US20120260634A1

US20120260634A1 - NOx CONTROL SYSTEMS AND METHODS FOR CONTROLLING NOx EMISSIONS

Info

Publication number: US20120260634A1
Application number: US13/537,351
Authority: US
Inventors: Maruthi Devarakonda; Gordon Parker; John Johnson
Original assignee: Michigan Technological University
Current assignee: Michigan Technological University
Priority date: 2007-11-26
Filing date: 2012-06-29
Publication date: 2012-10-18
Also published as: US8230677B2; WO2009070734A1; US20090133384A1

Abstract

An exhaust aftertreatment system comprising a selective reduction catalyst, a NO_xsensor or an NH₃sensor, a urea injector, and a dosing control unit, wherein the dosing control unit calculates the rate of urea injection by estimating the concentrations of NO and NO₂in the exhaust downstream of the SCR catalyst.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation of U.S. patent application Ser. No. 12/324,533, filed November 26, 2008, and claims priority to U.S. Provisional patent applications Ser. Nos. 60/990,211, filed Nov. 26, 2007, and 61/052,759, filed May 13, 2008, both of which are incorporated herein by reference in their entirety.

FIELD OF INVENTION

The present invention relates to methods for reducing nitrogen oxide emissions from combustion processes. In particular, the invention relates to the use of catalysts in combination with an ammonia-precursor injector to reduce nitrogen oxide emissions from engines, such as those used by diesel-powered vehicles, equipment, and generators.

BACKGROUND

Nitrogen oxides (NO_x) are a well-known class of atmospheric pollutants that pose risks to human health, agriculture, and property. Nitrogen oxides are key ingredients in the formation of photochemical smog, which damages respiratory functions and limits visibility. Additionally, nitrogen oxides form nitric acid when they react with moisture in the atmosphere, thus forming acid rain. They are also involved in the production of tropospheric ozone. Environmental agencies around the world have instituted standards for nitrogen oxide control in an effort to reduce nitrogen oxides' harmful effects.
Nitrogen oxides broadly include compounds such as NO₃and N₂O₅, but the bulk of atmospheric nitrogen oxides are N₂O, NO and NO₂. When dealing with reactive atmospheric nitrogen, it is often sufficient to deal only with NO and NO₂. (See Seinfeld and Pandis, Atmospheric Chemistry and Physics.)
While there are some natural sources of NO_x, most of the NO in urban environments is a by-product of fossil-fuel combustion. At temperatures found in engines, free-radical chemistry converts atmospheric nitrogen into nitrogen oxides. In particular, lean burning internal combustion engines, such as those used in diesel-powered vehicles, may produce NO_xat a many parts-per-million level. While combustion processes typically produce more NO than NO₂, the two easily interconvert in the presence of oxygen species. Thus, an effective emission control systems must account for both species. Additionally, NO₂-producing processes present a greater health risk than NO-producing processes, as is reported in the literature.
In view of the health risks posed by atmospheric NO_x, the U.S. Environmental Protection Agency regulates acceptable levels of NO_xemissions for U.S. cars and trucks.

SUMMARY OF THE INVENTION

In one embodiment, the invention comprises an exhaust aftertreatment control system for exhaust gases. The exhaust aftertreatment control system comprises a selective catalytic reduction (SCR) catalyst in fluid communication with an exhaust stream from an engine, a urea solution injector, upstream of the SCR catalyst, in fluid communication with the exhaust stream, a gas sensor, downstream of the SCR catalyst, in fluid communication with the exhaust stream, a diagnostic unit, operatively connected to the engine, and a dosing control unit, operatively connected to the gas sensor, the urea solution injector, and the diagnostic unit. The dosing control unit receives a signal indicative of NO, NO₂or NH₃concentrations from the gas sensor, receives a signal indicative of engine parameters from the diagnostic unit, converts the engine parameters to NO and NO₂concentrations, estimates NO and NO₂concentrations downstream from the SCR catalyst, estimates the fractional surface coverage of NH₃in the SCR catalyst, calculates a quantity of urea solution needed to minimize NO_xemissions, and signals the urea solution injector to inject the quantity of the urea solution into the exhaust stream.
In another embodiment, the invention comprises a method for controlling emission of NO_xfrom an engine with an exhaust aftertreatment system having a NO_xsensor. The method comprises inputting engine parameters into a dosing control unit, measuring NO_xconcentration in the exhaust, estimating NO and NO₂concentrations in the exhaust using the engine parameters and the measured NO_xconcentration, estimating fractional surface coverage of NH₃in the catalyst, calculating an amount of urea solution needed to minimize NO_xemissions using the estimate of NO and NO₂concentrations in the exhaust and the estimate of fractional surface coverage of NH₃in the catalyst, and injecting an amount of urea solution.
In another embodiment, the invention comprises a method for controlling emission of NO_xfrom an engine with an exhaust aftertreatment system having a NH₃sensor. The method comprises inputting engine parameters into a dosing control unit, measuring NH₃concentration in the exhaust, estimating NO and NO₂concentrations in the exhaust using the engine parameters and the measured NH₃concentration, estimating fractional surface coverage of NH₃in the catalyst, calculating an amount of urea solution needed to minimize NO_xemissions using the estimate of NO and NO₂concentrations in the exhaust and the estimate of fractional surface coverage of NH₃in the catalyst, and injecting an amount of urea solution.
In another embodiment, the invention comprises an exhaust aftertreatment control system for exhaust gases from an industrial burner. The burner may be fueled with a variety of fuels, including natural gas, heating oil, biomass, or municipal solid waste. The exhaust aftertreatment control system comprises a selective catalytic reduction (SCR) catalyst in fluid communication with an exhaust stream from a burner, a urea solution injector, upstream of the SCR catalyst, in fluid communication with the exhaust stream, a gas sensor, downstream of the SCR catalyst, in fluid communication with the exhaust stream, a diagnostic unit, operatively connected to the burner, and a dosing control unit, operatively connected to the gas sensor, the urea solution injector, and the diagnostic unit. The dosing control unit receives a signal indicative of NO, NO₂or NH₃concentrations from the gas sensor, receives a signal indicative of burner parameters from the diagnostic unit, converts the burner parameters to NO and NO₂concentrations, estimates NO and NO₂concentrations downstream from the SCR catalyst, estimates the fractional surface coverage of NH₃in the SCR catalyst, calculates a quantity of urea solution needed to minimize NO_xemissions, and signals the urea solution injector to inject the quantity of the urea solution into the exhaust stream.
Other aspects of the invention will become apparent by consideration of the detailed description and accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of an embodiment of an exhaust aftertreatment control system using a NO_xsensor.

