SOFTWARE, METHODS AND SYSTEMS INCLUDING SOOT LOADING METRICS

PRIORITY

The benefits and rights of priority of U.S. Patent Application No. 60/876,881 filed December 22, 2006 are claimed, and that application is incorporated by reference.

BACKGROUND

Internal combustion engines including diesel engines produce a number of combustion products including particulates, hydrocarbons ("HC"), carbon monoxide ("CO"), oxides of nitrogen ("NOx"), and oxides of sulfur ("SOx"). Aftertreatment systems may be utilized to reduce or eliminate emissions of these and other combustion products. Diesel particulate filters, such as catalyzed soot filters and others, can be used to trap diesel particulate matter and reduce emissions. Diesel particulate filters may undergo soot regeneration or desoot to eliminate trapped diesel particulate matter. There is a need for metrics operable to determine or estimate soot loading of diesel particulate filters or soot filters.

SUMMARY

One embodiment is a method including a unique soot loading metric. Further embodiments, forms, objects, features, advantages, aspects, and benefits shall become apparent from the following description and drawings.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 is a schematic of an integrated engine-exhaust aftertreatment system provided in a vehicle.

Fig. 2 is a schematic of an integrated engine-exhaust aftertreatment system operatively coupled with an engine control unit.

Fig. 3 is a graph of loss coefficient versus Reynolds number.

Fig. 4 is a graph of viscosity versus temperature.

Fig. 5 is a graph of soot load metric versus soot load.

Fig. 6 is a graph of soot load metric versus soot load.

Fig. 7 is a graph of soot load metric versus run point.

Fig. 8 is a graph of soot load metric versus soot load.

Fig. 9 is a graph of predicted soot load versus measured soot load.

Fig. 10 is a graph of soot load metric versus run point.

Fig. 1 1 is a graph of soot load metric versus time.

Fig. 12 is a graph of soot load metric versus time.

Fig. 13 is a graph of engine speed and final fueling versus time.

Fig. 14 is a graph of counter and soot load metric versus time.

DETAILED DESCRIPTION

For the purposes of promoting an understanding of the principles of the invention, reference will now be made to the embodiments illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention as thereby intended, such alterations and further modifications in the illustrated embodiments, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.

With reference to Fig. 1 , there is illustrated a schematic of a preferred integrated engine-exhaust aftertreatment system 10 provided in a vehicle 7. Aftertreatment subsystem 14 includes a diesel oxidation catalyst 16 which is preferably a close coupled catalyst but could be other types of catalyst units such as a semi-close coupled catalyst, a NOx adsorber or lean NOx trap 18, and a diesel particulate filter 20 which are coupled in flow series to receive and treat exhaust output from engine 12.

Diesel oxidation catalyst unit 16 is preferably a flow through device that includes a honey-comb like substrate. The substrate has a surface area that includes a catalyst. As exhaust gas from the engine 12 traverses the catalyst, CO, gaseous HC and liquid HC (unburned fuel and oil) are oxidized. As a result, these pollutants are converted to carbon dioxide and water. During operation, the diesel oxidation catalyst unit 16 is heated to a desired temperature.

NOx adsorber 18 is operable to adsorb NOx and SOx emitted from engine 12 to reduce their emission into the atmosphere. NOx adsorber 18 preferably includes catalyst

sites which catalyze oxidation reactions and storage sites which store compounds. After NOx adsorber 18 reaches a certain storage capacity it is regenerated through deNOx and/or deSOx processes. Other embodiments contemplate use of different NOx aftertreatment devices, for example, a converter such as a saline NOx catalyst.

Diesel particulate filter or soot filter 20 is preferably a catalyzed soot filter, but may include one or more of several types of filters. Diesel particulate filter 20 is utilized to capture unwanted diesel particulate matter from the flow of exhaust gas exiting engine 12. Diesel particulate matter includes sub-micron size particles found in diesel exhaust, including both solid and liquid particles, and may be classified into several fractions including: inorganic carbon (soot), organic fraction (sometimes referred to as SOF or VOF), and sulfate fraction (sometimes referred to as hydrated sulfuric acid). The regeneration of diesel particulate filter 20 is referred to as desoot or soot filter regeneration and may include oxidation of some or all of the trapped fractions of diesel particulate matter. Diesel particulate filter 20 preferably includes at least one catalyst to catalyze the oxidation of trapped particulate.

