The present invention relates to exhaust emission control systems for internal combustion engines; more particularly, to methods for regenerating a particulates filter for exhaust gas in an engine exhaust system; and most particularly, to a method for optimizing timing of such regeneration and for controlling temperature in a particulates filter during regeneration thereof to prevent thermal damage to the filter.
Internal combustion engine exhaust emissions, and especially diesel engine exhaust emissions, have recently come under scrutiny with the advent of stricter regulations, both in the U.S. and abroad. While diesel engines are known to be more economical to run than spark-ignited engines, diesel engines inherently suffer disadvantages in the area of emissions. For example, in a diesel engine, fuel is injected during the compression stroke, as opposed to during the intake stroke in a spark-ignited engine. As a result, a diesel engine has less time to thoroughly mix the air and fuel before ignition occurs. The consequence is that diesel engine exhaust typically contains incompletely burned fuel known as particulate matter, or “soot”.
It must be noted that other types of internal combustion engine ignition processes are also known to produce soot in the exhaust, for example, direct injection gasoline engines. Hence, the problem addressed by the present invention is broader than just diesel exhaust soot, although that is the largest application for the present invention at the present time. For this reason, the terms “catalytic diesel particulate filter (CDPF)” and “diesel particulate filter (DPF)” as used herein should not be limited to diesel engines but rather must be taken to mean a particulate filter for capturing soot particles in any internal combustion engine exhaust.
It is known to use catalytic particulate filters which physically trap soot particulates. However, such particulate filters progressively load up with accumulated soot and therefore must be repeatedly regenerated by burning off the trapped particulates, typically on a fixed schedule and by fuel and oxygen enrichment of the exhaust stream entering the CDPF and catalytically ignited in an integral diesel oxygen catalyst (DOC).
Typically, prior art regeneration systems are temperature based with the primary filter protection strategy being limitation of the quantity of soot allowed to accumulate. As shown below, such a strategy can under-utilize the filter capacity by frequent regeneration on a conservative schedule and thus result in a penalty in fuel economy.
A currently challenging durability issue in the CDPF art is cracking or melting of a CDPF substrate due to large temperature excursions within the bed of the filter during regeneration, especially when using an economical filter such as a cordierite monolith. These temperature excursions are caused by the exothermic reaction of carbon and oxygen due to the combined effects of the mass loading and distribution of wet volatile and dry soot within the CDPF, the operating condition of the engine, and the exhaust gas temperature and flow rate through the CDPF. Diesel engine exhaust temperatures are normally in the range of 200-500° C., depending in part on the amount of exhaust gas recirculation, throttle plate position (MVRV—Manifold Vacuum Regulator Valve) and fueling. These engine control parameters, in combination with the manipulation of both fuel quantity and timing in the main fueling and post fueling events, may be used to increase exhaust gas temperature in the range of 500-700° C. as an effective means of initiating a regeneration event and as a means of controlling exhaust gas temperatures supplied to the CDPF during the regeneration process. During regenerative events, when the exhaust gas contains sufficient available oxygen to support the O2 transport process (typically, 5-11%) and an adequate (actual mass depends upon wet/dry soot ratio and total mass) non-homogeneous distribution of wet and dry soot is resident within the CDPF, a highly non-uniform uncontrolled reaction can occur within the CDPF. This rapid, non-uniform reduction of wet and dry soot within the CDPF under various conditions of engine load and exhaust gas temperature and flow may result in excessive thermal gradients and peak monolith temperatures that exceed the material capabilities of the substrate material. This combination of events (rapid oxidation and inadequate heat transfer due insufficient exhaust gas flow) can result in excessive filter temperature and/or temperature gradients, resulting in substrate failure.
A factor not recognized in prior art CDPF regeneration is the relative combustibility difference between “wet” soot and “dry” soot, both of which can be present in a CDPF. By “wet soot” is meant soot particles coated with residual diesel fuel, such as may be generated during periods of high engine load but low engine speed, for example when pulling a heavy vehicle load up a substantial incline in a relatively high gear. Conversely, dry soot may be generated during periods of low engine load and high engine speed, such as at constant highway vehicle speeds. Wet soot burns substantially hotter that dry soot during catalytic regeneration. Indeed, wet soot is inherently rich in hydrocarbons that can explosively ignite, either spontaneously or when regeneration is started, and create an intense exothermic reaction within the CDPF in which temperatures can rise rapidly and uncontrollably (“flash-over”). Further, such intense combustion may occur nonuniformly over a CDPF, creating thermal stresses that can cause cracking or melting of the monolith, resulting in filter failure. Such flash-over is analogous to a creosote fire in a wood stove or fireplace chimney flue.
U.S. Pat. No. 6,735,941 B2 discloses a method for calculating the total soot mass accumulated in a CDPF by measuring differential pressure across the CDPF. This method does not recognize the functional (combustibility) difference between wet soot and dry soot; does not determine the percentage of total soot that is wet soot; and does not provide a strategy for burning off the wet soot in a controlled manner before completing oxidation of the dry soot, to protect against thermal damage to a CDPF.