FIG. 2 is a block diagram of an embodiment of a dosing control unit for the exhaust aftertreatment control system of FIG. 1.

FIG. 3 is a block diagram of an embodiment of an exhaust aftertreatment control system using a NH₃sensor.

FIG. 4 is a block diagram of an embodiment of a dosing control unit for the exhaust aftertreatment control system FIG. 3.

FIG. 5 depicts an experimental setup used to verify an estimator.

FIG. 6 shows the engine conditions used to test the four-state model in Example 1, described hereafter.

FIG. 7 shows the measured NO, NO₂, and NH₃input concentrations in Example 1, described hereafter.

FIG. 8 compares the estimated NO, NO₂, and NH₃output concentrations to experimental data in Example 1, described hereafter.

FIG. 9 compares urea consumption and fractional surface coverage of the SCR catalyst by NH₃for a NO_x-sensor-based and a NH₃-sensor-based exhaust aftertreatment system.

FIG. 10 is a block diagram of an embodiment of an exhaust aftertreatment control system using a NO_xsensor.

DETAILED DESCRIPTION

Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. While the following detailed description describes the invention through reference to embodiments involving an internal combustion engine exhaust treatment system, it should be understood that other systems, both stationary and mobile, where exhaust is treated for NO_xemissions, may also be suitable for a use in accordance with principles of the invention.
Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising,” or “having” and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. Unless specified or limited otherwise, the terms “mounted,” “connected,” “supported,” and “coupled” and variations thereof are used broadly and encompass both direct and indirect mountings, connections, supports, and couplings. Further, “connected” and “coupled” are not restricted to physical or mechanical connections or couplings.
Further, no admission is made that any reference, including any patent or patent document, cited in this specification constitutes prior art. In particular, it will be understood that, unless otherwise stated, reference to any document herein does not constitute an admission that any of these documents form part of the common general knowledge in the art in the United States or in any other country. Any discussion of the references states what their authors assert, and the applicant reserves the right to challenge the accuracy and pertinency of any of the documents cited herein.
One approach for controlling NO_xemissions from combustion processes involves adding ammonia (NH₃) to NO_xin the presence of a catalyst to reduce the NO_xto nitrogen and water. This process, known as selective catalytic reduction (SCR), is described with the balanced equation:
4NH₃+2NO+2NO₂→4N₂+6H₂O
A number of known catalysts may be used in the selective reduction process. See, for example, U.S. Pat. No. 5,451,387, which describes catalysts that may be used for this application.
Unfortunately, ammonia poses its own risks to human health and the environment. Ammonia is caustic, and irritates the lungs. Excess ammonia released into the environment often finds its way into the water system, where it threatens the health of plants and animals that contact the water. Because of this risk, selective catalytic reduction with ammonia requires careful monitoring of ammonia emissions as well as NO_xemissions. The unintended release of ammonia is known as ammonia “slip.”
For large-scale applications, such as power plants, compressed ammonia gas is used for the catalytic reduction of NO_xto nitrogen and water. However, for small-scale or mobile processes, it is often impractical to store/transport ammonia gas for selective catalytic reduction. In some small-scale systems, it is feasible to use highly concentrated ammonia solutions. Highly concentrated ammonia solutions are caustic, however, and present storage and handling issues.
Urea (NH₂CONH₂) solutions provide a suitable alternative to ammonia gas or ammonia solutions for SCR-based NO_xcontrol systems. In a urea SCR system, an aqueous urea solution is injected into a hot exhaust stream, where the urea thermally decomposes into ammonia and carbon dioxide. The resultant ammonia mixes with the exhaust gas prior to entering a catalytic unit where the NO_xis converted to nitrogen and water. Because urea solutions are unreactive, many of the storage and handling concerns are alleviated.
Urea-SCR systems have seen limited use in diesel trucks and diesel-powered marine vessels. Commercially-available urea-SCR systems typically comprise a urea injector (Bosch GmbH, Stuttgart, Germany), an SCR catalyst, and a NO_xsensor (Siemens VGO Trading Ltd., Schwalbach, Germany). State of the art systems employ a rudimentary look-up table based upon engine parameters to provide a guess for the appropriate urea injection rate. These look-up tables are often subject to error due to engine load and exhaust temperature. The NO_xsensor provides feedback as to when the system is exceeding NO_xlimits, allowing the injection rate to be adjusted in an iterative fashion. Alternatively, or in addition, an NH₃sensor may be used to provide feedback to minimize the amount of ammonia slip.
