWC catalyst with an ammonia generation catalyst, one cylinder may be operated with lean of stoichiometry AFR optimized to produce a desired amount of NOx. The remaining two cylinders can each be optimized to produce each half of the targeted amount of hydrogen. By sharing the hydrogen production requirement between two cylinders is in combination with the 3 and 4 to recognize that the cylinders can be operated with a less rich AFR than is the case with a single cylinder in order to achieve the required amount of Was to produce hydrogen. In this way, substance production requirements can be divided among cylinders to selectively direct different portions of the cylinders. In the alternative, a pair of cylinders can be used, each producing a required ratio of one of the substances, and a third cylinder can be selectively added according to the method of FIG 3 and 4 be tuned to produce additional amounts of both hydrogen and NOx. Similarly, a block comprising four or six cylinders feeding into a single catalyst can split the required production of substances into a number of configurations. It is additionally noted that the cylinders running at a higher AFR and the cylinders running at a lower AFR are preferably chosen to balance a resultant work production in the engine. It should also be noted that the cylinders running at a higher AFR and the cylinders running at a lower AFR do not have to be static, and cylinders running at a given AFR can cycle from combustion cycle to combustion cycle, so long the desired mixture of substances generated in the exhaust stream is maintained. The selection of cylinder-to-cylinder operation and injection schedules used to generate the required substances may be developed experimentally, empirically, predictively, by modeling, or by other techniques appropriate to engine operation and resulting composition of exhaust gas flow, and multiple injection schedules from the same engine can be used for different engine settings, conditions, or operating ranges.

A specific embodiment for using the methods described above is shown schematically in 10 shown in accordance with the present disclosure. The powertrain 600 includes an engine 610 , an aftertreatment system 620 , and an EGR loop 640 . Engine 610 controls exhaust gas flow paths 622, 624, 626 and 628. Aftertreatment system 620 includes an ammonia production catalyst 630 fed through exhaust gas flow paths 622 and 624, an ammonia production catalyst 632 fed through exhaust gas flow paths 626 and 628, and an SCR device 634. This particular embodiment includes an EGR loop 640 having an EGR valve 645 that selectively directs exhaust flow from the aftertreatment system 620 to the intake of the engine 610 . According to the methods described herein, the exhaust flowpaths 622 and 624 fed by a pair of cylinders feeding into a single catalyst can be modulated by modulating the AFR in the associated cylinders in the engine 610 to have hydrogen and NOx in different levels is included. Similarly, exhaust gas flow paths 626 and 628 are fed through a similar pair of cylinders. By modulating AFR values with the various cylinders of engine 610, increased levels of hydrogen and NOx can be generated and delivered to catalysts 630 and 632. In the particular embodiment of 10 1, exhaust gas flow paths 622 and 628 are shown with the associated cylinders being operated at stoichiometric AFR, thereby producing increased levels of NOx. Also shown are exhaust gas flow paths 624 and 626 with the associated cylinders being operated at a rich AFR (lambda equal to 0.90 to 0.95), thereby producing increased levels of hydrogen. By modulating an operation of the pairs of cylinders, the power train can 10 Generate hydrogen and NOx, thereby enabling generation of ammonia according to the methods described herein.

Hydrogen may be generated in a combustion chamber by injecting fuel in an amount according to a target AFR prior to the main combustion event. In the alternative, fuel may be injected in a split injection where a portion of the fuel is injected before the main combustion event and a portion is injected after the main combustion event. According to either method, higher levels of hydrocarbons in the combustion chamber increase levels of hydrogen production resulting from combustion or in-cylinder reforming. In the alternative, hydrocarbons may be contained in the exhaust flow by controlling the main combustion event by, for example, injection or spark timing, timing split injection, or direct injection into the exhaust flow. In such a configuration, where hydrocarbons are present in the exhaust gas flow, a hydrogen-forming catalyst that aids in reforming the hydrocarbons at the catalyst may be used upstream or in-line with the ammonia-forming catalyst as an alternative method to in-cylinder hydrogen generation. A resulting substance production in each cylinder and resulting processes, including post-combustion reforming, can be estimated and used to balance the total production of substances for the bank of cylinders feeding the respective catalyst.

The reforming of hydrocarbons on a catalyst is exothermic and can generate significant heat. The temperature of the catalyst is preferably monitored or estimated to protect the catalyst from an over temperature condition. An example method may switch between in-stroke injection and post-stroke injection based on relevant parameters, preferably including catalyst temperature. This catalyst for forming hydrogen is located upstream of or substantially the same as the catalyst used to form ammonia, but may exist either as a separate device or as a catalyst in the same unitary aftertreatment device. Furthermore, catalyst designs are known that produce hydrogen even in the presence of molecular oxygen, which improves the efficiency of hydrogen production by reducing the need to inject additional fuel to completely deplete oxygen.

11 12 schematically illustrates an exemplary arrangement of catalysts in an aftertreatment system for realizing generation of ammonia in a combustion chamber for use in an SCR device according to the present disclosure. The powertrain 300 includes an engine 310, a stage 1 catalyst 320, a stage 2 catalyst 330, a stage 3 catalyst 340 and a stage 4 catalyst 350. A flow of exhaust gas emanates from the engine 310 and travels through the four catalysts away. The powertrain 300 is shown optimized for late combustion hydrocarbon reforming, as described above. Each catalyst promotes a different reaction according to methods known in the art. In the example configuration of 11 stage 1 catalyst 320 is selected to promote ammonia production according to Equation 1, stage 2 catalyst 330 is selected to promote operation according to normal operation of a TWC, stage 3 catalyst 340 is an SCR - Device that stores and uses ammonia to react with NOx and the stage 4 catalyst 350 is used to eliminate excess ammonia escaping the SCR device. The stage 1 catalyst may be utilized close to the engine, for example in a device fluidly connected to an exhaust manifold. An exemplary selection of catalysts in the various stages is summarized in Table 1: Table 1 Catalytic Device designation Catalyst Metal (PGM, CU, Fe) Washcoat (Zir or Zeo) Preferred substrate (metal/ceramic) step 1 NH ₃ generation PGM/ possibly non-PGM Alumina based cordierite Level 2 TWC PGM Alumina with OSC cordierite level 3 SCR Fe or Cu zeo cordierite Level 4 NH3 elimination PGM aluminum oxide cordierite

In this way, catalysts can be used to generate and utilize ammonia through late combustion hydrocarbon reforming in an aftertreatment system. As described above, a hydrogen-forming catalyst can be used to reform hydrocarbons in the aftertreatment system. In a system configured in this way, 11 show such a catalyst as a separate device upstream of the stage 1 catalyst (as a "stage 0 catalyst") or as a feature in the stage 1 catalyst.

Further, it should be understood that the aftertreatment systems can be embodied in many configurations known in the art and the chemical reaction used to generate ammonia can take a number of forms requiring different catalysts and different operating conditions. For example, different devices are used in the exhaust stream of a gasoline engine, such as a TWC device, and a diesel engine, such as a DOC device. The example configuration of 11 and configurations described below represent exemplary embodiments by which generation of ammonia in an aftertreatment system may be achieved, and this disclosure is not intended to be limited to the embodiments described herein. Furthermore, other reactions known that can be used to produce ammonia. For example, another reaction that can be used includes the following. 2.5 _H2 + NO -> NHs + _H2 O [2]

This reaction has the advantage of being independent of the presence of CO, but requires higher amounts of molecular hydrogen. Another example reaction that can be used to produce ammonia includes the following. Ba(NO ₃ ) ₂ + 8H ₂ -> 2NH ₃ + BaO + 5H ₂ O [3]

Utilizing this reaction requires a device that includes barium. As one of ordinary skill in the art is aware, barium is not known to be present in devices that use a PGM catalyst, such as a TWC, a DOC, or certain LNT devices, but is known to be used in most LNT devices that use barium for storage of NOx is used during lean operation. It is further understood that each of these reactions may require different catalysts and powertrain operating conditions for normal operation. In addition, the different ratios of NO and molecular hydrogen change the AFR required to efficiently operate an ammonia production cycle.

