KR102431125B1 - NOx reduction system and control method thereof - Google Patents

Abstract

An embodiment of the present invention relates to a nitrogen oxide (NOx) reduction system which reduces NOx by performing a reductive reaction between NOx and a reducing agent. The NOx reduction system of the present invention comprises: an exhaust gas discharge unit discharging exhaust gas containing NOx; a reducing agent supply unit supplying a reducing agent to reduce the NOx; a reductive reaction unit performing a reductive reaction between the NOx and the reducing agent; a catalyst unit including a first catalyst layer which accelerates a speed of the reductive reaction; and a control unit measuring the activity of the first catalyst layer, controlling an amount of the NOx discharged by the exhaust gas discharge unit, based on the measured activity, and controlling an amount of the supplied reducing agent to correspond stoichiometrically to the amount of the NOx discharged by the exhaust gas discharge unit.

Description

Nitrogen oxide reduction system and its control method {NOx reduction system and control method thereof}

The present invention relates to a nitrogen oxide reduction system for reducing nitrogen oxides and a control method thereof.

In general, the necessity to maintain the air environment in a clean state by reducing nitrogen oxide, which is a harmful exhaust gas, to the maximum limit by mobilizing various methods among products generated by burning fossil fuels in factories or combustion devices, etc. is a reality.

As described above, as a representative method for reducing nitrogen oxides currently known, a selective catalytic reduction method and a selective non-catalytic reduction method are applied. The selective catalytic reduction method and the selective non-catalytic reduction method are well-known and well-known techniques, and detailed description thereof will be avoided.

As prior art related thereto, as described in Patent Document 1 (Title of the Invention: Nitrogen Oxide Reduction System Using Selective Catalytic Reduction Method) filed by the applicant on January 18, 2007, the selective catalytic reduction method (SCR; Selective Catalytic Reduction is designed to reduce nitrogen oxides by supplying ammonia as a reducing agent through an ammonia injection facility (AIG) downstream of the combustion device to cause a reduction reaction in the catalytic reaction tower. In this selective catalytic reduction method, when the combustion device is in a low load state or the temperature of the exhaust gas to be introduced into the catalytic reaction tower is low, the reaction of nitrogen oxides remarkably falls, so that the ammonia slip with the exhaust gas as it is. There is a problem that a slip phenomenon may occur, which may cause side effects to the downstream equipment.

In order to solve this problem, as disclosed in Patent Document 2 (Title of the Invention: Nitrogen Oxide Reduction System Using Urea Water) filed on July 14, 2017, the Applicant submitted carbonic acid from urea water at a relatively low temperature. A reduction reaction with nitrogen oxides was developed by carrying out the conversion of ammonium and the conversion of ammonium carbonate to ammonia.

However, these two technologies only improve the equipment before the catalytic reaction tower, and there is no description about the problems caused by the catalytic reaction tower. Catalyst reaction tower basically requires a catalyst layer to promote the reaction rate of nitrogen oxide and ammonia. Although the activity of the catalyst bed is still usable, it has a problem that it must be discarded.

Korean Patent Application No. 10-2007-0005662 Korean Patent Application No. 10-2017-0089829

An object of the present invention is to provide a nitrogen oxide reduction system for measuring the activity of a catalyst layer, and controlling the amount of nitrogen oxide emission and reducing agent supply, based on the activity.

In addition, an object according to an embodiment of the present invention is to measure the activity of the catalyst layer, and based on the activity, to control the amount of nitrogen oxide emission and reducing agent supply, or to control the flow path direction of nitrogen oxide emission and reducing agent supply amount It relates to a control method of a nitrogen oxide reduction system.

The nitrogen oxide reduction system for reducing the nitrogen oxides by reducing the nitrogen oxides and the reducing agent according to an embodiment of the present invention is an exhaust gas discharge unit for discharging exhaust gas containing nitrogen oxides, to reduce the nitrogen oxides A reduction supply unit for supplying a reducing agent for the purpose, a reduction reaction unit for reducing the nitrogen oxide and the reducing agent, a catalyst unit comprising a first catalyst layer for accelerating the reaction rate for the reduction reaction, and measuring the activity of the first catalyst layer And, based on the measured activity, it may include a control unit for controlling the amount of nitrogen oxide emission of the exhaust gas discharge unit and controlling the supply amount of the reducing agent to correspond stoichiometrically to the nitrogen oxide emission.