With reference to Fig. 2, there is illustrated a schematic of integrated engine- exhaust aftertreatment system 10 operatively coupled with an engine control unit ("ECU") 28. At least one temperature sensor 60 is connected with diesel oxidation catalyst unit 16 for measuring the temperature of the exhaust gas as it enters diesel oxidation catalyst unit 16. In other embodiments, two temperature sensors 60 are used, one at the entrance or upstream from diesel oxidation catalyst unit 16, and another at the exit or downstream from diesel oxidation catalyst unit 16. Information from temperature sensor(s) 60 is

provided to ECU 28 and used to calculate the temperature of diesel oxidation catalyst unit 16.

A first NOx temperature sensor 62 senses the temperature of flow entering or upstream of NOx adsorber 18 and provides a signal to ECU 28. A second NOx temperature sensor 64 senses the temperature of flow exiting or downstream of NOx adsorber 18 and provides a signal to ECU 28. NOx temperature sensors 62 and 64 are used to monitor the temperature of the flow of gas entering and exiting NOx adsorber 18 and provide signals that are indicative of the temperature of the flow of exhaust gas to ECU 28. An algorithm may then be used by ECU 28 to determine the operating temperature of the NOx adsorber 18.

A first oxygen sensor 66 is positioned in fluid communication with the flow of exhaust gas entering or upstream from NOx adsorber 18 and a second oxygen sensor 68 is positioned in fluid communication with the flow of exhaust gas exiting or downstream of NOx adsorber 18. Oxygen sensors are preferably universal exhaust gas oxygen sensors or lambda sensors, but could be any type of oxygen sensor. Oxygen sensors 66 and 68 are connected with ECU 28 and generate electric signals that are indicative of the amount of oxygen contained in the flow of exhaust gas. Oxygen sensors 66 and 68 allow ECU 28 to accurately monitor air-fuel ratios ("AFR") also over a wide range thereby allowing ECU 28 to determine a lambda value associated with the exhaust gas entering and exiting NOx adsorber 18.

Engine 12 includes a fuel injection system 90 that is operatively coupled to, and controlled by, ECU 28. Fuel injection system 90 delivers fuel into the cylinders of engine 12. Various types of fuel injection systems may be utilized in the present invention,

including, but not limited to, pump-line-nozzle injection systems, unit injector and unit pump systems, common rail fuel injection systems and others. The timing of the fuel injection, the amount of fuel injected, the number and timing of injection pulses, are preferably controlled by fuel injection system 90 and/or ECU 28.

Sensor 72 is a pressure differential sensor arrangement which is operable to sense a pressure differential, preferably a pressure differential across diesel particulate filter 20, and provide pressure differential information to ECU 28. Sensor 74 is a temperature sensor arrangement which is operable to sense a temperature of diesel particulate filter 20 and provide temperature information to ECU 28. ECU 28 can also receive temperature information from bed model virtual sensor 80. ECU 28 can receive ambient pressure information from sensor 80. ECU 28 can receive information about fuel flow rate from fuel flow rate sensor 82 which can be a physical or virtual sensor. ECU 28 can receive information about fresh air flow rate from fresh air flow sensor 84, which can be a mass flow rate sensor which is operatively coupled with a fresh air flow passage or a virtual sensor. In certain embodiments, some or all of the foregoing sensors are virtual sensors. In other embodiments some or all of the foregoing sensors are physical sensors. In further embodiments, a combination of virtual and physical sensors is used.