Catalyst design involves methods and preferences known in the art. Exemplary catalysts used in TWC design, which are used to produce ammonia as a result of a reaction described in Equation 1 above, as described above in connection with Table 1, preferably include a platinum and palladium-based (PGM) catalyst, however, the process can be utilized with certain non-PGM catalysts capable of producing the required reaction. The catalyst may be integrated into a close coupled catalyst device or a so-called transition catalyst device positioned near the exhaust manifold of the engine, or may be utilized in a separate device.

Ammonia generation cycles can be used as needed to provide ammonia to the SCR device. One method includes periodic ammonia production cycles based on periodic replenishment of an estimated demand. In the alternative, ammonia stored on the SCR catalyst or θ _NH3 can be estimated and used to schedule ammonia generation cycles as needed. Ammonia generation cycles that use stoichiometric or rich operation of the engine can be scheduled to use periods when such operation is already required according to powertrain performance requirements. Lean operation of an engine, particularly lean operation utilizing combustion processes such as homogeneous charge compression ignition or stratified charge modes, typically occurs at lower loads and lower engine speeds. For example, lean operation is often used during freeway driving where the engine is used in a steady state operation to maintain speeds. Rich operation is utilized where lean operation is not possible or preferred. For example, rich operation is often utilized in cases of acceleration, where generating force necessary to accelerate a vehicle requires high engine loads, and cycling through transmission operating range states requires engine speeds that include high engine speeds. Monitoring engine usage may enable initiation of an ammonia generation cycle in response to a transition to a rich operating mode. Additionally or alternatively, engine usage prediction may be statistical or coordinated with a 3D mapper, where ammonia production is predictively triggered based on expected engine usage already requiring high engine speeds or loads.

T₁ beschreibt die Temperatur des stromaufwärts der SCR-Vorrichtung gemessenen Abgasstroms, und T₂ beschreibt die Temperatur des stromabwärts der SCR-Vorrichtung gemessenen Abgasstroms. M_{DOT_ABG} beschreibt einen Massendurchsatz von Abgas durch die SCR-Vorrichtung und kann basierend auf dem Betrieb des Motors geschätzt oder modelliert werden. T_UMG beschreibt eine Temperatur von Umgebungsbedingungen zu dem Abgassystem und kann direkt gemessen oder basierend auf in üblicher Weise gemessenen Werten, beispielsweise der Ansauglufttemperatur, ermittelt werden. SV kann analog gemäß M_{DOT_ABG} und SCR-Geometrie vorhergesagt werden. Auf diese Weise kann eine Ammoniakproduktion zu Zeiten verwirklicht werden, bei denen ein übermäßiger Schlupf das Ammoniak nicht vorhersehbar aus der SCR-Vorrichtung abbaut.Ammonia generated by the above methods may be stored on a catalyst in an SCR device selected with an ammonia storage capability. As is known in the art, θ _NH3 depends on a number of properties of the exhaust flow, for example T _BED and SV. Increased catalyst bed temperatures or increased exhaust gas flow velocities in the SCR device cause slippage. Ammonia generation cycles may be scheduled predictively based on predicted ranges of T _BED and SV conducive to retaining stored ammonia. T _BED can be measured or predicted according to a model. An exemplary expression of T _BED can be given by the following functional relationship. $$T_{BETT}=f\left({\mathrm{<figure-callout\; id="1"\; label="T"\; filenames="DE102013113647B4\_0027.png,DE102013113647B4\_0029.png"\; state="\{\{state\}\}">T</figure-callout>}}_1,{\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="DE102013113647B4\_0030.png,DE102013113647B4\_0032.png"\; state="\{\{state\}\}">T</figure-callout>}}_2,M_{DOT\_ABG,}{T}_{uMG,}SC{R}_{GEOMETRIE}\right)$$

T ₁ describes the temperature of the exhaust stream measured upstream of the SCR device and T ₂ describes the temperature of the exhaust stream measured downstream of the SCR device. M _{DOT_ABG} describes a mass flow rate of exhaust gas through the SCR device and may be estimated or modeled based on engine operation. T _AMG describes a temperature of ambient conditions to the exhaust system and can be measured directly or determined based on values measured in a conventional manner, for example the intake air temperature. SV can be predicted analogously according to M _{DOT_ABG} and SCR geometry. In this way, ammonia production can be realized at times when excessive slip is not predictably depleting the ammonia from the SCR device.

Engine speeds and loads are important to ammonia generation cycles. In addition, engine operation can cause high temperature and high mass flow rates in the exhaust stream. The resulting conditions in the exhaust stream from operation of the engine may result in operating conditions that require wasteful injection of additional fuel or conditions that cause excessive slip in the SCR, causing ammonia breakdown. However, hybrid powertrains, which include an engine and other torque-generating devices, can provide a required output torque to a drive system while modulating the balance between the various devices of the powertrain. Other torque-generative devices may include an electric machine or machines that can operate in a torque-generative motor mode or an energy-regenerative generator mode. Such electrical machines are operatively connected to an energy storage device capable of supplying, receiving and storing electrical energy to the electrical machines. In this way, engine operation can be decoupled from required output torque to improve efficiency of ammonia production and storage in an aftertreatment system. For example, engine torque can be allowed to exceed required output torque, utilizing stoichiometric or rich engine operation conducive to ammonia production at high loads, and engine torque in excess of required output torque can be returned to the energy storage device by an electric machine. In this way, additional fuel used to generate hydrogen can produce stored energy rather than being completely dissipated as heat in the aftertreatment system. In another example, during high load operation, such as a vehicle towing a heavy object up a prolonged incline under wide open throttle conditions, exhaust gas temperatures resulting from high load operation of the engine can cause excessive slip in the engine Create SCR device. An electric machine or machines can be used to provide part of the required output torque, thereby reducing the load required of the engine, allowing the engine to operate at a transmission condition that permits lower engine speed and resultant temperatures in the exhaust gas reduced. In this way, a hybrid powertrain can be used to promote ammonia production and storage.

The methods described herein contemplate the production of ammonia by ammonia generation cycles that use constituents of the exhaust stream to maintain aftertreatment of NOx in an SCR device. It will be appreciated that these methods can be used separately from urea injection, with the methods described providing all of the required ammonia. Alternatively, the methods described herein can be used to supplement a urea injection system, increasing the range of the system between required filling of a urea storage tank while allowing a full range of engine and powertrain operation without significant monitoring of ammonia generation cycles and current due to available urea injection when needed storage capacity allows.