Because the current systems do not model the catalytic reactions in real time, current systems use more urea solution than necessary and have unacceptable levels of ammonia slip. Additionally, the current systems may be ineffective for controlling NOx emissions for vehicles that stop and start frequently, e.g., busses and garbage trucks.
The German Association of the Automobile Industry has championed urea-SCR NO_xcontrol, and advocates for a network of filling stations where diesel vehicles can replenish urea solutions. Suitable urea solutions are now commercially-available throughout Europe under the trade name AdBlue. AdBlue solutions comprise 32.5% urea in deionized water, and are sold by BASF, Yara, and GreenChem, among others. These solutions typically cost $4-7/gal. The rate of consumption varies tremendously depending upon the type of engine and the loading, however, it is not uncommon for an engine to consume 5 gallons of urea solution for each 100 gallons of diesel consumed. The high cost of urea solutions is motivating equipment manufacturers to maximize NO_xcontrol per unit of solution.
The invention described herein provides a closed-loop system for minimizing NO_xemissions from combustion processes while minimizing the consumption of urea solution and producing negligible ammonia slip. In one embodiment, the aftertreatment system comprises a selective catalytic reduction (SCR) unit, a urea injector, a gas sensor, and a dosing control unit. The gas sensor is sensitive to one or more gases that are to be controlled, including, but not limited to, NO, NO₂, or NH₃. The catalyst within the SCR unit facilitates the conversion of ammonia and NO_xto nitrogen and water. The catalyst may comprise any of a number of catalysts known to work for selective catalytic reduction, including, but not limited to, compounds of vanadium, iron, or copper, and metal zeolites. The urea injector is responsive to the dosing control unit which regulates the concentration of ammonia in the SCR unit by increasing or decreasing the rate of urea injection.
In one embodiment, the gas sensor is a NO_xsensor. The NO_xsensor provides NO_xconcentrations to a dosing control unit, thus allowing the dosing control unit to estimate the NO, and NO₂concentrations upstream from the SCR. In an alternate embodiment, the gas sensor is an NH₃sensor that provides the NH₃concentration to the dosing control unit, thus allowing the dosing control unit to estimate the NO, and NO₂concentrations upstream from the SCR. In both embodiments, the dosing control unit is capable of calculating the optimum urea injection rate to minimize NO_xemissions with minimal ammonia slip.
In many embodiments, the dosing control unit is suitably a single electronic unit. Nonetheless, the dosing control unit can be functionally described as comprising two components, an estimator and a controller. Using engine or burner operating parameters and total NO_xor total NH₃output, the estimator employs a mathematical model to estimate the concentrations of NO, NO₂, and NH₃downstream from the SCR catalyst as well as the fractional surface coverage of NH₃within the SCR catalyst. The estimator then passes these estimates to the controller which calculates the amount of NH₃needed to minimize NO_xoutput with minimal NH₃slip. Having calculated the optimum NH₃concentration, the controller then directs the urea injector to inject urea at an appropriate rate.
FIG. 1 is a block diagram of an embodiment of an exhaust aftertreatment system 10 employing a NO_xsensor according to the principles of the invention. The exhaust aftertreatment system 10 includes a urea injector 30, a selective catalytic reduction (SCR) catalyst 35, a NO_xsensor 40, and a dosing control unit 50. An internal combustion engine 15 produces an exhaust stream 20 containing nitrogen oxides, including NO and NO₂. As exhaust stream 20 moves through aftertreatment system 10, exhaust stream 20 is mixed with an atomized urea solution 25 from urea injector 30 to create urea-enriched exhaust 32. Urea-enriched exhaust 32 then enters SCR catalyst 35. SCR catalyst 35 contains metal zeolites arranged in a honeycomb such that the passing exhaust gasses will interact with a multitude of catalytic sites. As urea-enriched exhaust 32 is swept into SCR catalyst 35, the urea breaks down into ammonia (NH₃) and carbon dioxide (CO₂). Once inside SCR catalyst 35, the nitrogen oxides react with ammonia in the presence of the metal zeolite to produce nitrogen gas (N₂) and water vapor (H₂O). NO_x sensor 40 measures the NO_xin a scrubbed exhaust stream 37 exiting SCR catalyst 35, and provides a signal indicating the concentration of NO_xto dosing control unit 50. Dosing control unit 50 uses the measured NO_xconcentration, as well as operating parameters from engine 15, which originate from an onboard diagnostic unit (OBD) 60. Operating parameters may include exhaust gas temperature, speed, load, air-flow rate, fuel-flow rate and/or exhaust gas pressure. Dosing control unit 50 then calculates the rate at which atomized urea solution 25 should be mixed with exhaust stream 20, and adjusts urea injector 30 accordingly.