The soot load of a diesel particulate filter or soot-filter such as diesel particulate filter 20 can be correlated to the pressure drop across the soot filter divided by the volumetric flow through the filter VP IQ . This metric for soot loading applies if the pressure loss characteristics correspond to that of flow through a permeable media where the soot layer and filter porosity can be envisioned as being made up of a series of capillary

tubes. Under such conditions VP / g can be determined or calculated according to Equation 1 :

VP _ ILε

— = μ ^~ 7^ (Equation 1)

where

VP = soot filter differential pressure

Q = volumetric flow μ = absolute viscosity

A = cross-sectional area of filter ε= void fraction or porosity

R _h = hydraulic radius of passage

L = length of passage

The geometric parameters void fraction, passage radius and length, can be combined with viscosity if it is assumed constant, into a single value that is a function of the soot load as described by Equation 2:

— = K _S ' (Equation 2)

where K _s ' is a value that correlates to the soot load.

Using VP / Q as a soot-loading metric assumes that pressure loss between pressure taps is entirely from laminar wall friction, viscosity is constant, flow is adiabatic (no heat transfer or heat source), and density is constant.

Preferred embodiments include methods, systems, and software which include an improved soot loading metric. In a preferred embodiment, the improved metric can include the effects of the turbulent pressure loss mechanism and fluid viscosity. The relationship between the soot-filter pressure loss coefficient, C _L , to Reynolds number, Re, may be

experimentally determined. The loss coefficient and Reynolds number are defined according to Equation 3 and Equation 4, respectively;

C - ^V/> (Equation 3) pV

Re = -t— (Equation 4) μ where,

V = Average velocity in reference cross-section p = Fluid density

D = Diameter of reference cross-section

For many flow paths the power-law relationship given by Equation 5 applies:

Q

C _L = —ς, (Equation 5)

Ke where C and b are constants.

The constant b will equal one if the pressure loss is entirely from laminar wall friction and zero if the pressure loss is entirely from turbulence. This value of the constant b can be determined experimentally.

With reference to Fig. 3, there is illustrated a graph of loss coefficient as a function of Reynolds number showing the loss characteristics of a soot filter. A full engine map was run in regeneration mode on a clean filter. The data of this map are the points which fit the line indicated with the arrow and label "deSoot mode, clean filter." From this line, the value of the exponent coefficient b was empirically determined to be 0.763. Less extensive maps, the smaller data sets of Fig. 3, were run with soot loaded filters which were loaded with various soot loads. This shortened the run time and prevented the soot load from

changing significantly during the map run time. The data of from these maps are the points which fit the group of lines indicates with the bracket and label "various soot loads (color coded)". These groups gave greater accuracy of the metric when the exemplary filter was clean.

The inventors determined that the pressure loss follows the power law equation. The exponent coefficient b is 0.763 which indicates that the primary pressure loss is by laminar wall friction, but there is also a significant turbulence pressure loss mechanism. The b exponent was established using the clean filter data. For each loaded condition data set, the C factor was fit using least squares and assuming the b coefficient from the regeneration data applied.

As shown in Fig. 3, the coefficient C can be used to estimate the soot loading state of the filter. In a preferred embodiment, the improved soot-loading metric can be determined as follows. Solve for C in Equation 5 as se forth in Equation 6:

C = C ₁ Re* (Equation 6)

Substitute the definitions of the loss coefficient, Equation 3, and Reynolds number,

Equation 4, to arrive at Equation 7:

(Equation 7)

From the continuity equation, Equation 8 follows:

V = - , (Equation 8) pA where

m - Mass flow rate

A - Cross-section of reference diameter.

Substitute Equation 8 into 7 yields Equation 9:

C = 2D ^b A ² - ^b \ ^P . ^P , (Equation 9) m μ

Density can be computed from the gas temperature and pressure using the ideal gas law as set forth in Equation 10:

p

P = -± , (Equation 10)

Kl where,

P _s - Static pressure at reference cross-section of a filter T - Temperature at reference cross-section of a filter R - Ideal gas constant

Substituting Equation 10 into Equation 9 yields Equation 1 1 :

2D ^b A ^2~b VPP

C = Tjd-J (Equation 1 1)

R m 2-b Tμ"

For an ideal gas, viscosity is a function of temperature alone. Fig. 4 shows the viscosity of air as a function of temperature. Note that viscosity can change by almost a factor of 2 over the temperature range of diesel exhaust and can be accounted for. Equation 12 relates a burnt air/fuel mixture to that of air:

M _nrod = . (Equation 12) ^μprod

where φ is the equivalence ratio. Diesel engines run at equivalence ratios less than 1, so the exhaust gas viscosity will not differ from that of air by more that 2.7%. This difference can be neglected in the improved soot-load metric.