Detection of NOx is important for understanding the operation of the aftertreatment system and for controlling NOx as a component of ammonia production. A NOx sensor or an oxygen sensor increases the cost and weight of a vehicle, and such sensors often require a certain operating temperature range, reached after a certain warm-up time, in order to be functional. As described above, a virtual NOx sensor can be used to estimate the presence of NOx in an aftertreatment system. 12 12 schematically depicts an exemplary NOx model module utilized in an engine control module that determines a NOx creation estimate, in accordance with the present disclosure. The example NOx model module 500 operates in a NOx creation estimation system 510 and includes a model module 520 and a NOx estimation module 530. Engine sensor inputs x ₁ through x _n are inputs to the NOx model module and can be a number of factors including temperatures, pressures, engine control settings including valve and ignition settings, and other measurements indicative of a combustion condition in the combustion chamber. The model module 520 receives these inputs and applies known relationships to determine a number of parameters to describe combustion in the combustion chamber. Examples of these descriptive parameters include EGR%, the percentage of exhaust gas that is recirculated to the combustion chamber to control the combustion process; an air-to-fuel charge ratio (AFR) describing the mixture of air and fuel present in the combustion chamber; combustion temperature metrics including, for example, either combusted gas combustion temperature or average combustion temperature; a combustion timing metric tracking the progression of combustion during a combustion process, for example CA50, a measurement at which crank angle 50% of the mass of fuel originally present in the combustion chamber is burned; and fuel rail pressure indicative of the pressure of fuel available to fuel injectors to be injected into the combustion chamber. These descriptive parameters can be used to estimate conditions present in the combustion chamber during the combustion process. As described above, conditions present in the combustion chamber affect the production of NOx in the combustion process. These descriptive parameters may be provided to the NOx estimation module 530 where programmed calculations use the descriptive parameters as inputs to generate an estimate of NOx creation due to the combustion process. However, as described above, models that analyze a variable that describes the combustion process may involve complex calculations that take longer to calculate than is necessary to produce real-time results that require large amounts of processing capability and are only as accurate as the pre-programmed algorithm permitted. Due to these challenges and a need for accurate and timely information, estimation of NOx creation in an ECM as part of an aftertreatment control strategy is not currently preferred.

Monitoring NOx through a virtual NOx sensor may require monitoring the combustion process to accurately estimate the engine's NOx production. Additionally, accurate control of multiple injections as described in the method above can be aided by monitoring the combustion process. Various engine sensor inputs can be used to quantify the parameters that describe the combustion process. However, it is difficult to directly monitor the combustion occurring in the engine. Sensors can detect and measure fuel flow and air flow into the cylinder, a sensor can monitor a specific electrical voltage applied to a spark plug, or a processor can collect a sum of information that would predict conditions necessary to produce auto-ignition , but these combustion readings taken together are only predictive and do not measure actual combustion results. An example method that measures actual combustion results uses pressure measurements taken in the combustion chamber during a combustion process. Cylinder pressure readings provide tangible readings that describe conditions within the combustion chamber. Based on an understanding of the combustion process, cylinder pressures can be analyzed to estimate the state of the combustion process in a particular cylinder, describing the combustion in terms of both combustion phasing and combustion severity. Combustion of a known charge at known times under known conditions produces a predictable pressure in the cylinder. By describing the phase and magnitude of combustion at specific crank angles, the initiation and progression of a specific combustion process can be described as an estimated state of combustion. By estimating the state of the combustion process for a cylinder, factors affecting NOx creation during the combustion process can be determined and made available for use in estimating NOx creation.

A known method for monitoring combustion phasing is to estimate the mass fraction burn ratio for a given crank angle based on known parameters. The mass fraction burn ratio describes what percentage of the charge was burned in the combustion chamber and serves as a good estimate of combustion phasing. 13 1 graphically depicts an exemplary mass fraction burn curve, in accordance with the present disclosure. At a given crank angle, the curve depicted describes the estimated percentage of a fuel/air mixture in the charge that was burned for that combustion process. To be used as a measure of combustion phasing, it is known to establish either a particular mass fraction burn percentage of interest or a particular crank angle of interest. 13 determines CA50% as the crank angle at which the mass fraction burn is equal to 50%. By checking that particular level over multiple combustion processes in that cylinder or over a number of cylinders, the comparative phase control of the specific combustion processes can be described.

As described above, combustion phasing can be used to estimate the state of a particular combustion process. An example method of monitoring combustion phasing to diagnose ineffective combustion is disclosed, whereby combustion is monitored in an engine, mass fraction burn ratios are generated for each cylinder combustion process, and combustion phasing is compared across cylinders. If the combustion phase for a cylinder at a certain crank angle for that first cylinder differs by more than a threshold phase difference from the combustion phase for another cylinder at the same crank angle for that second cylinder, anomalous combustion can be inferred. Many sources of abnormal combustion can be diagnosed by this method. For example, if a certain condition causes pre-ignition or knocking in the combustion chamber, the cylinder pressure readings will have different values than normal combustion. In addition, fuel system injection timing errors that cause the charge to be injected at incorrect times cause anomalous cylinder pressure readings. Furthermore, if a cylinder misfires or never achieves combustion, the cylinder pressure readings will show different readings than normal combustion. Similarly, pressure waveforms may be used to diagnose other abnormal combustion conditions, such as air/fuel mixture changes, camshaft phasing changes, and maintenance failures of related components. All such combustion health diagnostics have implications for NOx and can be helpful in estimating NOx production.

Many methods of estimating mass fraction burn are known. One method examines pressure data from the combustion chamber, including analyzing the pressure rise in the chamber attributable to combustion. There are several methods to quantify an increase in pressure in a cylinder that is attributable to combustion. Pressure Ratio Management (PRM) is a technique based on the Rassweiler approach, which states that mass fraction burn can be approximated by the fractional pressure increase due to combustion. Combustion of a known charge at a known time under known conditions tends to produce a consistently predictable pressure rise in the cylinder. PRM derives a pressure ratio (PR) from the ratio of a measured cylinder pressure under combustion at a given crank angle (P _CYL (θ)) to a calculated cylinder pressure without combustion, the so-called "motored pressure", where it, when in the cylinder combustion has not occurred, estimates a pressure value at a given crank angle (P _MOT (θ)), giving the following equation. $$PR\left(\theta \right)=\frac{P_{ZYL}\left(\theta \right)}{P_{MOT}\left(\theta \right)}$$

14 graphically illustrates an example cylinder pressure during a combustion process plotted against crank angle, in accordance with the present disclosure. P _MOT (θ) has a smooth inverse parabolic peak from the piston compressing a retained gas pocket without combustion. With the piston at BDC, all valves are closed, the piston moves up compressing the gas, the piston reaches TDC at the peak of the pressure curve, and the pressure decreases as the piston falls off TDC. P _CYL (θ) represents a pressure rise above P _MOT (θ). Combustion times vary from application to application. In this particular example curve, P _CYL (θ) begins to increase around TDC from P _MOT (θ), which describes an ignition event at a time before TDC. When the charge burns, heat and work result from the combustion, leading to an increase in pressure in the combustion chamber. PR is a ratio of P _MOT to P _CYL and P _MOT is a component of P _CYL . Net Combustion Pressure (NCP(θ)) is the difference between P _CYL (θ) and P _MOT (θ), or the pressure rise in the combustion chamber attributable to combustion at a given crank angle. It is understood that by subtracting one from PR, a ratio of NCP to P _MOT can be found as follows. $$PR\left(\theta \right)-1=\frac{P_{ZYL}\left(\theta \right)}{P_{MOT}\left(\theta \right)}-\frac{P_{MOT}\left(\theta \right)}{P_{MOT}\left(\theta \right)}=\frac{NCP\left(\theta \right)}{P_{MOT}\left(\theta \right)}$$

The PR measured by the above equation can therefore be used to directly describe the amount of combustion in a cylinder. Normalizing PR minus one at crank angle θ to an expected or theoretical maximum PR value minus one gives a fractional Pressure ratio of the pressure increase due to combustion at crank angle θ to the expected total pressure increase due to combustion at the termination of the combustion process. This normalization can be expressed by the following equation. $$fPR\left(\theta \right)=\frac{PR\left(\theta \right)-1}{PR\left(90°\right)-1}\propto Massenanteilve\mathrm{right}b\mathrm{right}enn\mathrm{and}nG\left(\mathrm{\theta }\right)$$

This fractional pressure ratio describes the mass fraction burn for that particular combustion process by equating the pressure increase attributable to combustion as combustion progresses. Using PRM, pressure readings from a cylinder can be used to estimate mass fraction burn for that cylinder.