Dosing control unit 50 is an electronic control unit comprising random access memory (RAM) 52, read only memory (ROM) 55, and a microprocessor (CPU) 58. Functionally, dosing control unit 50 may be described as comprising an estimator 70 and a controller 80, as shown in FIG. 2. Estimator 70 collects information about the engine from OBD 60, as well as information from NO_x sensor 40. Using this information, estimator 70 estimates the concentration of NO, NO₂, and NH₃in exhaust stream 20, prior exhaust stream 20 entering SCR catalyst 35. Estimator 70 also estimates the NH₃loading, θ, in the catalyst using equation VIII, discussed hereafter. Controller 80 then uses the estimated NH₃loading and the NO, NO₂and NH₃concentrations as inputs to equation XIV, described hereafter. Controller 80 then adjusts the rate at which atomized urea solution 25 is injected exhaust stream 20 to minimize NO, NO₂and NH₃emissions in scrubbed exhaust stream 37.
The estimator employs a four-state model to estimate the concentrations of NO, NO₂and NH₃within SCR catalyst 35. The four-state model encompasses six chemical equations (I-VI) which describe the simultaneous reactions within SCR catalyst 35
4NH₃+2NO+2NO₂→4N₂+6H₂O (I)
4NH₃+4NO+O₂→4N₂+6H₂O (II)
4NH₃+3NO₂→7/2N₂+6H₂O (III)
4NH₃+3O₂→2N₂+3H₂O (IV)
NH₃+S→NH₃* (V)
NH₃*→NH₃+S (VI)
wherein S represents the free active sites in the catalyst, and NH₃* is NH₃adsorbed at an active site in the catalyst. Incorporating the appropriate rate constants for chemical reactions I-VI, the time-dependent states of NO, NO₂, NH₃, and the ammonia storage in the catalyst can be represented as
Ċ_NO=− QC_NO −k ₁ΩθC_NOC_NO ₂ −k ₂ΩθC_NOC_O ₂+ QC_NO,in
Ċ_NO ₂=− QC_NO ₂ −k ₁ΩθC_NOC_NO ₂ −k ₃ΩθC_NO ₂+ QC_NO ₂ _,in
{dot over (θ)}=−(k ₆ +k ₄)θ+k ₅C_NH ₃ −k ₁C_NOC_NO ₂ θ−k ₂C_NOC_O ₂ θ−k ₃C_NO ₂ θ−k ₅θC_NH ₃
Ċ_NH ₃ =k ₆Ωθ−(k ₅Ω+ Q)C_NH ₃ +k ₅ΩθC_NH ₃+ Q k ₅C_NH ₃ _,in (VII)
wherein
Q is the normalized flow rate into the SCR catalyst 35 in sec⁻¹, wherein Q=Q/V, and Q is the actual volumetric flow rate in actual m³/sec, and V is the volume of the exhaust gas in the substrate channel in m³,
C_iis the concentration of species i in moles/m³, where i=NO, NO₂, NH₃, or O₂,
C_i,inis the concentration of species i entering SCR catalyst 35, where i=NO, NO₂, NH₃,
θ is the fractional surface coverage of SCR catalyst 35 by NH₃,
Ω is SCR catalyst 35 adsorption capacity in moles of NH₃/m³,
k₁is the rate constant for equation I,
k₂is the rate constant for equation II,
k₃is the rate constant for equation III,
k₄is the rate constant for equation IV,
k₅is the rate constant for equation V,
k₆is the rate constant for equation VI.
In one embodiment, an estimator would employ a converging linear state equation of the form
$\begin{matrix} {\begin{matrix} {\dot{C}}_{NO, est} \\ {\dot{C}}_{{NO}_{2}, est} \\ {\dot{θ}}_{est} \\ {\dot{C}}_{{NH}_{3}, est} \end{matrix}} = \overset{->}{f} (C_{NO, est}, C_{NO_{2}, est}, θ_{est}, C_{NH_{3}, est}, C_{NO, in}, C_{NO_{2}, in}) + \overset{->}{L} (C_{NO, meas} + C_{NO_{2}, meas} - C_{NO, est} - C_{NO_{2}, est}) & (VIII) \end{matrix}$
to estimate θ and the concentrations of NO, NO₂, and NH₃downstream from the SCR catalyst. {right arrow over (L)} serves to stabilize the estimator. That is, when the system begins, it is necessary to employ an estimate for the NO and NO₂, however, once the system is running, the measured and estimated values of NO and NO₂should converge, and {right arrow over (L)} will consequently disappear. The estimator's ability to predict concentrations of NO, NO₂, and NH₃was verified experimentally, as described hereafter in Example 1.
For some applications, it may be computationally expensive to estimate the four states using equation VIII. In an alternative embodiment, an acceptable estimate may be obtained quicker and/or with less computational resources by solving Ċ_NO, Ċ_NO ₂, and Ċ_NH _Has steady state expressions. This approximation results in a single state model for the fractional NH₃coverage in the catalyst
{dot over (θ)}=(−k ₆ +k ₄)θ+k ₅C_NH ₅ −k ₂C_NOC_NO ₂ θ−k ₂C_NOC_O ₂ θ−k ₃C_NO ₂ θ−k ₅θC_NH ₃ (IX)
and a corresponding linear state estimator of the form
{dot over (θ)}_est −{right arrow over (f)}(C_NO, C_NO ₂, θ, C_NH ₃, C_NO,in, C_NH ₃ _,in)+{right arrow over (L)}(C_NO _x _,meas−C_NO _x _,est) (X)
where C_NO _x _,measis the measurement from the NO_xsensor 16. Because the NO_xsensor is cross-sensitive to NH₃, C_{NO ,est}must actually incorporate an estimate of the downstream NH₃concentration. Thus, C_NO _x _,estis
C_NO _x _,est=A₁C_NO+A₂C_NO ₂+A₃(α)C_NH ₃ (XI)
where A₁, A₂, and A₃(α) are specific to the aftertreatment control system.
