Ammonia injection amount optimization control device and control method based on scr denitration system

Abstract

The application discloses an ammonia injection amount optimization control device and method based on an SCR denitration system, which comprises an ammonia injection amount prediction system, a protection logic and an SCR denitration system. The ammonia injection amount prediction system comprises at least one neural network model, the prediction results of the ammonia injection amount are calculated through mutual calling of the prediction results of each model. The protection logic is used for judging the prediction results of the ammonia injection amount, generating an ammonia injection amount prediction value signal and preventing the ammonia injection amount from being substandard. The SCR denitration system controls the ammonia injection amount of the SCR denitration reactor according to the ammonia injection amount prediction value signal. Through the ammonia injection amount optimization control method provided by the application, the NO x export emission is up to standard, the optimal ammonia injection amount is obtained, the ammonia utilization rate is improved, the ammonia escape is reduced, and the stability of the operation of the denitration system is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of thermal power generation, in particular to a device and method for optimizing ammonia injection amount based on an SCR denitration system. BACKGROUND

[0002] At present, selective catalytic reduction (SCR) technology is widely used in domestic coal-fired units to reduce the value of nitrogen oxides (NO x ) in flue gas, so as to achieve the goal of environmental protection and emission reduction. The basic principle of SCR technology is to convert NO x in flue gas into non-polluting N2 and H2O through oxidation-reduction reaction. The SCR reaction process is complex, and the denitration efficiency is affected by many factors, such as ammonia injection amount, reaction temperature, flue gas velocity, etc., among which the main factor affecting the efficiency is the ammonia injection amount, which is determined according to the obtained inlet NO x concentration. Due to the dynamic characteristics of large inertia, large delay and strong disturbance of the SCR system, when the combustion condition changes, the inlet NO x value will fluctuate greatly, and there is a risk that the ammonia injection amount is inaccurate and not timely, resulting in large fluctuations in outlet NO x concentration and difficulty in controlling denitration efficiency. SUMMARY

[0003] Therefore, the present application provides a device and method for optimizing ammonia injection amount based on an SCR denitration system, which overcomes the defects of inaccurate and untimely ammonia injection amount in the prior art, resulting in large fluctuations in outlet NO x concentration and difficulty in controlling denitration efficiency.

[0004] To achieve the above purpose, the present application provides the following technical solutions:

[0005] In a first aspect, the present application provides a device for optimizing ammonia injection amount based on an SCR denitration system, comprising: an ammonia injection amount prediction system, a protection logic, and an SCR denitration system.

[0006] The ammonia injection amount prediction system comprises at least one neural network model, and the prediction results of each model are mutually called to calculate the prediction results of ammonia injection amount.

[0007] The protection logic is used to judge the prediction results of ammonia injection amount, generate an ammonia injection amount prediction value signal, and prevent the ammonia injection amount from being substandard.

[0008] The SCR denitration system controls the ammonia injection amount of the SCR denitration reactor according to the ammonia injection amount prediction value signal.

[0009] Optionally, the ammonia injection amount prediction system comprises: 1 inlet NO x prediction model and 3 ammonia injection amount prediction models;

[0010] The 3 ammonia injection amount prediction models are ammonia injection amount prediction model 1, ammonia injection amount prediction model 2 and ammonia injection amount prediction model 3 respectively.

[0011] The ammonia injection amount prediction model 1, the ammonia injection amount prediction model 2 and the ammonia injection amount prediction model 3 are connected in series and are connected in parallel with the inlet NO x prediction model.

[0012] Optionally, the inlet NO x prediction model comprises: 1 input layer, a plurality of hidden layers and 1 output layer; wherein the input layer contains 5 neurons; each hidden layer contains a plurality of neurons; and the output layer contains 1 neuron.

[0013] Optionally, the ammonia injection amount prediction model comprises: 1 input layer, a plurality of hidden layers and 1 output layer; wherein the input layer contains 7 neurons; each hidden layer contains a plurality of neurons; and the output layer contains two neurons.

[0014] Optionally, the input parameters of the ammonia injection amount prediction model 1 are: the load, the oxygen amount, the total fuel amount, the total air amount, the inlet NO x concentration value, the outlet NO x concentration set value and the ammonia injection amount at the current time (t); and the output parameters are the ammonia injection amount prediction value and the outlet NO x concentration prediction value at the time (t+1).

[0015] The input parameters of the ammonia injection amount prediction model 2 are: the load, the oxygen amount, the total fuel amount, the total air amount, the inlet NO x concentration prediction value, the outlet NO x concentration set correction value and the ammonia injection amount prediction value at the time (t+1); and the output parameters are the ammonia injection amount prediction value and the outlet NO x concentration prediction value at the time (t+2).