In addition, the nitrogen oxide reduction system further includes a driving characteristic information generating unit to control the NOx emission, and the driving characteristic information generating unit includes a driving characteristic information collecting unit collecting driving characteristic information, and collecting nitrogen oxide emission information a nitrogen oxide emission information collecting unit, and a neural network model trained to derive specific driving characteristic information corresponding to specific nitrogen oxide emission information by using the collected driving characteristic information and fine oxide emission information, The characteristic information may include at least one of an exhaust gas amount, a combustion temperature, a combustion speed, and an exhaust gas exhaust load related to a driving operation of the exhaust gas exhaust part.

In addition, the activity may be measured from the nitrogen oxide removal efficiency and the by-product generated in excess of the supply amount of the reducing agent.

In addition, when the control unit determines that the reaction rate cannot be accelerated with the measured activity of the first catalyst layer, the control unit determines the amount of nitrogen oxide that can promote the reaction rate only with the measured activity of the first catalyst layer. Calculating, inputting nitrogen oxide emission information for the calculated NOx emission into the learned neural network model, and deriving driving characteristic information corresponding to the NOx emission information, so that the current driving with the derived driving characteristic information By adjusting the characteristic information, it is possible to adjust the supply amount of the reducing agent to correspond stoichiometrically to the calculated nitrogen oxide emission amount.

In addition, the catalyst section changes the direction of the second catalyst layer for accelerating the reaction rate for the reduction reaction, and the direction of the nitrogen oxide and reducing agent flowing from the reduction reaction section to the first catalyst layer or the second catalyst layer. Further comprising a part, wherein when the control unit determines that the reaction rate cannot be accelerated by the measured activity of the first catalyst layer, the control unit determines the reaction rate for the reduction reaction of some of the nitrogen oxide emission amount and the reducing agent supply amount. In order to promote the activity, the path direction change part is controlled to flow the part to the first catalyst layer, and the reaction rate for the reduction reaction of the remaining amount of the nitrogen oxide emission and the reducing agent supply amount is promoted by the activity of the second catalyst layer. For this purpose, the remaining amount may flow into the second catalyst layer by controlling the path direction change unit.

The control method of the nitrogen oxide reduction system for reducing the nitrogen oxides by reducing the nitrogen oxides and the reducing agent according to an embodiment of the present invention includes the steps of obtaining nitrogen oxide emissions based on the operation characteristic information of the exhaust gas discharge unit , obtaining a reducing agent supply amount stoichiometrically corresponding to the nitrogen oxide emission amount, measuring the activity of the catalyst layer, and determining the reaction rate for the reduction reaction when the nitrogen oxide emission amount and the reducing agent supply amount reduce reaction as the activity Determining whether it can be promoted, and when it is determined that the reaction rate can be promoted to the activity, providing the nitrogen oxide emission amount and the reducing agent supply amount to the catalyst layer.

In addition, when it is determined that the activity cannot be promoted, the method may further include controlling the nitrogen oxide emission amount and the reducing agent supply amount, or controlling a flow path direction of the nitrogen oxide emission amount and the reducing agent supply amount.

In addition, the step of controlling the nitrogen oxide emission amount and the reducing agent supply amount is a step of calculating the nitrogen oxide emission amount that can be promoted by the activity of the reaction rate, the nitrogen oxide emission information for the calculated nitrogen oxide emission amount learned Input to a neural network model, deriving driving characteristic information corresponding to the inputted nitrogen oxide emission information, adjusting current driving characteristic information with the derived driving characteristic information, and the obtained reducing agent supply amount, the It may include adjusting the reducing agent supply amount stoichiometrically corresponding to the calculated nitrogen oxide emission amount.

In addition, the driving characteristic information may include at least one of an exhaust gas amount, a combustion temperature, a combustion speed, and a load of the exhaust gas exhaust part related to the operation operation of the exhaust gas exhaust part.

In addition, the nitrogen oxide reduction system further includes an additional catalyst layer having the same function as the catalyst layer, and a path direction change unit for changing the flow direction of nitrogen oxides and reducing agents to the catalyst layer or the additional catalyst layer, And the step of controlling the flow path direction of the reducing agent supply amount, in order to promote the reaction rate for the reduction reaction of a part of the nitrogen oxide emission and the reducing agent supply amount to the measured activity of the catalyst layer, by controlling the path direction change part Flowing the part into the catalyst layer, and controlling the path direction change part to control the path direction change part so that the reaction rate for the reduction reaction of the residual amount of the nitrogen oxide discharge amount and the reducing agent supply amount is promoted by the activity of the additional catalyst layer. flowing to an additional catalyst bed.

The features and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings.

Prior to this, the terms or words used in the present specification and claims should not be construed in a conventional and dictionary meaning, and the inventor may properly define the concept of the term to describe his invention in the best way. It should be interpreted as meaning and concept consistent with the technical idea of the present invention based on the original side.