The initial concentrations of NO and NO₂are calculated by estimator 70 using a model specific for internal combustion engine 15 and additional emission equipment that may lie between internal combustion engine 15 and SCR 35 (e.g. a catalyzed particulate filter). Onboard diagnostic unit 60 reports engine parameters, such as engine speed, load, exhaust gas temperatures, flow rates and exhaust gas pressure, to estimator 70. Estimator 70 uses these engine parameters to produce an estimate of NO and NO₂concentrations for the purpose of estimating C_NO _x. Estimator 70 also estimates the NH₃entering SCR catalyst 35 based upon the rate that urea injector 30 is injecting atomized urea solution 25 into exhaust stream 20 previous to urea-enriched exhaust 32 entering SCR catalyst 35.
Controller 80 functions to minimize the NO, NO₂, and NH₃in scrubbed exhaust stream 37 while using as little atomized urea solution 25 as necessary. The controller's efficiency at converting NO and NO₂to N₂per mole of NH₃can be expressed as
$\begin{matrix} η_{T} = \frac{C_{NO, i n} + C_{NO_{2}, i n} - C_{NO, out} - C_{NO_{2}, out} - λ C_{NH_{3}, out}}{C_{NO, i n} + C_{NO_{2}, i n}} & (XIII) \end{matrix}$
wherein λ accounts for NH₃slipping out of the catalyst.
Defining a new function e_p= p _des− p, where p is a linear combination of C_NO, C_NO ₂, C_NH ₃, and p _desis a linear combination of the desired concentrations, i.e., p _des=C_NO,des+C_NO ₂ _,des+αC_NH ₃ _,des, the controller's goal can be represented as e_p=ė_p=0.
Substituting η_T=1−p_des, the dynamic concentration of NH₃within the SCR catalyst 15 for the goal e_p=ė_p=0 can be expressed as,
$\begin{matrix} C_{NH_{3}, i n, d y n} = C_{NH_{3}, est} + \frac{1}{λ} (C_{NO, est} + C_{NO_{2}, est} - C_{NO, i n} - C_{NO_{2}, i n}) + \frac{1}{Q} (k_{5} Ω (1 - θ_{est}) C_{NH_{3}, est} - k_{6} {Ωθ}_{est}) + \frac{1}{λ \overline{Q}} ({\dot{\overline{p}}}_{des} + 2 {Ωθ}_{est} k_{1} C_{NO, est} C_{NO_{2}, est} + {Ωθ}_{est} k_{2} C_{NO, est} C_{O_{2}} + {Ωθ}_{est} k_{3} C_{NO_{2}, est}) & (XIV) \end{matrix}$
As the system, in accordance with the invention, is subject to nonlinear dynamics, controller 80 must employ an additional correction term to ensure stability in the event of measurement errors or other transient disturbances. Therefore, an additional term is added that penalizes deviations from the objective e_p=0,
C_NH ₃ _,in=C_NH ₃ _,in,dyn −Γsgn(e _p) (XV)
where Γ is a control variable that can be tuned to meet the control objective. Based upon the sign of sgn(e_p), the signum function changes:
$\begin{matrix} sgn (x) = {\begin{matrix} - 1 : x < 0 \\ 0 : x = 0 \\ 1 : x > 0 \end{matrix}} & (XVI) \end{matrix}$
Thus, given estimates of NO, NO₂, NH₃and free catalytic sites from estimator 70, controller 80 calculates the optimum concentration of ammonia to achieve its efficiency goals. Upon calculating this value, controller 80 adjusts the rate of injection of atomized urea solution 25 from injector 30 assuming a 2:1 molar ratio of ammonia to atomized urea solution 25.
As an alternative to NO_x sensor 40 downstream of SCR catalyst 35, an NH₃sensor 140 can be used for SCR control applications. See FIG. 3. NH₃sensor 140 has the additional, beneficial, characteristic of having minimal cross-sensitivity toward NO and NO₂species, as is reported in the literature (See Wang, DEER 2007; Moos, Sensors and Actuators, 2002.) Because of the minimal cross-sensitivity, ammonia-sensor-based exhaust aftertreatment system 110 requires fewer variables than exhaust aftertreatment system 10, employing NO_x sensor 40.
FIG. 3 is a block diagram depicting an ammonia-sensor-based exhaust aftertreatment system 110 embodying the principles of the invention for catalytic reduction of NO_xin the presence of ammonia. The ammonia-sensor-based exhaust aftertreatment system 110 includes urea injector 30, SCR catalyst 35, an NH₃sensor 140, and an ammonia-sensor-based dosing control unit 150. Internal combustion engine 15 produces exhaust stream 20 containing nitrogen oxides, including NO and NO₂. As exhaust stream 20 moves through the ammonia-sensor-based aftertreatment system 110, exhaust stream 20 is mixed with atomized urea solution 25 from urea injector 30 to create urea-enriched exhaust 32. Urea-enriched exhaust 32 then enters SCR catalyst 35. SCR catalyst 35 contains metal zeolites arranged in a honeycomb such that the passing exhaust gasses will interact with a multitude of catalytic sites. As urea-enriched exhaust 32 is swept into SCR catalyst 35, the urea breaks down into ammonia (NH₃) and carbon dioxide (CO₂). Once inside SCR catalyst 35, the nitrogen oxides react with ammonia in the presence of the metal zeolites to produce nitrogen gas (N₂) and water vapor (H₂O). NH₃sensor 140 measures the NH₃exiting SCR catalyst 35, providing a signal indicating the concentration of NH₃in scrubbed exhaust stream 37 to the ammonia-sensor-based dosing control unit 150. Ammonia-sensor-based dosing control unit 150 uses the measured NH₃concentration, as well as internal combustion engine 15 operating parameters, which originate from OBD 60. Engine parameters may include exhaust gas temperature, speed, load, air-flow rate, fuel-flow rate and/or exhaust gas pressure. Ammonia-sensor-based dosing control unit 150 then calculates the rate at which the atomized urea solution 25 should be mixed with exhaust stream 20, and adjusts injector 30 accordingly.