[0016] The input parameters of the ammonia injection amount prediction model 3 are: the load, the oxygen amount, the total fuel amount, the total air amount, the inlet NO x concentration prediction value, the outlet NO x concentration set correction value and the ammonia injection amount prediction value at the time (t+2); and the output parameters are the ammonia injection amount prediction value and the outlet NO x concentration prediction value at the time (t+3).

[0017] Optionally, the inlet NO xThe historical data for training the prediction model and the ammonia injection amount prediction model are preprocessed before training.

[0018] Optionally, the inlet NO x The neural network model for predicting the ammonia injection amount and the outlet NOx concentration value has an absolute deviation between the predicted value and the actual value of no more than 8%.

[0019] The neural network model for predicting the ammonia injection amount and the outlet NOx concentration value has an absolute deviation between the predicted value and the actual value of no more than 8%.

[0020] In a second aspect, the embodiments of the present application provide an ammonia injection amount optimization control method based on a denitration system, and the optimization control method is based on the SCR denitration system ammonia injection amount optimization control device of the first aspect, and the optimization control method comprises the following steps.

[0021] Step 1: According to the historical data of the actual operation of the thermal power generating unit, a neural network model for predicting the inlet NO x concentration and three neural network models for predicting the ammonia injection amount are constructed.

[0022] Step 2: According to the real-time operation data at the current time (t), the inlet NO x concentration prediction model, the inlet NO x concentration prediction value at the time (t+1) is generated; according to the real-time operation data at the time (t), the outlet NO x concentration setpoint at the time (t+1) is generated; according to the ammonia injection amount and the outlet NO x concentration prediction model 1, the outlet NO x concentration prediction value at the time (t+1) and the ammonia injection amount prediction value are generated.

[0023] Step 3: According to the real-time operation data at the time (t), the inlet NO x concentration prediction value at the time (t+1), the outlet NO x concentration setpoint correction value at the time (t+2) and the ammonia injection amount prediction value, the ammonia injection amount prediction model 2 is used to generate the ammonia injection amount prediction value at the time (t+2).

[0024] Step 4: According to the real-time operation data at the time (t), the inlet NO x concentration prediction value at the time (t+1), the outlet NO x concentration setpoint correction value at the time (t+3) and the ammonia injection amount prediction value, the ammonia injection amount prediction model 3 is used to generate the ammonia injection amount optimization value at the time (t+3) and the outlet NO x concentration prediction value.

[0025] Step 5: The SCR denitration system controls the ammonia injection amount of the SCR denitration reactor according to the ammonia injection amount prediction value provided by the prediction model.

[0026] Optionally, the exit NO at time (t+1) in step 2 can be calculated using the following formula. x Concentration setpoint:

[0027] date Std (t+1) = date Std -(date NOx -date Std )×α

[0028] Where, date Std (t+1) represents the exit NO at time (t+1). x Concentration setpoint, date Std For export NO x Concentration meets the set value, date Nox Let NO be the exit at time (t). x The actual concentration value, where α is a set parameter.

[0029] Optionally, the exit NO at time (t+2) in step 3 can be calculated using the following formula. x Concentration setting correction value:

[0030] date Std (t+2) = date Std (t+1)-(date sim (t+1)-date Std (t+1))×β

[0031] Where, date Std (t+2) represents the exit NO at time (t+2). x Concentration setting correction value, date Std (t+1) represents the exit NO at time (t+1). x Concentration setpoint, date Sim (t+1) represents the exit NO at time (t+1). x The concentration prediction value, where β is the first preset parameter.

[0032] Optionally, the exit NO at time (t+3) in step 4 can be calculated using the following formula. x Concentration setting correction value:

[0033] date Std (t+3) = date Std (t+2)-(date sim (t+2)-date Std (t+2))×ε

[0034] Where, date Std (t+3) represents the exit NO at time (t+3). xConcentration setting correction value, date Std (t+2) is the outlet NOx concentration at time (t+2) x Concentration setting correction value, date Sim (t+2) is the outlet NOx concentration at time (t+2) x Concentration prediction value, ε is a second preset parameter.