According to an embodiment of the present invention, by using a neural network model in which the correlation between nitrogen oxide emission information and driving characteristic information is machine-learned, specific driving characteristic information suitable for a specific NOx emission amount can be derived, and specific driving characteristic information to adjust the current driving characteristic information.

According to an embodiment of the present invention, the activity of the catalyst layer can be maximally used by generating a specific nitrogen oxide emission amount and a reducing agent supply amount based on the measured activity of the catalyst layer.

According to an embodiment of the present invention, by installing catalyst layers in each of the passages branching from each other, even when the activity life of one catalyst layer has expired, the other catalyst layer can be used immediately, and one catalyst layer can be replaced while the other catalyst layer is being used. Therefore, the nitrogen oxide reduction system can be continuously operated without having to be separately stopped for replacing the catalyst layer.

The accompanying drawings are included to provide a further understanding, and are incorporated in and constitute a part of this specification. The drawings illustrate one or more embodiment(s) and together with the description serve to explain the principles and operation of the various embodiments. As such, the present invention will be more fully understood from the following detailed description taken in conjunction with the accompanying drawings.
1 is a block diagram of a nitrogen oxide reduction system according to an embodiment of the present invention.
2 is a diagram schematically illustrating data of a driving characteristic information generating unit according to an embodiment of the present invention.
3 is a diagram illustrating a machine learning process of a neural network model according to an embodiment of the present invention.
4 is a view illustrating a gas from which nitrogen oxides are removed and discharged through a first catalyst layer of a catalyst unit according to an embodiment of the present invention.
5A and 5B are views illustrating a first catalyst layer and a second catalyst layer of a catalyst unit according to an embodiment of the present invention.
6 is a flowchart of a control method of a nitrogen oxide reduction system for reducing nitrogen oxides according to an embodiment of the present invention.

The objects, specific advantages and novel features of one embodiment of the present invention will become more apparent from the following detailed description and preferred embodiments taken in conjunction with the accompanying drawings. In the present specification, in adding reference numbers to the components of each drawing, it should be noted that only the same components are given the same number as possible even though they are indicated on different drawings. In addition, terms such as "one side", "the other side", "first", "second" are used to distinguish one component from another component, and the component is limited by the terms not. Hereinafter, in describing an embodiment of the present invention, detailed descriptions of related known technologies that may unnecessarily obscure the gist of an embodiment of the present invention will be omitted.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic diagram illustrating that exhaust gas containing nitrogen oxides is discharged through a nitrogen oxide reduction system 100 according to an embodiment of the present invention. Note that the solid line connecting each component in FIG. 1 indicates that each component is actually connected in order for the gas of nitrogen oxide and reducing agent described below to flow, and the dotted line connecting each component may represent a communication line or an electric line do.

Referring to FIG. 1 , it is a schematic diagram illustrating that nitrogen oxides of the exhaust gas discharged from the exhaust gas discharge unit 110 are removed and discharged to the stack 1000 .

The nitrogen oxide reduction system 100 includes an exhaust gas discharge unit 110 , an operation characteristic information generation unit 120 , a reduction supply unit 130 , a reduction reaction unit 140 , a catalyst unit 150 , and a control unit 160 . include

The exhaust gas discharge unit 110 may discharge an exhaust gas containing nitrogen oxides, etc. while burning a fossil fuel composed of at least one of carbon, hydrogen, oxygen, nitrogen, and sulfur. In addition, the exhaust gas discharge unit 110 may include a management device capable of measuring and monitoring at least one of an exhaust gas amount related to a driving operation, a combustion temperature, a combustion speed, and an exhaust gas discharge unit load, in order to calculate the NOx emission amount. can Furthermore, the exhaust gas discharge unit 110 includes at least one of an exhaust gas amount, a combustion temperature, a combustion speed, and an exhaust gas discharge unit load to the driving characteristic information generation unit 120 at the request of the driving characteristic information generation unit 120 . It is possible to provide operation characteristic information and nitrogen oxide emission information obtained based on operation characteristic information. The NOx emission information can represent a specific amount of NOx emission.

The driving characteristic information generator 120 may include a driving characteristic information collection unit 122 that collects driving characteristic information, and a nitrogen oxide emission information collection unit 124 that collects nitrogen oxide emission information. The driving characteristic information generating unit 120 additionally utilizes the driving characteristic information of the driving characteristic information collecting unit 122 and the nitrogen oxide emission information of the nitrogen oxide emission information collection unit 124 to correspond to the desired NOx emission information. In order to derive driving characteristic information, the learned neural network model 126 may be included.