Ammonia-sensor-based dosing control unit 150 is an electronic control unit comprising random access memory (RAM) 152, read only memory (ROM) 155, and a microprocessor (CPU) 158. Functionally, the ammonia-sensor-based dosing control unit 150 may be described as comprising an ammonia-sensor-based estimator 170 and an ammonia-sensor-based controller 180. The functions of ammonia-sensor-based estimator 170 and ammonia-sensor-based controller 180 are illustrated in FIG. 4. Ammonia-sensor-based estimator 170 collects information about the engine from OBD 60, as well as information from NH₃sensor 140. Using this information, ammonia-sensor-based estimator 170 estimates the concentration of NO, NO₂, and NH₃in exhaust stream 20, prior to exhaust stream 20 entering SCR catalyst 35. Ammonia-sensor-based estimator 170 also estimates the NH₃loading, θ, in the catalyst using equation VIII, discussed above. Ammonia-sensor-based controller 180 then uses the estimated NH₃loading and the NO, NO₂and NH₃concentrations as inputs to equation XIV, described previously. Ammonia-sensor-based controller 180 then adjusts the rate at which atomized urea solution 25 is injected into exhaust stream 20 to minimize NO, NO₂and NH₃in scrubbed exhaust stream 37.
Ammonia-sensor-based estimator 170 employs the same four-state model, described above with respect to equations (I)-(VII), to estimate the concentrations of NO, NO₂and NH₃within SCR catalyst 35. However, ammonia-sensor-based estimator 170 employs a modified converging linear state equation of the form
$\begin{matrix} {\begin{matrix} {\dot{C}}_{NO, est} \\ {\dot{C}}_{{NO}_{2}, est} \\ {\dot{θ}}_{est} \\ {\dot{C}}_{{NH}_{3}, est} \end{matrix}} = \overset{->}{f} (C_{NO, est}, C_{NO_{2}, est}, θ_{est}, C_{NH_{3}, est}, C_{NO, in}, C_{NO_{2}, in}) + \overset{->}{L} (C_{NH_{3}, meas} - C_{NH_{3}, meas}) & (XVII) \end{matrix}$
to estimate θ and the concentrations of NO, NO₂, and NH₃downstream from the SCR catalyst. {right arrow over (L)} serves to stabilize the estimator. That is, when the system begins, it is necessary to employ an estimate for the NH₃, however, once the system is running, the measured and estimated values of NH₃should converge, and {right arrow over (L)} will consequently disappear.
For some applications, it may be computationally expensive to estimate the four states using equation XVII. In an alternative embodiment, an acceptable estimate may be obtained quicker and/or with less computational resources by solving Ċ_NO, Ċ_NO ₂, and Ċ_NH _Has steady state expressions. Using equation (IX), above, to describe a single state model for the fractional NH₃coverage in the catalyst, a corresponding linear state estimator for the use of an NH₃sensor 140 can be determined
{dot over (θ)}_est ={right arrow over (f)}(C _NO, C_NO ₂, θ_est, C_NH ₃, C_NO,in, C_NO ₂ _,in, C_NH ₃ _{, in})+{right arrow over (L)}(C_NH _3x _{, meas}−C_NH ₃ _,est) (XVIII)
where C_NH ₃ _,measis the NH₃concentration measured by the NH₃sensor 140, and C_NH ₃ _,estis calculated based upon an NH₃sensor model given by
C_NH ₃ _,est=A₃(α)C_NH ₃ (XIX)
where A₃is determined as a function of α, which is specific to ammonia-sensor-based aftertreatment control system 110.
The initial concentrations of NO and NO₂are calculated by ammonia-sensor-based estimator 170 using a model specific for internal combustion engine 15 and additional emission equipment that may lie between internal combustion engine 15 and SCR 35 (e.g. a catalyzed particulate filter). OBD 60 reports engine parameters, such as engine speed, load, exhaust gas temperatures, flow rates and exhaust gas pressure, to ammonia-sensor-based estimator 170. Ammonia-sensor-based estimator 170 uses these engine parameters to produce an estimate of NO and NO₂concentrations for the purpose of estimating C_NH ₃. The estimator also estimates the NH₃entering SCR catalyst 35 based upon the rate that urea injector 30 is injecting atomized urea solution 25 into the exhaust stream previous to entering SCR catalyst 35.
Ammonia-sensor-based controller 180 functions to minimize the NO, NO₂, and NH₃leaving the aftertreatment system, while using as little atomized urea solution 25 as necessary, as is described in equation (XIII) above. However, because NH₃sensor 140 measures the NH₃slip directly, it is unnecessary to calculate a dynamic NH₃concentration after the SCR as is done in the NO_x-sensor-based exhaust aftertreatment system 10. Thus, given estimates of NO, NO₂, NH₃and free catalytic sites from ammonia-sensor-based estimator 170, ammonia-sensor-based controller 180 can calculate the optimum concentration of ammonia to achieve its efficiency goals. Upon calculating this value, ammonia-sensor-based controller 180 adjusts the rate of atomized urea solution 25 injection from injector 130 assuming a 2:1 molar ratio of ammonia to atomized urea solution 25.