The above method utilizing PRM is applicable to wide ranges of temperature, cylinder charge and timing associated with compression ignition engines, with the added benefit of not requiring calibrated pressure sensors. Since PR is a ratio of pressures, an uncalibrated linear pressure transducer can be used to collect pressure data readings from each cylinder.

Another method of estimating mass fraction burn is to directly use the Rassweiler approach to determine mass fraction burn by calculating the total heat released at a given crank angle. The Rassweiler approach uses pressure readings from a cylinder to approximate the incremental heat release in the cylinder. This approach is given by the following equation. $$Q_{f\mathrm{right}eiGeset\mathrm{e.g}t}\left(\mathrm{\theta }\right)=\sum P_{k+1}-P_{k-1}{\left(\frac{V_{k-1}}{V_k}\right)}^{\mathrm{right}}$$

Mass fraction burn, a measure of how much of the charge is burned at a given crank angle, can be approximated by determining what fraction of heat release has occurred for a combustion process at a given crank angle. The incremental heat release determined by the Rassweiler approach can be summed over a range of crank angles, compared to the total expected or theoretical heat release for the combustion process, and used to estimate mass fraction burn. For example, if 75% of the total expected heat release was realized at a given crank angle, we can estimate that 75% of the combustion for the cycle took place at that crank angle.

Other methods can be used to estimate mass fraction burn. One method quantifies the rate of change of energy in the combustion chamber due to combustion through analysis of classical heat release metrics based on analysis of heat released and work done during combustion of the charge. Such analyzes are based on the first law of thermodynamics, which states that the net change in energy in a closed system is equal to the sum of the heat and work added to the system. Applied to a combustion chamber, the energy increase in the combustion chamber and the trapped gases is equal to the heat transferred to the walls of the chamber and the gases plus the expansive work performed by the combustion.

An example method that uses these classical heat release metrics to approximate a mass fraction burn estimate analyzes the rate of heat release from charge combustion throughout the combustion process. This rate of heat release, dQ _ch / dθ, can be integrated over a range of crank angles to describe the net energy released in the form of heat. By deductions well known in the art, this heat release can be expressed by the following equation. $$Q={\displaystyle \int \frac{\mathrm{i.e}{Q}_{cH}}{\mathrm{i.e}\theta }}=\int (\frac{g}{g-1}p\frac{\mathrm{i.e}V}{\mathrm{i.e}\theta }+\frac{1}{g-1}V\frac{\mathrm{i.e}p}{\mathrm{i.e}\theta }$$

Gamma, γ, includes a ratio of specific heats and is chosen nominally as that for air at the temperature corresponding to those used to calculate the signal bias and without EGR. Thus, nominally or initially γ = 1.365 for diesel engines and nominally γ = 1.30 for conventional gasoline engines. However, these can be based on data from the specific heats for air and stoichiometric products using an estimate of the equivalence ratio Φ and the target EGR mole fraction for the operating condition and using the relationship that [γ = 1+ (R/ c _v )], where R is the universal gas constant, and the weighted mean of air and product properties can be fitted by the following expression, $$c_v\left(T\right)=1\left(1.0-\Phi *AGR\right)*c_{vl\mathrm{and}ft}\left(T\right)+\left(\Phi *AGR\right)*c_{vst\ddot{O}cHp\mathrm{right}O\mathrm{i.e}}\left(T\right)$$

The plot is evaluated at the gas temperature corresponding to that for pressures taken for the signal bias calculation.

Whether calculated by the above method or by another method known in the art, the calculation of energy released in the combustion process at a given crank angle can be compared to an expected or theoretical total energy release for the combustion process. This comparison yields an estimate of mass fraction burn for use in describing combustion phasing.

The methods described hereinabove are easily reduced to be programmed into a microcontroller or other device for execution during ongoing operation of an internal combustion engine as follows.

Once a mass fraction burn curve is generated for a particular combustion process, the curve is useful to assess combustion phasing for that particular combustion process. Referring again to 13 a reference point is taken from which to compare mass fraction burn estimates from different combustion processes. In this particular embodiment, CA50%, representing the crank angle at which 50% of the charge is burned, is selected. Other metrics can be chosen as long as the same metric is used for each comparison.

The determination of mass fraction burn values is a procedure that is well known in the art. Although example methods for determining mass fraction burn are described above, the methods disclosed herein for using mass fraction burn values to diagnose cylinder combustion issues may be used with any method for determining mass fraction burn. Any approach to developing mass fraction burn can be used, and this disclosure is not intended to be limited to the specific methods described herein.

There are other methods of analyzing cylinder pressure signals. Methods are known for processing complex or noisy signals and reducing them to useful information. One such method involves spectral analysis by Fast Fourier Transform (FFT). FFTs reduce a periodic or repetitive signal to a sum of harmonic signals that are useful for converting the signal into components of its frequency spectrum. Once the components of the signal are determined, they can be analyzed and information taken from the signal.

Pressure readings from the pressure transducers located in or associated with the combustion cylinders contain information directly related to the combustion occurring in the combustion chamber. However, engines are very complex mechanisms and these pressure readings may include a variety of pressure fluctuations from other sources in addition to a measure P _CYL (θ). Fast Fourier Transforms (FFT) are mathematical methods well known in the art. An FFT technique, known as spectral analysis, analyzes a complex signal and separates the signal into its component parts, which can be represented as the sum of harmonics. The spectral analysis of a pressure transducer signal, represented by f(θ), can be represented as follows. $$ffT\left(f\left(\mathrm{\theta }\right)\right)=\begin{array}{l}{\text{A}}_0+\left({\text{A}}_1\text{sin}\left({\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="DE102013113647B4\_0025.png,DE102013113647B4\_0026.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_0\mathrm{\theta +}{\mathrm{\Phi }}_1\right)\right)+\left({\text{A}}_2\text{sin}\left({\mathrm{<figure-callout\; id="0"\; label="\omega "\; filenames="DE102013113647B4\_0025.png,DE102013113647B4\_0026.png"\; state="\{\{state\}\}">\omega </figure-callout>}}_0\mathrm{\theta +}{\mathrm{\Phi }}_2\right)\right)\hfill \\ +...+\left({\text{A}}_{\text{N}}\text{sin}\left(\text{<figure-callout id="0" label="N ω" filenames="DE102013113647B4\_0025.png,DE102013113647B4\_0026.png" state="{{state}}">N</figure-callout>}{\mathrm{\omega }}_{}{\mathrm{}}_0\mathrm{\theta +}{\mathrm{\Phi }}_{\text{N}}\right)\right)\hfill \end{array}$$

Each component N of the signal f(θ) represents a periodic input to the pressure in the combustion chamber, with each increasing increment comprising N signals or higher in frequency. An experimental analysis has shown that the combustion caused by the different stages of the Pressure variation caused by the combustion process P _CYL (θ) moving piston tends to be the first, lowest frequency harmonic. By isolating this first harmonic signal, P _CYL (θ) can be measured and evaluated. As is well known in the art, FFTs provide information regarding the magnitude and phase of each detected harmonic, captured as the term Φ in each harmonic of the above equation. The angle of the first harmonic or Φ ₁ is therefore the dominant term tracking combustion phasing information. By analyzing the component of the FFT output associated with P _CYL , the phasing information of that component can be quantified and compared to either expected phasing or the phasing of other cylinders. This comparison allows evaluating the measured phasing values and displaying a warning if the difference is greater than a threshold phasing difference, indicating combustion problems in that cylinder.