2 is a driving characteristic information generating unit 120 including driving characteristic information 210 when an arbitrary fossil fuel is used and nitrogen oxide emission information 220 corresponding to the driving characteristic information according to an embodiment of the present invention. ) is a diagram schematically illustrating the data 200 of .

In the second to N+1th rows of the data 200 shown in FIG. 2 , the driving characteristic information 210-1 to 210-N and the corresponding nitrogen oxide emission information 220-1 to 220-N are obtained. is showing For example, in the second row, under the condition that the exhaust gas amount is 100,000 m3/h, the combustion temperature is 1,300° C., the combustion rate is 40 cm/sec, and the exhaust gas discharge part load is 70 MW, the exhaust gas discharge unit 110 is When operating, NOx emissions are measured at 500 ppm. The NOx emission corresponding to the conditions of various driving characteristics information is data collected and stored in the process of operating the exhaust gas emission unit 110 .

As described above, the exhaust gas amount of 100,000 m3/h, the combustion temperature of 1,300° C., the combustion rate of 40 cm/sec, and the exhaust gas exhaust load of 70 MW are one of the operating characteristic information 210 of the operating characteristic information 210. -1), and the nitrogen oxide emission of 500 ppm corresponding to the condition of the driving characteristic information may be one of the nitrogen oxide emission information 220 - 1 among the nitrogen oxide emission information 220 .

Referring back to FIG. 1 , the reduction supply unit 130 may supply a reducing agent (eg, ammonia) to the reduction reaction unit 140 to stoichiometrically correspond to nitrogen oxide in order to reduce nitrogen oxide. The reduction supply unit 130 may precisely spray the reducing agent with a plurality of injectors (or injectors) in order to supply the reducing agent from the storage unit (not shown) in which the reducing agent is stored. The reduction supply unit 130 may adjust the injection speed and injection pressure of the injector, the nozzle size of the injector, etc. in order to adjust the amount of the reducing agent to be sprayed.

The reduction reaction unit 140 may reduce the nitrogen oxide and the reducing agent. The reduction reaction unit 140 may measure and adjust the internal temperature and pressure so as to reduce the nitrogen oxide and the reducing agent. In addition, the reduction reaction unit 140 may include a turbulence providing device for sufficiently mixing the nitrogen oxide and the reducing agent.

The catalyst unit 150 is communicated from the reduction reaction unit 140 , so that the nitrogen oxide and the reducing agent mixed in the reduction reaction unit 140 may be supplied. The catalyst unit 150 may include a first catalyst layer 152 in order to accelerate the reaction rate for the reduction reaction of nitrogen oxide and the reducing agent under a predetermined temperature. According to an embodiment, the catalyst unit 150 may further include a second catalyst layer 154 having the same function as the first catalyst layer 152 (see FIGS. 5A and 5B ), and in this case, the first The catalyst layer 152 and the second catalyst layer 154 may be installed in each of the passages branching from each other along the reduction reaction unit 140 passage, as will be described below. Such a catalyst layer may be composed of a honeycomb type, a corrugated cardboard type, a plate type, and the like. The catalyst carrier of the catalyst layer may be ceramic, fiber and stainless steel. In addition, the active material enabling the activity of the catalyst layer may include, for example, TiO ₂ , V ₂ O ₅ , WO ₃ .

The activity of a catalyst may be an important design factor of a catalyst capable of removing nitrogen oxides. For example, the higher the activity of the catalyst, the more nitrogen oxides can be removed. On the other hand, the lower the activity of the catalyst, the less nitrogen oxides can be removed. As a result, the amount of nitrogen oxide removed may vary depending on the activity of the catalyst.

Such activity can be measured from the nitrogen oxide removal efficiency and the by-product generated in excess of the supply amount of the reducing agent, that is, ammonia slip. To this end, in order to measure the removal efficiency of nitrogen oxides and ammonia slip, before and after the first catalyst layer 152 and/or the second catalyst layer 154, the amount before and after the removal of nitrogen oxide can be sensed, It may include a plurality of sensors capable of measuring ammonia slip.

The control unit 160 includes all units of the nitrogen oxide reduction system 100 , that is, the exhaust gas discharge unit 110 , the operation characteristic information generation unit 120 , the reduction supply unit 130 , the reduction reaction unit 140 , and the catalyst unit ( 150) can be controlled in general. For example, the control unit 160, the exhaust gas discharge unit 110 measures at least one of the amount of exhaust gas, combustion temperature, combustion speed, and exhaust gas exhaust load, and calculates nitrogen oxide emissions based on these measurements, , driving characteristic information generating unit 120 collects driving characteristic information and nitrogen oxide emission information, and driving characteristic information corresponding to nitrogen oxide emission information for nitrogen oxide emission that can be processed (ie, removed) with usable activity , the reduction supply unit 130 provides a reducing agent supply amount stoichiometrically to the nitrogen oxide emission, the reduction reaction unit 140 measures the internal temperature and pressure, and the catalyst unit 150 measures the activity. control so that The control unit 160 may include a memory (not shown) capable of storing necessary information or data, or may be connected to such a memory. For example, the control unit 160 may obtain the supply amount of the reducing agent to stoichiometrically correspond to the nitrogen oxide emission and the nitrogen oxide emission based on the operation characteristic information of the exhaust gas discharge unit 110 and store it in the memory.