EXAMPLE 1

Verification of Four-State Model

FIG. 5 depicts an experimental setup in accordance with the invention. A Navistar 16 7.6L diesel engine 210 was connected to a diesel oxidation catalyst 230 and a catalyzed particulate filter 240, and then outfitted with urea injector 30 and a first SCR catalyst 35 and a second SCR catalyst 38. [Generally an exhaust aftertreatment system would have only one SCR catalyst, as is depicted in FIG. 1.] The exhaust concentrations of NO and NO₂were measured before first SCR catalyst 35 with a first emission test bench 250 and after second SCR catalyst 38 with a second emission test bench 255. The NH₃concentrations were measured before first SCR catalyst 35 with a first Fourier Transform Infrared Spectrometer (FTIR) 260 and after second SCR catalyst 38 with a second Fourier Transform Infrared Spectrometer (FTIR) 265.
Engine 210 was run at a variety of speed and load conditions as shown in FIG. 6, resulting in a variety of engine temperatures and exhaust gas flow rates. The rate of urea injection was adjusted manually. FIG. 7 shows the NO and NO₂concentrations measured with first emission test bench 250 during the engine conditions depicted in FIG. 6. The NH₃concentration in FIG. 7 was calculated from the rate of urea injection in FIG. 6.
The resulting NO and NO₂concentrations were measured with second emission test bench 255. The resulting NH₃concentrations were measured with second FTIR 265. These measurements are illustrated with dashed lines in FIG. 8. For comparison, the NO, NO₂, and NH₃concentrations after first and second SCR catalysts 35 and 38 were estimated using the four state model (Equation VIII), engine 210 parameters, and the rate of urea solution injection. The estimated concentrations are depicted with a solid line.
As illustrated in FIG. 8, the four-state model provides a suitable method for estimating NO, NO₂, and NH₃concentrations.
As a final verification of the robustness of the invention described herein, the NO_xsensor model, described by equation (VIII) was used to control the rate of urea injection in the setup shown in FIG. 5, and then the NH₃sensor model, described by equation (XVII) was used to control the rate of urea injection in a similar experimental setup. As can be seen in FIG. 9, both sensor models result in similar rates of urea injection and fractional surface coverage, θ, of the SCR catalyst by NH₃.