Signals analyzed by FFT are estimated most efficiently when the input signal is at a steady state. Transient effects of a changing input signal can create errors in the estimates made. While methods are known for compensating for the effects of transient input signals, the methods disclosed herein are best performed at either idle or stable, mid-range engine speed conditions where the effects of transients are eliminated. A known method of accomplishing the test in a reasonably stable test period is to take samples and use an algorithm in the control module to either validate or invalidate the test data taken during a stable engine operation period.

It should be noted that while the test data is preferably taken at idle or stable engine operation, information derived from these analyzes can be used through complex programmed calculations or engine models to effect more precise engine control during various ranges of engine operation. For example, if testing and analysis at idle shows that cylinder number 4 has a partially plugged injector, the fuel injection times for that cylinder could be modified during different operating regimes to compensate for the perceived problem.

Once cylinder pressure signals have been analyzed by FFT, information from the pressure signal can be used in a variety of ways to analyze the combustion process. For example, the analyzed pressure signal may be used to generate a fractional pressure ratio, as discussed in the methods above, and is used to describe percentage of mass fraction burn to describe the progression of the combustion process.

RPM oder die Motordrehzahl ist durch einen Kurbelwellendrehzahlsensor leicht messbar, wie vorstehend beschrieben ist. P_im oder Ansaugkrümmerdruck wird typischerweise mit der Motorsteuerung in Verbindung stehend gemessen und ist ein leicht verfügbarer Term. ṁ_a oder der Anteil an frischem Luftmassenstrom der in den Zylinder strömenden Ladung ist ebenfalls ein Term, der häufig in dem Luftansaugsystem des Motors gemessen wird, oder kann alternativ aus P_im, Umgebungsluftdruck und bekannten Eigenschaften des Luftansaugsystems leicht abgeleitet werden. Eine andere Variable, die den Verbrennungsprozess beschreibt und die aus Zylinderdrücken und anderen leicht verfügbaren Sensormesswerten abgeleitet werden kann, ist die Ladungsströmung in den Zylinder ṁ_c.ṁ_c kann durch die folgende Gleichung ermittelt werden. $${\dot{m}}_c=\frac{P_{im}\cdot rpm\cdot D\cdot \eta }{2R{T}_{im}}$$

D ist gleich der Verdrängung des Motors. R ist eine Gaskonstante, die aus dem Stand der Technik gut bekannt ist. Tim ist ein Temperaturmesswert von dem Ansaugkrümmer. Eine andere Variable, die den Verbrennungsprozess beschreibt und die von Zylinderdrücken sowie anderen leicht verfügbaren Sensormesswerten abgeleitet werden kann, ist AGR% bzw. der Prozentsatz von Abgas, der in den Abgasrückführungskreislauf zurückgeleitet wird. AGR% kann durch die folgende Gleichung ermittelt werden. $$\text{AGR\%}=1-\frac{{\dot{m}}_a}{{\dot{m}}_c}$$

Once a measure, such as pressure readings, are available, other descriptive parameters related to a combustion process can be calculated. Sub-models describing certain characteristics of a combustion process may be deployed using physical properties and relationships well known in the art to convert cylinder pressures and other readily available engine sensor terms into a variable that describes the combustion process. For example, volumetric efficiency, a ratio of air/fuel charge flowing into the cylinder compared to the capacity of the cylinder, can be expressed by the following equation. $${\mathrm{n}}_{\text{PU}}=\text{f}\left(\text{RPM}{\text{,P}}_{\text{in the}\text{,}{\dot{m}}_a}\right)$$

RPM or engine speed is easily measurable by a crankshaft speed sensor as described above. P _im or intake manifold pressure is typically measured in connection with engine control and is a readily available term. ṁ _a or the fresh mass air flow fraction of the charge flowing into the cylinder is also a term often measured in the engine's air intake system, or can alternatively be easily derived from P _im , ambient air pressure and known properties of the air induction system. Another variable that describes the combustion process and that can be derived from cylinder pressures and other readily available sensor readings is the charge flow into the cylinder ṁ _c .ṁ _c can be found by the following equation. $${\dot{m}}_c=\frac{P_{im}\cdot \mathrm{right}pm\cdot D\cdot n}{2R{T}_{im}}$$

D is equal to the displacement of the motor. R is a gas constant well known in the art. Tim is a temperature reading from the intake manifold. Another variable describing the combustion process that can be derived from cylinder pressures as well as other readily available sensor readings is EGR% or the percentage of exhaust gas that is fed back into the exhaust gas recirculation circuit. EGR% can be found by the following equation. $$\text{EGR\%}=1-\frac{{\dot{m}}_a}{{\dot{m}}_c}$$

Yet another variable that describes the combustion process and that can be derived from cylinder pressures and other readily available sensor readings is CAx, where x equals a desired fractional pressure ratio. CAx can be found by the following equation. $$Z=\frac{P_{ZYL}\left(\theta \right)}{P_{MOT}\left(\theta \right)}-1$$

Filling in the desired fractional pressure ratio as Z and solving for θ gives CAx. For example, CA50 can be determined as follows. $$\frac{P_{ZYL}\left(\theta \right)}{P_{MOT}\left(\theta \right)}=1.5$$

Pmax ist der maximale Druck, der in dem Brennraum durch den Verbrennungsprozess erreicht wird. PPL ist ein Maß des Kurbelwinkels, bei dem Pmax auftritt. V(PPL) ist das Volumen des Zylinders an dem Punkt, an dem Pmax auftritt. T_u, die durchschnittliche Temperatur des noch nicht verbrannten oder unverbrannten Teils der Ladung in dem Brennraum, kann durch die folgende Gleichung ermittelt werden. $$T_u=\frac{\mathrm{1,05}\ast {\dot{m}}_c}{\mathrm{1,05}\ast {\dot{m}}_c-a\cdot {\dot{m}}_{ƒ}{\lambda }_s}\left[\mathrm{0,5}\beta T_{Abg}+{\mathrm{0,95}}_{im}\right]{\left(\frac{P_{max}-\Delta P}{P_{im}}\right)}^{\frac{r-1}{r}}$$

ṁ _ƒ ist der Kraftstoffmassenstrom und kann entweder aus einem bekannten Kraftstoffverteilerrohrdruck kombiniert mit bekannten Eigenschaften und dem Betrieb der Kraftstoffeinspritzvorrichtungen oder aus ṁ_c und ṁ_a ermittelt werden, α und β sind Kalibrierungen, die auf Motordrehzahl und Last beruhen und können experimentell, empirisch, prädiktiv, durch Modellierung oder anderen Techniken entwickelt werden, die adäquat sind, um Motorbetrieb genau vorherzusagen, und es könnten eine Vielzahl von Kalibrierungskurven von dem gleichen Motor für jeden Zylinder und für unterschiedliche Motoreinstellungen, -bedingungen oder - betriebsbereiche verwendet werden. λ_s ist das stöchiometrische Luft/ Kraftstoff-Verhältnis für den bestimmten Kraftstoff und umfasst Werte, die aus dem Stand der Technik gut bekannt sind. T_Abg ist eine gemessene Abgastemperatur. T_im und P_im sind Temperatur- und Druckmesswerte, die an dem Ansaugkrümmer genommen sind. P_max - ΔP beschreibt den Druck in dem Brennraum kurz vor Beginn der Verbrennung. γ ist eine spezifische Wärmekonstante, die vorstehend beschrieben ist. T_b, die durchschnittliche Temperatur des verbrannten Teils der Ladung in dem Brennraum, kann durch die folgende Gleichung ermittelt werden. $$T_b=\frac{T_a-\left(1-x_b\right)T_u}{x_b},x_b=\frac{a\cdot {\dot{m}}_{ƒ}\left(1+{\lambda }_s\right)}{\mathrm{1,05}{\dot{m}}_c}$$