3 is a diagram illustrating a process in which the driving characteristic information generator 120 trains the neural network model 126 composed of a machine learning-based neural network under the control of the controller 160 according to an embodiment of the present invention. it is one drawing

Machine learning (machine learning) refers to a field that defines various problems dealt with in the field of artificial intelligence and studies methodologies to solve them. Machine learning is also defined as an algorithm that improves the performance of a certain task through constant experience.

An artificial neural network (ANN) is a model used in machine learning, and may refer to an overall model having problem-solving ability, which is composed of artificial neurons (nodes) that form a network by combining synapses. An artificial neural network may be defined by a connection pattern between neurons of different layers, a learning process that updates model parameters, and an activation function that generates an output value.

The artificial neural network may include an input layer, an output layer, and optionally one or more hidden layers. Each layer includes one or more neurons, and the artificial neural network may include neurons and synapses connecting neurons. In the artificial neural network, each neuron may output a function value of an activation function with respect to input signals input through a synapse, a weight, and a bias.

Model parameters refer to parameters determined through learning, and include the weight of synaptic connections and the bias of neurons. In addition, the hyperparameter refers to a parameter that must be set before learning in a machine learning algorithm, and includes a learning rate, the number of iterations, a mini-batch size, an initialization function, and the like.

The purpose of learning the artificial neural network can be seen as determining the model parameters that minimize the loss function. The loss function may be used as an index for determining optimal model parameters in the learning process of the artificial neural network.

Machine learning can be classified into supervised learning, unsupervised learning, and reinforcement learning according to a learning method.

Supervised learning refers to a method of learning an artificial neural network in a state where a label for learning information is given. can mean Unsupervised learning may refer to a method of learning an artificial neural network in a state where no label for learning information is given. Reinforcement learning can refer to a learning method in which an agent defined in an environment learns to select an action or sequence of actions that maximizes the cumulative reward in each state.

Among artificial neural networks, machine learning implemented as a deep neural network (DNN) including a plurality of hidden layers is also called deep learning (deep learning), and deep learning is a part of machine learning. Hereinafter, machine learning may be used in a sense including deep learning.

Referring to FIG. 3 , the driving characteristic information generating unit 120 uses each information of the collecting units 122 and 124 to obtain input data (Input Data, ID), which is nitrogen oxide emission information, and the nitrogen oxide emission information. Correct answer data (Answer Data, AD), which is driving characteristic information that can be calculated, can be generated as learning data (LD).

For example, when referring together with FIG. 2 , the nitrogen oxide emission information 220-1 to 220-N may be input data ID, and the driving characteristic information 210-1 to 210-N may include nitrogen oxides. It may be correct answer data AD corresponding to the oxide amount information 220-1 to 220-N.

The driving characteristic information generator 120 may train the neural network model 126 as the learning data LD. The driving characteristic information generating unit 120 inputs the correct answer data AD of the training data LD and compares the output data OD output by the neural network model with the correct answer data AD of the learning data LD. Thus, the neural network model can be trained repeatedly.

The driving characteristic information generating unit 120 may output specific driving characteristic information corresponding thereto when specific nitrogen oxide emission information is input to the neural network model using the learned neural network model.

As described above, when NOx emission information for NOx emission that can be processed (that is, removed) at the currently measured activity level is input to the learned neural network model, driving characteristic information corresponding to the NOx emission information is derived can be

4 is a gas discharged by removing nitrogen oxides and a reducing agent from the reduction reaction unit 140 through the first catalyst layer 152 of the catalyst unit 150 according to an embodiment of the present invention. It is the drawing shown.

As shown in FIG. 1 , the control unit 160 may measure the activity of the first catalyst layer 152 in the catalyst unit 150 , and based on the measured activity, the current nitrogen oxide emission amount and the reducing agent supply amount and It can be determined whether the reaction rate of

If the control unit 160 determines that the reaction rate can be accelerated with the measured activity, the reducing agent supply amount stoichiometrically corresponding to the current NOx emission and NOx emission is transferred to the catalyst unit 150 through the reduction reaction unit 140. can be provided as Accordingly, the provided nitrogen oxide amount and the supply amount of the reducing agent may be treated or removed in the first catalyst layer 152 . After that, the exhaust gas from which nitrogen oxides have been removed may be discharged to the stack 1000 .