What is needed in the art is a method for continuously calculating the total soot load and the wet soot fraction of the soot load in a CDPF and determining a relative Combustibility Index for the overall soot content.
It is a principal object of the present invention to prevent damage to a CDPF substrate by overheating during regeneration thereof, by continuously calculating a Combustibility Index for the soot load within a CDPF.
It is a further object of the present invention to improve engine fuel economy by conducting CDPF regeneration only when needed, as indicated by the Combustibility Index, rather than on a fixed schedule.
Briefly described, a method in accordance with the invention for triggering a new regeneration event in a soot-trapping device disposed in an exhaust gas stream of an internal combustion engine comprises the steps of:
a) determining instantaneous engine speed and engine load;
b) determining instantaneous mass fractions for wet soot and for dry soot in said exhaust gas stream for said instantaneous engine speed and load;
c) determining instantaneous concentrations of wet and dry soot particles in said exhaust gas;
d) determining the rates of accumulation of wet soot and dry soot in said soot-trapping device;
e) determining the total amounts of wet soot and dry soot accumulated in said soot-trapping device during all engine operation conditions since the latest previous regeneration event; and
f) triggering said new regeneration event when said total amounts of wet soot and dry soot exceed a permissible value.
The present invention will now be described, by way of example, with reference to the accompanying drawing, in which:
The present invention recognizes that combustibility of a soot load in a CDPF at any point in time during operation of a diesel engine is a function of both the Total Soot Mass and the Wet Soot Percentage. In addition, a forecast of instantaneous combustibility in the near future may be made by determining the rate of change in the Wet Soot Percentage and the Equivalent Accumulation Rate of soot in the CDPF.
Referring to
Presently, there exist no modeling or predictive representations that can accurately estimate exhaust stream particulate emissions or soot distribution within the DPF. However, as these tools are developed they may be directly applied to this methodology. Additionally, the selection of nine regions for the maps 10,20 depicted in
Referring to
As any given accumulation of wet and dry soot produced by the engine is time dependent, the total mass of wet and dry soot per unit time can be determined by integration of the respective components (ppm or micrograms/sec) over discrete time steps within the control embodiment on the order of ten milliseconds (0.01 seconds) or less. Consequently, the mass quantity of both wet soot and dry soot can be determined for a measured period of operation. Additionally, since the effect of exhaust gas temperature and flow on wet soot phase conversion are known (conversion of wet soot in the CDPF to dry soot by exhaust drying), these integrators (up/down) can be modified to account for wet soot drying and reduction effects and allow for even greater accuracy.
Once the fractional relationship of wet soot and dry soot production for each region of
Since volatility is the primary indicator for DPF control decisions in this strategy, the Effective Volatility of the soot emissions produced by the engine for any given time interval is arrived at by means of the following relationship:
This measure of instantaneous effective volatility represents flammability of the physical soot composition accumulated in a DPF and is the indicator that is employed as a decision trigger in various embodiments of DPF control strategies. As effective volatility is a proportion of wet soot fraction to total soot, it will always fall between the values of 0-1.0. Once determined, this index value is a universal value within the control context regardless of the absolute magnitude of mass accumulation.
How is effective volatility used? Once the Instantaneous Volatility Index Value is known, it can be directly referenced to the Instantaneous Wet Soot Equivalent Accumulation Rate of mass production via the relationship indicated in
How can this virtual soot sensor be utilized? The Wet Soot Mass Up/Down Integrator 30 represented schematically in
Note that pre-emptive regen is necessary whenever adequate wet soot is present on the filter, although not necessarily the allowable based upon the Total Soot Mass Regen Target value, and the engine operating condition is atypical. Such a situation exists in an event such as a diesel pulling a heavy load with a relatively high wet soot mass on the filter. If this vehicle were to chance encounter a steep grade over an extended period of time, the engine will begin to operate at extreme power and rpm levels and produce extreme exhaust temperatures and emission products. This may be adequate to light off the Diesel Oxygen Catalyst and the soot load within the DPF in an uncontrolled manner. The wet soot quantity has not met the Target Soot Mass Regen Target value for the active regen process, but the engine operating condition is at an extreme operating point for an extended period, an outlier condition.
This is not the only such extreme condition. One cannot base all potential region heuristics on this case as the system would then encounter unnecessary regen cycles under normal, typical, conditions. However, provisions must be made for this type of control scenario as it will occur periodically and can result in a melted or cracked filter monolith if left undetected and uncontrolled.
In
A second example 36, shown as Case B, is a soot accumulation profile wherein 60/40 wet/dry soot is accumulated for the first 1500 seconds of engine operation. At point 37, the engine then shifts into a operating regime for the next 600 seconds wherein the soot comprises 100% wet soot. At point 38, the engine shifts to a dry soot operating condition. An Equivalent Soot Mass Target Load 39 (Point B) of 33 grams is reached after a total operating time of 2900 seconds, triggering a regeneration event before a dangerously combustible condition develops in the DPF.