Various temperatures within the combustion chamber may also be estimated from cylinder pressures and other readily available sensor readings. 15 FIG. 12 depicts a number of different temperatures that are estimable within the combustion chamber and are useful in describing the combustion process, in accordance with the present disclosure. T _a , the average temperature in the combustion chamber, can be found by the following equation. $$T_a\frac{T_{max}\cdot V\left(PPL\right)}{1.05\ast {\dot{m}}_{c}R}$$

Pmax is the maximum pressure that is reached in the combustion chamber by the combustion process. PPL is a measure of the crank angle at which Pmax occurs. V(PPL) is the volume of the cylinder at the point where Pmax occurs. T _u , the average temperature of the unburned or unburned portion of the charge in the combustion chamber, can be found by the following equation. $$T_{\mathrm{and}}=\frac{1.05\ast {\dot{m}}_c}{1.05\ast {\dot{m}}_c-a\cdot {\dot{m}}_{ƒ}{\lambda }_s}\left[0.5\beta T_{AbG}+{0.95}_{im}\right]{\left(\frac{P_{max}-\Delta P}{P_{im}}\right)}^{\frac{\mathrm{right}-1}{\mathrm{right}}}$$

_ṁƒ is the mass fuel flow and can be determined either from a known fuel rail pressure combined with known characteristics and operation of the fuel injectors or from _ṁc and _ṁa , α and β are calibrations based on engine speed and load and can be experimental, empirical, predictive , be developed through modeling or other techniques adequate to accurately predict engine operation, and a variety of calibration curves could be used from the same engine for each cylinder and for different engine settings, conditions or operating regions. λ _s is the stoichiometric air/fuel ratio for the particular fuel and includes values well known in the art. T _exhaust is a measured exhaust gas temperature. T _im and P _im are temperature and pressure readings taken at the intake manifold. P _max - ΔP describes the pressure in the combustion chamber just before the start of combustion. γ is a specific heat constant described above. T _b , the average temperature of the burned portion of the charge in the combustion chamber, can be found by the following equation. $$T_b=\frac{T_a-\left(1-x_b\right)T_{\mathrm{and}}}{x_b},x_b=\frac{a\cdot {\dot{m}}_{ƒ}\left(1+{\lambda }_s\right)}{1.05{\dot{m}}_c}$$

Note that the above equations are simplified in a method well known in the art by neglecting the heat loss to the cylinder wall. Methods for compensating for this simplification are well known in the art and are not described in detail herein. By using the above relationships and derivatives, cylinder pressure and other readily available sensor readings can be used to determine a number of parameters that describe the combustion process being monitored.

As described above, cylinder pressure measurements may be used to describe a combustion condition occurring in the combustion chamber for use as a factor in estimating NOx creation. Also as described above, a number of other factors are relevant to accurately estimating NOx creation. 16 12 is a graphical representation of example modeled results describing standardized effects of a number of inputs on NOx emissions under a given set of conditions, in accordance with the present disclosure. As described above, methods are known that use a model module and a NOx estimation module to simulate or estimate NOx creation based on known characteristics of an engine. The model used in this particular example analysis to characterize NOx creation by a combustion process can be characterized by the following expression. $$NOx=NNT\left(Pmax,CA\mathrm{50,}CApmax,AGR\%,AfR\right)$$

As in the graphical results of 16 a number of factors have been shown to have differential effects on NOx generation. Under this particular set of conditions, EGR% has the greatest effect on NOx creation for the modeled engine. In this case, by methods well known in the art, recirculating a certain amount of exhaust gas back into the combustion chamber through the EGR loop lowers the adiabatic flame temperature of the combustion process, thereby lowering the temperatures to which nitrogen and oxygen molecules are exposed to combustion and thereby lowering the rate of NOx generation. By examining such models under various engine operating conditions, the most useful inputs can be provided to the neural network to provide accurate estimates of NOx creation. Furthermore, examining such models provides information useful for selecting input data for initial training of the neural network, for manipulating inputs and providing corresponding outputs to sensor inputs and descriptive parameters most likely to affect NOx production.

wobei gegenwärtige NOx-Messwerte und eine Ausgabe y von einem eingelernten neuronalen Netz genutzt werden, um eine Änderung der NOx-Erzeugung zu schätzen. Eine solche Änderungsvariable kann verwendet werden, um die NOx-Erzeugung inkremental zu schätzen, oder kann verwendet werden, um Schätzungen von NOx-Erzeugung zu prüfen oder zu filtern. 17 stellt schematisch ein beispielhaftes System dar, das eine Schätzung von NOx-Erzeugung erzeugt, wobei es Modelle in einem neuronalen Netz nutzt, um Schätzungen von NOx-Erzeugung zu erzeugen, und ein dynamisches Modellmodul umfasst, um Schätzungen der NOx-Erzeugung bezüglich Wirkungen von dynamischen Motor- und Fahrzeugbedingungen gemäß der vorliegenden Offenbarung auszugleichen [ engl. „to compensated“]. Das NOx-Erzeugungsschätzsystem 400 umfasst ein Modellmodul 410, ein Modul 420 für neuronales Netz und ein dynamisches Modellmodul 430. Faktoren, die sich unter den aktuellen Betriebsbedingungen am wahrscheinlichsten auf die Schätzung von NOx-Erzeugung unter dynamischen oder sich ändernden Bedingungen auswirken, können experimentell, empirisch, prädikativ, durch Modellierung oder andere Techniken ermittelt werden, die adäquat sind, um Motorbetrieb genau vorherzusagen. Eingaben bezüglich dieser Faktoren werden dem dynamischen Modellmodul 430 zusammen mit der Ausgabe von dem neuronalen Netzmodul 420 eingespeist, und die Rohausgabe von dem neuronalen Netz kann basierend auf den projizierten Wirkungen der dynamischen Bedingungen, die durch das dynamische Modellmodul 430 ermittelt werden, angepasst, gefiltert, gemittelt, entpriorisiert oder anderweitig modifiziert werden. Auf diese Weise können die Wirkungen von dynamischen Motor- oder Fahrzeugbetriebsbedingungen bei der Schätzung von NOx-Erzeugung berücksichtigt werden.Through the methods described above, estimates of NOx creation can be generated for a range of engine sensor inputs. As will be understood by those of ordinary skill in the art, equations and model predictions of engine operation often work most effectively when the engine is operating at or near steady state power. However, observations and predictions can be made regarding the effects of transient or dynamic engine operation on NOX production estimates or the accuracy thereof. An example expression describing a dynamic model or dynamic filter module is shown by the following: $$\frac{\mathrm{i.e}NOx}{\mathrm{i.e}t}=ƒ\left(NOx,y,AGR\%,AfR,Ta,RPM\right)$$

using current NOx measurements and an output y from a trained neural network to estimate a change in NOx production. Such a changing variable may be used to incrementally estimate NOx creation, or may be used to test or filter estimates of NOx creation. 17 FIG. 12 schematically illustrates an example system that generates an estimate of NOx creation using models in a neural network to generate estimates of NOx creation and including a dynamic model module to generate estimates of NOx creation relative to effects of dynamic Balancing engine and vehicle conditions in accordance with the present disclosure. "to be compensated"]. The NOx creation estimation system 400 includes a model module 410, a neural network module 420, and a dynamic model module 430. Factors most likely to affect the estimation of NOx creation under dynamic or changing conditions under the current operating conditions can be experimental, be determined empirically, predictively, through modeling, or other techniques adequate to accurately predict engine operation. Inputs regarding these factors are fed to the dynamic model module 430 along with the output from the neural network module 420, and the raw output from the neural network can be adjusted, filtered, based on the projected effects of the dynamic conditions determined by the dynamic model module 430, averaged, deprioritized, or otherwise modified. In this way, the effects of dynamic engine or vehicle operating conditions can be taken into account when estimating NOx creation.