On the other hand, if the control unit 160 determines that the reaction rate cannot be accelerated with the measured activity, the control unit 160 determines the amount of nitrogen oxide that can promote the reaction rate with the measured activity based on the measured activity. can be calculated Thereafter, the control unit 160 inputs nitrogen oxide emission information for the calculated nitrogen oxide emission into the learned neural network model 126 of the driving characteristic information generating unit 120, and driving characteristics corresponding to the nitrogen oxide emission information. information can be derived. The control unit 160 may control the exhaust gas discharge unit 110 to operate according to the derived driving characteristic information. The exhaust gas discharge unit 110 driven by the derived driving characteristic information may discharge nitrogen oxides as much as the calculated amount of nitrogen oxides. The control unit 160 may adjust the supply amount of the reducing agent to correspond stoichiometrically to the calculated nitrogen oxide emission.

When the controller 160 determines that the reaction speed cannot be accelerated by the measured activity, the controller 160 adjusts the current driving characteristic information to the neural network model 126 described above according to another embodiment of the present invention. Without it, the current NOx emissions can be treated. To this end, the catalyst unit 150 may further include a second catalyst layer 154 . A detailed description thereof will be described with reference to FIGS. 5A and 5B.

5A and 5B are views illustrating the catalyst unit 150 connected to the reduction reaction unit 140 passage according to an embodiment of the present invention.

The catalyst unit 150 includes a first passage 151 and a second passage 153 branched from each other along the passage of the reduction reaction unit 140, and a first catalyst layer 152 and a second catalyst layer ( 154) may be included. The catalyst unit 150 may further include a path direction change unit 156 for changing the directions of the nitrogen oxide and the reducing agent flowing from the reduction reaction unit 140 .

As described above, when the control unit 160 determines that the reaction rate cannot be accelerated by the measured activity of the first catalyst layer 152 , the control unit 160 controls the path direction change unit 156 to control nitrogen oxides. Only an amount capable of accelerating the reaction rate with the measured activity among the amount of emission and the amount of reducing agent supplied is flowed into the first catalyst layer 152, and the reaction rate of the remaining amount of the amount of nitrogen oxide and the amount of supply of reducing agent is promoted to the activity of the second catalyst layer 154. For this, the remaining amount may be flowed into the second catalyst layer 154 .

Specifically, when the controller 160 determines that the reaction rate cannot be accelerated by the measured activity of the first catalyst layer 152, the controller 160 adjusts the reaction rate to the measured activity based on the measured activity. Calculate the amount of nitrogen oxide emission and the reducing agent supply amount that can be promoted, and only the calculated nitrogen oxide emission amount and the reducing agent supply amount can flow into the first catalyst layer 152 by controlling the path direction change unit 156 (FIG. 5A) refer), the remaining nitrogen oxide emission amount and the reducing agent supply amount (= 'current nitrogen oxide emission amount and reducing agent supply amount' - 'the nitrogen oxide emission amount and reducing agent supply amount flowing into the first catalyst layer 152) The reaction rate of the remaining amount is calculated as the second catalyst layer ( 154), the remaining amount may be flowed into the second catalyst layer 154 by controlling the path direction change unit 156 (see FIG. 5B).

Referring to FIG. 5A , nitrogen oxides and a part of the reducing agent among the nitrogen oxides and the reducing agent flowed from the reduction reaction unit 140 pass through the first catalyst layer 152 of the catalyst unit 150 to remove the nitrogen oxides to the stack 1000. can be emitted.

Referring to FIG. 5B , the remaining nitrogen oxides and reducing agent among the nitrogen oxides and reducing agents flowing from the reduction reaction unit 140 pass through the second catalyst layer 154 by the change of the path direction change unit 156 to remove nitrogen oxides. It may be discharged to the stack (1000).

In one embodiment of the present invention, a blocking device is included to block the gas flow to the first passageway 151 or the second passageway 153, but this is only an example, and a device capable of changing the gas flow direction, for example For example, of course, it may include a valve device such as a three-way valve.

6 is a flowchart of a control method of a nitrogen oxide reduction system for reducing the nitrogen oxides by reducing the nitrogen oxides and the reducing agent according to an embodiment of the present invention. Such a control method may be performed by the controller 160 .

First, in step S610 , the nitrogen oxide emission amount may be obtained and stored in the memory based on the operation characteristic information of the exhaust gas emission unit 110 . As described above, the driving characteristic information 210 is information related to the operation of the exhaust gas discharge unit, and may include at least one of an exhaust gas amount according to a fuel type, a combustion temperature, a combustion speed, and an exhaust gas discharge unit load.