Fig. 4 shows that the viscosity is related to temperature by the power law given by Equation 13:

(Equation 13)

Finally, substituting Equation 13 into Equation 1 1 yields the final form for the C coefficient given by Equation 14:

(Equation 14)

To arrive at the improved soot load metric, M, the first fractional term in Equation 14 can be omitted since it is a constant and has no effect on changes due to soot load. The improved soot load metric is thus defined according to Equation 15:

(Equation 15)

Table 1 below shows the units for the inputs and engine sensors according to a preferred embodiment.

Table 1

The sensors listed in Table 1 can be the sensors described above in connection with Fig. 2, for example, or other sensors operable to provide appropriate information. The calculations to solve equation 15 for M can be performed by ECU 28 described above in connection with Fig. 2, for example, or can be performed by another controller, processor, or other device. In order to better approximate the gas properties in the soot filter, the average of the inlet and outlet pressure (assuming negligible exhaust system losses downstream of filter) and bed temperature estimate can be used.

The improved metric has been validated using a test cell. Several test-cell data sets have been taken to gain confidence in using the improved metric as a soot-load predictor in both base-engine run mode and in regeneration mode.

Soot-Load Metric Behavior with Soot Load Mass

Fig. 5 shows an exemplary improved soot loading metric as a function of soot-load mass. The metric varies linearly from approximately 0.3 clean to approximately 0.55 fully loaded at 55 grams. Each data group was collected at the run modes shown in the smaller plot. For each data group the soot loading mass was measured at the beginning and end of the map. The soot mass of the intermediate points was assumed to vary linearly from the

beginning of the map to the end. The arrows indicate whether the soot load increased or decreased from beginning to end. Note that at high soot loadings the soot load decreased whereas the soot load increased when the map was started at a low soot load. The engine torque was limited for this map to keep the temperature low enough to prevent cleaning the filter before the map completed. This plot shows a highly linear correlation of the soot load to an exemplary improved metric.

Fig. 6 shows similar data as Fig. 5 collected on four additional particulate filter units to indicate the part-to-part variability of the metric value to soot load. The part-to- part variability is much smaller than the change due to the soot load.

Clean Filter Metric Variation

Fig. 7 shows an exemplary metric values of a clean filter in regeneration mode. The engine run map extensively covers the engine operating range as shown in small inserted plot. The exemplary metric variation over this map is small compared to the range of the metric value corresponding to a clean and dirty filter. A number of comparisons of the improved metric and the VPIQ Metric were conducted.

Loaded Filter Metric Comparison

Fig. 8 shows the correlation of an exemplary improved-metric to soot-load compared to the correlation of a VPIQ metric to soot load. The same measurement data were used to calculate both metrics. A correlation line has been fit through each of the run map end points since these are the points where the soot mass was actually measured. For the intermediate points the soot load was assumed to follow a linear line from the starting mass to the ending mass. Note that the VPIQ intermediate points fall further away from this correlation line than do the improved-metric points.

Using the correlation line of Fig. 8, the intermediate point soot load can be predicted for each metric value. Fig. 9 shows these results. It is expected that predicted soot load should fall between the starting and ending soot load values. The boxes in Fig. 9 show the expected soot load prediction ranges. The VP IQ points fall much further outside the expected range than the improved-metric predictions. These results indicate that the VPIQ metric is more affected by the change in flow field properties than is an exemplary improved metric.