As described above, the integration can be used as a low-pass filter in the comparison of an actual conversion efficiency with a malfunction conversion efficiency. The data generated can often be chopped up with a number of spikes. The interpretation of the different signals, especially a comparison of the different predicted NOx values at a given point in time, is prone to misinterpretation or incorrect determinations. A comparison of the data curves generated by integration is greatly simplified and the potential for misinterpretation or incorrect determinations in a comparison is greatly reduced.

Determining conversion efficiency may be helpful to operate ammonia production cycles, for example to predict a required timing and duration of ammonia production cycles required to efficiently operate the SCR. Conversion efficiency is described as the efficiency with which an aftertreatment device can convert NOx into other molecules. The example aftertreatment system described above describes a measured or estimated NOx content of the exhaust stream measured upstream of the aftertreatment device being analyzed. This level of NOx entering the aftertreatment system at any time 't' can be described as x(t). The example aftertreatment system described above describes a measured or estimated NOx content of the exhaust stream measured downstream of the aftertreatment device being analyzed. This level of NOx exiting the aftertreatment system at any given time can be described as y(t). The conversion efficiency at any given time by the following equation. $${\mathrm{n}}_{\text{ACT}\ddot{\text{A}}\text{CHLICH}\left(\text{t}\right)}=1-\frac{\text{y}\left(\text{t}\right)}{\text{x}\left(\text{t}\right)}$$

It is understood that this equation provides the conversion efficiency at any point in time. Such instantaneous measurements or calculations are prone to error based on signal noise. Methods for applying a low-pass filter are known from the prior art. An integration of x(t) or y(t) gives a description of an amount of actual NOx entering or exiting the aftertreatment system during a period of time, respectively. An exemplary equation for determining an integrated conversion efficiency where anomalous measurements in x(t) and y(t) are filtered can be described as follows. $${\mathrm{n}}_{\text{ACT}\ddot{\text{A}}\text{CHLICH}\left(\text{t}\right)}=\frac{\int y\left(t\right)*\mathrm{i.e}t}{\int x\left(t\right)*\mathrm{i.e}t}$$

In this manner, measured or estimated values of NOx entering or exiting the aftertreatment system may be used to determine an estimated or calculated actual conversion efficiency of the aftertreatment system.

A properly operating or fresh aftertreatment device will operate at some maximum achievable conversion efficiency for a given set of conditions. However, it is understood that aftertreatment devices, particularly devices utilizing a catalyst, are subject to degraded performance over time and particularly when exposed to high temperatures. Determining a malfunctioning catalyst is desirable in maintaining low NOx emissions and continuing to enable fuel efficient engine operating modes.

Conversion efficiency in a fresh SCR device is affected by a number of environmental or operational factors. The conversion efficiency for an example SCR can be determined by a model expressed by the following function. $$\mathrm{n}=f\left(T_{BETT},SV,{\mathrm{<figure-callout\; id="3"\; label="\theta \; N\; H"\; filenames="DE102013113647B4\_0027.png,DE102013113647B4\_0030.png"\; state="\{\{state\}\}">\theta </figure-callout>}}_{NH}x\left(t\right),V_{HARNSTOff},{\rho }_{ZELLE}\right)$$

T _BED describes the temperature of the catalyst bed in the SCR. This temperature can be measured directly or can be estimated based on the temperature, flow rate and other properties of the exhaust stream. SV describes the superficial velocity of the exhaust gas flowing through the SCR device and can be determined as a function of properties of the exhaust gas flow, including temperature and flow rate. _θNH3 describes an amount of ammonia storage on the catalyst bed and is an adequate presence of ammonia is required at the SCR to achieve the desired NOx conversion response. _θNH3 can be estimated, for example, by analyzing absorption and desorption rates of ammonia, NOx conversion rates, and oxidation rates of adsorbed ammonia. As described above, x(t) describes the presence of NOx in the exhaust stream entering the aftertreatment system. Low NOx levels are easily responded to in a properly functioning SCR, while NOx levels above a certain threshold are more difficult to respond and correspond to lower conversion efficiencies. An example of a factor that limits treatment of NOx above certain levels includes limited ammonia present in an SCR. V UREA describes the volume of injected urea. While V UREA describes a presence of ammonia analogous to θ _NH3 , V UREA includes a present level of urea that is injected and may better describe a transient indicator of ammonia expected to be present in the near future. ρ _CELL describes the density of catalyst material in the SCR and therefore describes an ability of the SCR to catalyze the desired reaction.

The above model describing conversion efficiency includes factors that can be assumed or validated in normal operation of an SCR. This allows the model to be simplified, thereby reducing a processing load required to analyze conversion efficiency by the model. For example, VUREA can be monitored by operation of the urea dosing module, and given VUREA values within a certain target range, the resulting conversion efficiency calculations should be unaffected. In some embodiments, V UREA is controlled to be substantially directly proportional to x(t). Further, in some embodiments, θ _NH3 may be estimated based on V UREA, monitored exhaust flow and SCR characteristics such as temperature, and x(t). With _θNH3 values in a normal range, _θNH3 can be reduced to a part of the functional model that depends on T _BED . A value for x(t) as described above may be monitored by an upstream NOx sensor or a virtual NOx sensor. ρ _CELL is a property of the SCR device and is a known value. Given these known or estimable factors, the conversion efficiency for an example SCR can be determined by a model expressed by the following function. $$\mathrm{n}=f\left(T_{BETT,}SV,{\theta }_{NH3}\right)$$

In this way, the conversion efficiency of the SCR can be precisely determined as an on-board diagnostic function by keeping other factors within known or calibrated ranges.

As described above, an exhaust flow containing a mixture of molecular hydrogen and NOx may be used to generate ammonia through an ammonia generation catalyst. Exemplary embodiments are directed to initiating an ammonia production cycle after depleting oxygen from an ammonia production catalyst. As previously mentioned, a TWC can be used to contain an ammonia generation catalyst. Depending on the configuration of the engine, one or more TWC's can be used based on the number of exhaust manifolds. For example, a "V" configuration using two exhaust manifolds may use two TWC's, one for each exhaust manifold. Likewise, an "in-line" configuration that has a single exhaust manifold to channel exhaust from the engine into the exhaust aftertreatment system may use a TWC.