EXAMPLE 2

Controlling NO_xEmissions from Industrial Burner

FIG. 10 is a block diagram of a prophetic embodiment of an exhaust aftertreatment system 310 employing a NO_xsensor according to the principles of the invention. The exhaust aftertreatment system 310 will include a urea injector 30, a selective catalytic reduction (SCR) catalyst 35, a NO_xsensor 40, and a dosing control unit 50. An oil-fueled industrial burner 315 (used to power a 66 MW steam generator) will produce an exhaust stream 20 containing nitrogen oxides, including NO and NO₂. Exhaust stream 20 will move through aftertreatment system 310, exhaust stream 20 will be mixed with an atomized urea solution 25 from urea injector 30 to create urea-enriched exhaust 32. Urea-enriched exhaust 32 will then enter SCR catalyst 35. SCR catalyst 35 will contain metal zeolites arranged in a honeycomb such that the passing exhaust gasses will interact with a multitude of catalytic sites. As urea-enriched exhaust 32 is swept into SCR catalyst 35, the urea will break down into ammonia (NH₃) and carbon dioxide (CO₂). Once inside SCR catalyst 35, the nitrogen oxides will react with ammonia in the presence of the metal zeolite to produce nitrogen gas (N₂) and water vapor (H₂O). NO_x sensor 40 will measure the NO_xin a scrubbed exhaust stream 37 exiting SCR catalyst 35, and will provide a signal indicating the concentration of NO_xto dosing control unit 350. Dosing control unit 350 will use the measured NO_xconcentration, as well as operating parameters from industrial burner 315, which will originate from a burner diagnostics unit 360. Operating parameters will include exhaust gas temperature, steam load, steam temperature, air-flow rate, fuel-flow rate and exhaust gas pressure. Dosing control unit 350 then will calculate the rate at which atomized urea solution 25 will be mixed with exhaust stream 20, and will adjust urea injector 30 accordingly. Dosing control unit 350 is an electronic control unit comprising random access memory (RAM) 52, read only memory (ROM) 55, and a microprocessor (CPU) 58. Functionally, dosing control unit 350 will be nearly identical to dosing control unit 50, shown in FIG. 2. Dosing control unit will be programmed to estimate the concentration of NO_xin exhaust stream 20 based upon empirical measurements of NO_xin exhaust stream 20 under various operating conditions.
Thus, the invention provides systems and methods for the reduction of NO_xemissions from a combustion process by measuring engine parameters and concentrations of NO_xor NH₃, estimating NO and NO₂concentrations, injecting a urea solution into the exhaust stream, and selectively reducing NO_xto N₂in the presence of a catalyst. The methods and systems minimize urea solution consumption at the same time they minimize NO_xand ammonia emissions.
The above embodiments and examples do not serve as limitations on the invention, as other embodiments may employ the principles of the invention described herein. For example, the principles of the invention may be employed to create an exhaust aftertreatment system for NO_xreduction in industrial burners, such as those used in furnaces, boilers, or power plants. These burners may be powered by a variety of fuels, including, but not limited to, natural gas, heating oil, biomass, or municipal solid wastes.
Furthermore, Example 1 does not indicate that the invention is limited to diesel engines such as those used for over-the-road trucking With only minor modifications, one skilled in the art could apply the principles of the invention described herein to reduce NO_xemissions from any diesel powered vehicle or equipment, including locomotives, construction equipment, earthmoving equipment, or marine craft.
All publications, patents, and patent applications are herein expressly incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated by reference. In case of conflict between the present disclosure and the incorporated patents, publications and references, the present disclosure should control.
Various features and advantages of the invention are set forth in the following claims.

Claims

1. A method for minimizing NO_xin an exhaust stream from an engine having an exhaust aftertreatment system comprising a urea solution injector and a catalyst, the method comprising:

collecting engine operating parameters;

measuring NO_xconcentrations in the exhaust stream;

estimating a NO concentration and a NO₂concentration in the exhaust stream using the engine operating parameters and the measured NO_xconcentration;

estimating a fractional surface coverage of NH₃in the catalyst;

calculating an amount of urea solution needed to minimize NO_xemissions using the estimate of the NO concentration and the estimate of the NO₂concentration in the exhaust and the estimate of fractional surface coverage of NH₃in the catalyst; and

injecting an amount of urea solution into the exhaust stream.

2. The method of claim 1, wherein the engine is a diesel engine.

3. The method of claim 1, wherein the engine parameters are selected from the group consisting of engine speed, engine load, engine temperature, exhaust gas temperature, exhaust flow rate, and exhaust pressure.

4. The method of claim 1, wherein a state estimator is used to calculate the quantity of urea solution needed to minimize NO_xemissions.

5. A method for minimizing NO_xin an exhaust stream from an engine having an exhaust aftertreatment system comprising a urea solution injector and a catalyst, the method comprising:

collecting engine operating parameters;

measuring NH₃concentrations in the exhaust stream;

estimating a NO concentration and a NO₂concentration in the exhaust stream using the engine operating parameters and the measured NH₃concentration;

estimating a fractional surface coverage of NH₃in the catalyst;

calculating an amount of urea solution needed to minimize NO_xemissions using the estimate of the NO concentration and the NO₂concentration in the exhaust and the estimate of fractional surface coverage of NH₃in the catalyst; and

injecting an amount of urea solution into the exhaust stream.

6. The method of claim 5, wherein the engine is a diesel engine.

7. The method of claim 5, wherein the engine parameters are selected from the group consisting of engine speed, engine load, engine temperature, exhaust gas temperature, exhaust flow rate, and exhaust pressure.

8. The method of claim 5, wherein a state estimator is used to calculate the quantity of urea solution needed to minimize NO_xemissions.