How are effective volatility, integrated wet and dry soot mass and accumulated wet and dry soot mass values used in a grand control scheme? Once these parameters are known, it is possible to control a downstream device such as a DPF based upon a primary indicator rather than a proxy indicator such as delta-pressure. In this control method, the knowledge of the proportional relationship between wet (very volatile) soot and dry (less volatile) soot is used as a primary indicator of the need for DPF regeneration at any given time. If a relatively low quantity of total soot is resident on the filter monolith, but is composed of a high proportion of wet volatile soot, the control algorithm can intervene and prevent an uncontrolled flashover (high temperature thermal gradient) by initiating an active controlled regeneration process. Additionally, under certain circumstances dependent upon total soot mass, wet soot proportion and engine operating condition, this information can be used to allow an un-commanded, opportunistic regen to occur due to normal exhaust gas temperature rise without incurring the associated fuel penalty of an active regeneration, thus saving on fuel expenditure per unit of engine operating time. Finally, if the filter soot load accumulation is determined to be composed mostly of dry, less volatile soot, with an adequately low proportion of wet volatile soot, the accumulation period can be extended beyond any “scheduled” regeneration trigger to maximize the time between regenerations. This enables full utilization of the DPF capacity (high efficiency operation) resulting in a further reduced fuel penalty by minimizing the total number of regens required over a given operating cycle. This also has the associated benefit of reduced monolith and catalyst aging effects associated with large numbers of regeneration cycles and results in increased filter durability and longevity.
The threshold at which active regen intervention is required based upon the effective_volatility index-Instantaneous Percent Wet Soot Resident On DPF Filter (%) (fraction) is illustrated in
In a continuous time domain, the equation form is:
Where: MAF=instantaneous engine mass air flow value
Converting this equation form to a discrete time domain that is usable within a controls environment yields:
Case 1: Integration series of instantaneous wet soot mass. The following example illustrates a low wet soot accumulation scenario:
Engine operation: low load and rpm operations predominately within Regions I, II, III, and IV in
Hence, utilizing the data values (non-interpolated) from
For a time step equal to 0.2 seconds, the first five elements of the series would represent the operational dither of wet soot production over a period of 1 second. Carrying this series forward over a time of N seconds would result in a wet soot mass value of 4.3 grams.
Concurrent with the calculation of wet soot mass accumulation, the dry soot mass accumulation is integrated by the same method in the form:
This yields the following:
(Note: Data are drawn from
For a time step equal to 0.2 seconds, the first five elements of the series would represent the operational dither of dry soot production over a period of 1 second. Carrying this series forward over a time of N seconds would result in a dry soot mass value of 38.7 grams.
As established above, the effective_volatility is a determinant control metric that equates the relative instantaneous volatility of the total soot load resident on the filter as:
As the DPF accumulation times are significant, it is impractical to tabulate the entire time series data; for the purposes of example, the calculated effective volatility over N seconds for the sample case above results in an index value of 0.10.
Referencing Case 1 in
Case 2: Integration Series of Instantaneous Wet Soot Mass.
Alternately, if the engine is operating in a region of high load, low rpm, the index and accumulation values would be, respectively:
For a time step equal to 0.2 seconds, the first five elements of this series would represent the operational dither of wet soot production over a period of 1 second. Carrying this series forward over a time of N seconds would result in a wet soot mass value of 7.2 grams.
For the same period and a time step equal to 0.2 seconds, the first five elements of this series would represent the operational dither of dry soot production over a period of 1 second. Carrying this series forward over a time of N seconds would result in a dry soot mass value of 4.8 grams.
In this extreme illustration of extended high load operation, for the purposes of example, the calculated effective volatility over N seconds for the sample case above results in an index value of 0.60.
Referencing Case 2 in
This, in essence, is the usefulness of the volatility index method of predicting soot production and hence the timing of the next required regeneration event. The metrics of instantaneous volatility and effective volatility are used in conjunction with traditional engine mapping or simulation to produce a method of emission estimation measures for downstream device control and tailpipe emissions.
While the invention has been described by reference to various specific embodiments, it should be understood that numerous changes may be made within the spirit and scope of the inventive concepts described. Accordingly, it is intended that the invention not be limited to the described embodiments, but will have full scope defined by the language of the following claims.
Number | Name | Date | Kind |
---|---|---|---|
6735941 | Saito et al. | May 2004 | B2 |
7231761 | Okugawa | Jun 2007 | B2 |
20060042237 | Nonoyama et al. | Mar 2006 | A1 |
20060059901 | Saito et al. | Mar 2006 | A1 |
20060225409 | Kuboshima et al. | Oct 2006 | A1 |
Number | Date | Country |
---|---|---|
2003254042 | Sep 2003 | JP |
Number | Date | Country | |
---|---|---|---|
20090188242 A1 | Jul 2009 | US |