Referring again to the 3 and 4 In order to produce ammonia in the TWC, the component substances for the reaction used to produce ammonia must be present in the TWC, and the use of different cylinders to optimally produce molecular hydrogen and NOx must be in the same Feed ammonia production catalyst. Accordingly, a portion of the plurality of cylinders may be operated at an air/fuel ratio in a first range from stoichiometric to rich conducive to the production of NOx, and a remaining portion of the plurality of cylinders may be operated at an air/fuel -Ratio in a second region having a substantially richer air/fuel ratio than the first region conducive to the production of hydrogen, the produced NOx and the produced hydrogen must be fed into the same ammonia production catalyst in order to effectively convert ammonia to manufacture. For example, in the "V" configuration, multiple cylinders coordinated to produce NOx at the first range from stoichiometric to rich and the remaining cylinders coordinated to produce hydrogen at the second range, which is an air/ Fuel ratio that is richer than the first region, feed into the same catalyst to effectively generate ammonia.

Further, cylinders may be controlled in pairs to promote hydrogen and NOx production, with one cylinder controlled at the air/fuel ratio in the first stoichiometric range to produce NOx and the other cylinder at the air/fuel -ratio is controlled in the richer second region to produce hydrogen as described by example Equations 1, 2 and 3. In one example, in a "V-six" configuration where three cylinders feed into a single TWC catalyst in an ammonia generation catalyst, one cylinder may be operated with an air/fuel ratio from stoichiometric to rich that is optimized to to generate a desired amount of NOx. The remaining two cylinders can be optimized to each produce one-half the desired amount of hydrogen, with each of the remaining two cylinders operating at an air/fuel ratio richer than the air/fuel ratio of one cylinder. By splitting the hydrogen production requirement between two cylinders, it becomes associated with the 3 and 4 It can be seen that each of the cylinders operating in the richer second region can be operated at an air/fuel ratio less rich than would be the case for a single cylinder to produce the required amount of hydrogen. Additionally, note that cylinders operating at a higher air/fuel ratio (ie, first range from stoichiometric to rich) and cylinders operating at a lower air/fuel ratio (ie, second range) are preferred are chosen to balance a resultant work production in the engine and need not be static, whereby cylinders operating at a particular air/fuel ratio can alternate from combustion cycle to combustion cycle as long as the desired mixture of substances present in of the exhaust gas flow is maintained. The selection of cylinder-to-cylinder operation and injection schedules used to generate the required substances may be developed experimentally, empirically, predictively, through modeling, or through other techniques appropriate to engine operation and the resulting composition of exhaust gas flow, and a variety of injection schedules can be used by the same engine for different engine settings, conditions or operating ranges.

Referring back to 10 For example, the exhaust gas flow paths 622 and 624 fed through the pair of cylinders and feeding into the ammonia generating catalyst 630 can be modulated to contain hydrogen and NOx at different levels by modulating the air/fuel ratio in associated cylinders in the engine 610. Likewise, exhaust flow paths 626 and 628 are fed by a like pair of cylinders. By modulating air/fuel ratios with the various cylinders of engine 610, increased levels of hydrogen and NOx can be generated and delivered to catalytic converters 630 and 632. In an exemplary embodiment, exhaust gas flow paths 622 and 628 are shown with the associated cylinders operating at air/fuel ratios in a first range from stoichiometric to rich that is conducive to NOx production. For example, the first range may include lambda values from 0.96 to 1.00, where exhaust flow paths 622 and 628 may have the same air/fuel ratio within the first range, or each of the paths 622 and 628 may have a different air/fuel have a ratio within the first range. Exhaust gas flow paths 624 and 626 are also shown with the associated cylinders operating at air/fuel ratios in a second region having a richer air/fuel ratio than the first region conducive to the production of hydrogen. For example, the second range may include lambda values from 0.90 to 0.95, where exhaust flow paths 624 and 626 may have the same air/fuel ratio in the second range, or each of paths 622 and 628 may have a different air/fuel -Have ratio in the second region.

Claims (10)

A method for controlling a powertrain that includes an internal combustion engine (10, 300, 600) having multiple cylinders (20) and an exhaust aftertreatment system (15) that includes a device (220) for selective catalytic reduction that uses ammonia as a reductant, wherein the method comprises: depleting oxygen from an ammonia production catalyst (630, 632) disposed between the engine (10) and the selective catalytic reduction device (220) and connected to the plurality of cylinders (20), comprising that for the cylinders (20) are selected to have an air/fuel ratio with an operating range from stoichiometric to rich; after depleting oxygen from the ammonia production catalyst (630, 632), initiating an ammonia production cycle, comprising: operating a plurality of the cylinders (20) cooperatively, a proportion of the plurality of cylinders (20) operating at air/fuel ratios in a first range from stoichiometric to rich conducive to producing NOx, and wherein a remaining proportion of the plurality of cylinders (20) at air/fuel ratios in a second range that having a richer air/fuel ratio than the first range, conducive to producing molecular hydrogen; and using the ammonia production catalyst (630, 632) to produce ammonia. procedure after claim 1 wherein the proportion of the plurality of cylinders (20) operating at air/fuel ratios in the first range from stoichiometric to rich conducive to producing NOx comprises each cylinder (20) of the proportion at the same Air/fuel ratio works. procedure after claim 1 wherein the proportion of the plurality of cylinders (20) operating at air/fuel ratios in the first range from stoichiometric to rich conducive to producing NOx comprises at least two cylinders (20) of the proportion at different air/fuel ratios. procedure after claim 1 wherein the remaining portion of the plurality of cylinders (20) operating at air/fuel ratios in the second region having a richer air/fuel ratio than the first region conducive to producing molecular hydrogen generate comprises each cylinder (20) of the remaining portion operating at the same air/fuel ratio. procedure after claim 1 wherein the remaining portion of the plurality of cylinders (20) operating at air/fuel ratios in the second region having a richer air/fuel ratio than the first region conducive to producing molecular hydrogen generate comprises at least two cylinders (20) of the remaining portion operating at different air/fuel ratios. procedure after claim 1 wherein the cylinders (20) operating at air/fuel ratios in the first range from stoichiometric to rich conducive to producing NOx and the cylinders (20) operating at air/fuel ratios in the work second area conducive to produce molecular hydrogen can switch from combustion cycle to combustion cycle. procedure after claim 1 wherein the cylinders (20) operating at air/fuel ratios in the second range conducive to producing molecular hydrogen are operated with a split fuel injection strategy. procedure after claim 7 , wherein the split fuel injection strategy includes late combustion hydrocarbon reforming. procedure after claim 7 , wherein the split fuel injection strategy includes post-combustion hydrocarbon reforming. Apparatus for controlling a powertrain comprising an internal combustion engine (10) having multiple cylinders (20) and an exhaust aftertreatment system (15), comprising: a direct injection fuel injection system; the exhaust aftertreatment system (15) comprising: a selective catalytic reduction device (220) using ammonia as a reductant, and a first ammonia production catalyst (630, 632) disposed between the engine (10) and the selective catalytic reduction device is; and a controller configured to: monitor ammonia production requirements for the selective catalytic reduction device, impoverish oxygen from the first ammonia production catalyst (630, 632), comprising, for a first pair of cylinders (20), an air/fuel ratio is selected in an operating range from stoichiometric to rich, and after depleting oxygen from the first ammonia generating catalyst (630, 632), controlling the direct injection fuel injection system comprising that different air/fuel ratios in the first pair of cylinders (20) being effected comprising: operating one of the first pair of cylinders (20) at an air/fuel ratio in a first range from stoichiometric to rich conducive to producing NOx based on the ammonia generation requirements, and operating the other of the first pair of cylinders (20) at an air/fuel ratio in a second range having a richer air/fuel ratio than the first range and conducive to producing molecular hydrogen , based on the ammonia production requirements.

Images