Clean Filter Metric Comparison

Fig. 10 presents an exemplary improved metric and the WPIQ metric during a regeneration event on a clean filter for a map that covers an entire engine operating region (see Fig. 7 for run modes). The VPIQ metric value changes extensively over this map due to the changing flow field properties associated with different run modes. The change in the improved metric is much smaller when compared to the range in metric value from a clean to dirty filter. Therefore the improved metric is capable of detecting a clean filter. Conversely, there would be a great deal of inaccuracy using dP/Q to determine soot loading if a filter was clean.

Behavior of Metrics During a Regeneration Event

Fig. 11 shows time traces of an exemplary improved metric during a regeneration event for five different run modes. The filters were loaded to 60 grams. The engine was run briefly in base run mode and then regeneration was triggered. Note that the beginning metric value was the same for all the run modes except for the 1500@l 50 mode. For the 1500@150 test, the filter was loaded one day, the engine sat overnight, and then the test was conducted. For all the other tests the filter was loaded and the test was run

immediately after. The difference in metric value could be due to changes in the soot consistency as it sat overnight. All of the traces collapse to the same metric value as the regeneration event nears completion. A jump in the traces occurs in the metric value near the end of regeneration. The cause could be due to: change from heat source, dosing, to no heat source, change in physical dimensions of the filter flow paths as the temperature of the gas and filter element decreases, filling of the filter pores with soot.

Fig. 12 shows the VP IQ metric calculated from the same test data as Fig. 11. The starting metric values are more varied, and the traces do not collapse near the end of the regeneration event, another indication that the VP I Q metric is affected more by the flow field parameter change with run mode.

City Stop-and-Go Cycle Behavior of Improved Metric

Fig. 13 presents the speed and torque time traces of a test cell simulated engine stop- and-go route. Fig. 14 presents an exemplary improved metric value for this route. The blue points indicate a believable metric value, where the volumetric flow rate is above the threshold of 0.15 m ³ /sec. The exemplary improved metric responds accurately over the rapid changing speed and load. The metric value monotonically increases during the portion of the cycle outside of regeneration and decreases monotonically during the regeneration mode.

Various embodiments contemplated use of the improved metric to initiate, request and/or command desoot operations both alone or in combination with other metrics. Certain embodiments contemplate use of the improved metric to initiate regeneration based upon an experimentally determine the minimum metric value that corresponds to a fully- loaded mal-distributed filter. Various embodiments contemplate use of the improved

metric and an algorithm that corrects for zero drift when the engine is off. Certain preferred embodiments contemplate commanding, initiating, or requesting, a soot filter regeneration event can be initiated when the metric is above a threshold. Certain preferred embodiments contemplate command an end to or terminating a desoot or regeneration event when the metric is below a threshold. Certain preferred embodiments contemplate commanding, initiating, or requesting, a soot filter regeneration event can be initiated when the metric is above a threshold and terminated when the metric is below a threshold. Another preferred embodiment contemplates using a soot pressure differential, an exhaust flow rate, and exhaust flow temperature and pressure, to determine or calculate the value of a metric to provides a an indication of actual soot load in the filter. Additional embodiments contemplate using that metric value can then be to initiate and/or terminate a regeneration event and/or to prevent overloading the soot filter and/or to save fuel used for regeneration.

While the invention has been illustrated and described in detail in the drawings and foregoing description, the same is to be considered as illustrative and not restrictive in character, it being understood that only the preferred embodiments have been shown and described and that all changes and modifications that come within the spirit of the inventions are desired to be protected. It should be understood that while the use of words such as preferable, preferably, preferred or more preferred utilized in the description above indicate that the feature so described may be more desirable, it nonetheless may not be necessary and embodiments lacking the same may be contemplated as within the scope of the invention, the scope being defined by the claims that follow. In reading the claims, it is intended that when words such as "a," "an," "at least one," or "at least one portion" are

used there is no intention to limit the claim to only one item unless specifically stated to the contrary in the claim. When the language "at least a portion" and/or "a portion" is used the item can include a portion and/or the entire item unless specifically stated to the contrary. While equations, theory, and experimental results and validations have been presented to aid in an understanding of the principles of the invention they are not to be considered restrictive unless provided to the contrary.