Then, in step S620, the reducing agent supply amount is obtained stoichiometrically corresponding to the nitrogen oxide emission obtained in step S610 can be stored in the memory. In step S630, the activity of the first catalyst layer 152 is measured, and the reaction rate for the reduction reaction is currently measured when the amount of nitrogen oxide and the supply amount of the reducing agent (or the amount of the mixed gas) is reduced. It can be determined whether activity can be promoted. When it is determined that the reaction rate can be promoted by the activity, in step S640, the amount of nitrogen oxide discharged and the supply amount of the reducing agent may be provided to the catalyst layer.

If it is determined in step S630 that the reaction rate cannot be promoted to activity, the method may proceed to step S650.

In step S650, the nitrogen oxide emission amount and the reducing agent supply amount may be controlled, or the flow path direction of the nitrogen oxide emission amount and the reducing agent supply amount may be controlled. When the nitrogen oxide emission amount and the reducing agent supply amount are controlled, the method proceeds to step S652, in which the nitrogen oxide emission amount at which the reaction rate can be promoted with activity can be calculated. Thereafter, in step S654 , the NOx emission information for the calculated NOx emission is input to the trained neural network model 126 , and driving characteristic information corresponding to the inputted NOx emission information may be derived. Since the description of the neural network model 126 has been previously described, a detailed description thereof will be omitted. The driving characteristic information may include at least one of an exhaust gas amount related to a driving operation of the exhaust gas exhaust unit, a combustion temperature, a combustion speed, and an exhaust gas discharge unit load. In operation S656, current driving characteristic information may be adjusted with the derived driving characteristic information. Then, in step S658, the reducing agent supply amount obtained from step S620 may be adjusted to the reducing agent supply amount stoichiometrically corresponding to the nitrogen oxide emission calculated from step S652.

In step S650, if the flow path direction of the nitrogen oxide emission amount and the reducing agent supply amount is controlled, the method proceeds to the S670 step, in which the reaction rate for the reduction reaction of some of the nitrogen oxide emission amount and the reducing agent supply amount is measured. In order to promote the activity of the first catalyst layer 152 , the path direction change unit 156 may be controlled so that a portion thereof may flow into the first catalyst layer 152 . In other words, based on the measured activity of the first catalyst layer 152, the amount of nitrogen oxide emission and reducing agent supply amount that can promote the reaction rate for the reduction reaction with the measured activity of the first catalyst layer 152 is calculated and only the calculated nitrogen oxide amount and the reducing agent supply amount may flow to the first catalyst layer 152 under the control of the path direction change unit 156 . After that, in step S672, reduction of the nitrogen oxide emission amount and the reducing agent supply amount, which are the remaining amounts (= 'current nitrogen oxide emission amount and reducing agent supply amount' - 'the nitrogen oxide emission amount and reducing agent supply amount flowing into the first catalyst layer 152) other than the part In order to promote the reaction rate for the reaction with the activity of the second catalyst layer 154 , the path direction change unit 156 may be controlled so that the remaining amount may flow to the second catalyst layer 154 .

Although the present invention has been described in detail through specific examples, it is intended to describe the present invention in detail, and the present invention is not limited thereto, and by those of ordinary skill in the art within the technical spirit of the present invention. It is clear that the modification or improvement is possible.

All simple modifications or changes of the present invention fall within the scope of the present invention, and the specific scope of protection of the present invention will be made clear by the appended claims.

100: nitrogen oxide reduction system 110: exhaust gas discharge unit
120: driving characteristic information generating unit 122: driving characteristic information collecting unit
124: nitrogen oxide emission information collection unit 126: neural network model
130: reduction supply unit 140: reduction reaction unit
150: catalyst unit 152: first catalyst layer
154: second catalyst layer 160: control unit
200: data of the driving characteristic information generation unit

Claims (10)

In the nitrogen oxide reduction system for reducing the nitrogen oxide by reducing the nitrogen oxide and the reducing agent,
an exhaust gas discharge unit for discharging exhaust gas containing nitrogen oxides;
a reduction supply unit for supplying a reducing agent to reduce the nitrogen oxide;
a reduction reaction unit for reducing the nitrogen oxide and the reducing agent;
a catalyst unit including a first catalyst layer for accelerating the reaction rate for the reduction reaction; and
Measuring the activity of the first catalyst layer, based on the measured activity, the control unit for controlling the amount of nitrogen oxide emission of the exhaust gas discharge unit and controlling the supply amount of the reducing agent to correspond stoichiometrically to the amount of nitrogen oxide; ,
Further comprising a driving characteristic information generating unit including a neural network model in which driving characteristic information and nitrogen oxide emission information are used to control the NOx emission,
The driving characteristic information generating unit generates the input data (Input Data, ID) that is the nitrogen oxide emission information and the correct answer data (Answer Data, AD) that is the driving characteristic information as learning data (Learning Data, LD), and the learning Input the correct answer data AD of the data LD, and the input data (Input Data, ID) is input to the neural network model and output data (Output Data, OD) and the correct answer of the learning data (LD) Iteratively trains the neural network model by comparing the data AD,
When the control unit determines that the reaction rate cannot be accelerated by the measured activity of the first catalyst layer,
When the nitrogen oxide emission information is input to the learned neural network model, the control unit outputs driving characteristic information corresponding to the nitrogen oxide emission information, and adjusts current driving characteristic information with the output driving characteristic information, nitrogen Oxide reduction system. The method according to claim 1,
The driving characteristic information generating unit
Further comprising: a driving characteristic information collection unit for collecting the driving characteristic information; and a nitrogen oxide emission information collection unit for collecting the nitrogen oxide emission information;
The driving characteristic information is
The nitrogen oxide reduction system, comprising at least one of an exhaust gas amount, combustion temperature, combustion speed, and exhaust gas exhaust load related to the operation operation of the exhaust gas exhaust part. The method according to claim 1,
The activity is
The nitrogen oxide removal efficiency and the nitrogen oxide reduction system, which is measured from the by-products generated in excess of the supply amount of the reducing agent. The method according to claim 1,
When the controller determines that the reaction rate cannot be accelerated by the measured activity of the first catalyst layer,
the control unit
Calculate the amount of nitrogen oxide that can promote the reaction rate only with the measured activity of the first catalyst layer, and input the information on the amount of nitrogen oxide on the calculated amount of nitrogen oxide into the learned neural network model, and the amount of nitrogen oxide By deriving operating characteristic information corresponding to the information, the current operating characteristic information is adjusted with the derived operating characteristic information, and the reducing agent supply amount is adjusted to correspond stoichiometrically to the calculated nitrogen oxide emission amount, a nitrogen oxide reduction system . delete In the control method of a nitrogen oxide reduction system for reducing the nitrogen oxide by reducing the nitrogen oxide and a reducing agent,
obtaining nitrogen oxide emissions based on the operation characteristic information of the exhaust gas emission unit;
obtaining a reducing agent supply amount stoichiometrically corresponding to the nitrogen oxide emission;
measuring the activity of the catalyst layer, and determining whether the reaction rate for the reduction reaction can be accelerated to the activity when the amount of nitrogen oxide discharged and the supply amount of the reducing agent is reduced;
providing the nitrogen oxide emission amount and the reducing agent supply amount to the catalyst layer when it is determined that the reaction rate can be promoted to the activity; and
When it is determined that the activity cannot be promoted, the nitrogen oxide emission information for the nitrogen oxide emission is input to the trained neural network model, driving characteristic information corresponding to the input nitrogen oxide emission information is output, and the outputted In order to adjust the current driving characteristics information to the driving characteristics information, the step of controlling the amount of nitrogen oxide emission and the reducing agent supply;
The learned neural network model compares the input data (Input Data, ID), which is the nitrogen oxide emission information, into the neural network model and output data (Output Data, OD), and the correct answer data (AD), which is the driving characteristic information. A method of controlling a nitrogen oxide reduction system that has been repeatedly learned by doing so. 7. The method of claim 6,
The activity is
The nitrogen oxide removal efficiency and the control method of the nitrogen oxide reduction system, which is measured from the by-products generated in excess of the supply amount of the reducing agent. 7. The method of claim 6,
The step of controlling the nitrogen oxide emission amount and the reducing agent supply amount,
calculating an amount of nitrogen oxides in which the reaction rate can be promoted with the activity;
inputting nitrogen oxide emission information for the calculated nitrogen oxide emission amount into the learned neural network model, and deriving driving characteristic information corresponding to the input nitrogen oxide emission information;
adjusting current driving characteristic information with the derived driving characteristic information; and
Adjusting the obtained reducing agent supply amount to the reducing agent supply amount stoichiometrically corresponding to the calculated nitrogen oxide emission; including; a control method of a nitrogen oxide reduction system. 7. The method of claim 6,
The driving characteristic information is
The control method of a nitrogen oxide reduction system, comprising at least one of an exhaust gas amount, a combustion temperature, a combustion speed, and an exhaust gas exhaust load related to the operation of the exhaust gas exhaust part. delete

Images