[0035] The technical scheme of the present application has the following advantages:

[0036] The ammonia injection amount optimization control device and control method based on the SCR denitration system provided by the present application ensure that the outlet NOx concentration is up to standard while obtaining the optimal ammonia injection amount, which can improve the ammonia utilization rate, reduce ammonia escape, and improve the stability of the denitration system operation. x The ammonia injection amount optimization control device and control method based on the SCR denitration system provided by the present application ensure that the outlet NOx concentration is up to standard while obtaining the optimal ammonia injection amount, which can improve the ammonia utilization rate, reduce ammonia escape, and improve the stability of the denitration system operation. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the specific embodiments of the present application or the technical scheme in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0038] Figure 1 A component diagram of one specific example of an ammonia injection amount optimization control device based on the SCR denitration system provided by the present application;

[0039] Figure 2 A flowchart of one specific example of an ammonia injection amount optimization control method provided by the present application. DETAILED DESCRIPTION

[0040] The technical scheme of the present application will be described in detail below with reference to the drawings. Obviously, the described embodiments are some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0041] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0042] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components; and they can refer to a wireless connection or a wired connection. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0043] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0044] Example 1

[0045] This invention provides an optimized ammonia injection control device based on an SCR denitrification system, such as... Figure 1 As shown, the system includes: an ammonia injection rate prediction system, protection logic, and an SCR denitrification system. The ammonia injection rate prediction system includes at least one neural network model, for example, it can consist of four neural network models. Multi-step prediction and optimization calculation of the ammonia injection rate are achieved through the mutual calling of the prediction results of the four models. This is just an example and not a limitation; in practical applications, the appropriate number of neural network models should be selected according to actual needs. The protection logic is used to judge the predicted ammonia injection rate and generate a predicted ammonia injection rate signal to prevent the ammonia injection rate from falling short of the standard. "Failure to meet the standard" means preventing the ammonia injection rate from being too high or too low. The values ​​for "too high" or "too low" are not limited here and are selected according to the actual situation. The SCR denitrification system controls the ammonia injection rate of the SCR denitrification reactor based on the predicted ammonia injection rate signal. This invention guarantees that the ammonia injection rate in the NO... x While achieving compliance with export emission standards, obtaining the optimal ammonia injection rate can improve ammonia utilization, reduce ammonia escape, and enhance the operational stability of the denitrification system.

[0046] In this embodiment of the invention, the ammonia injection quantity prediction system includes: one inlet NO... xThe prediction model and the three ammonia injection amount prediction models are connected in series. x The prediction model and the three ammonia injection amount prediction models are connected in parallel.

[0047] In the embodiment of the present application, the prediction model for the inlet NO x The prediction model uses the parameters having a strong correlation with the change of the inlet NO x concentration as input parameters, uses historical data to train the model, and constructs the prediction model for the inlet NO x concentration and the parameters having a strong correlation with the change of the inlet NO x concentration value; the input parameters of the neural network model are the load, oxygen amount, total fuel amount, total air amount, inlet NO x concentration value, and the output parameter is the inlet NO x concentration value. In order to further improve the prediction accuracy of the prediction model, the difference between the predicted value and the actual value of the previous moment is used to correct the predicted value of the inlet NO x concentration of the next moment.

[0048] In the embodiment of the present application, the prediction model for the inlet NO x The prediction model comprises one input layer, a plurality of hidden layers and one output layer. The input layer comprises five neurons; each hidden layer comprises a plurality of neurons; and the output layer comprises one neuron. For example, the prediction model comprises one input layer, one to three hidden layers and one output layer; the input layer comprises five neurons, i.e. five input parameters; each hidden layer comprises ten to fifteen neurons; and the output layer comprises one neuron, i.e. one output parameter. The above is only an example, and the number of hidden layers and the number of neurons can be selected according to actual needs in practical applications.

[0049] In the embodiment of the present application, the prediction model for the inlet NO x The prediction model comprises one input layer, a plurality of hidden layers and one output layer. The input layer comprises five neurons; each hidden layer comprises a plurality of neurons; and the output layer comprises one neuron. For example, the prediction model comprises one input layer, one to three hidden layers and one output layer; the input layer comprises five neurons, i.e. five input parameters; each hidden layer comprises ten to fifteen neurons; and the output layer comprises one neuron, i.e. one output parameter. The above is only an example, and the number of hidden layers and the number of neurons can be selected according to actual needs in practical applications.

[0050] In this embodiment of the invention, the input parameters of the ammonia injection rate prediction model 1 are: the load, oxygen content, total fuel quantity, total air volume, and inlet NO at the current time (t). x Concentration value, export NO x The concentration setpoint and ammonia injection rate are set, and the output parameters are the predicted ammonia injection rate and outlet NO at time (t+1). x Predicted concentration values.

[0051] In this embodiment of the invention, the input parameters of the ammonia injection quantity prediction model 2 are: the load, oxygen quantity, total fuel quantity, and total air volume at the current time (t), and the inlet NO at time (t+1). x Concentration forecast, export NO x The concentration setting correction value and the predicted ammonia injection rate at time (t+1) are used as the output parameters, which are the predicted ammonia injection rate at time (t+2) and the outlet NO. x Concentration prediction.

[0052] In this embodiment of the invention, the input parameters of the ammonia injection quantity prediction model 3 are: the load, oxygen quantity, total fuel quantity, and total air volume at the current time (t), and the inlet NO at time (t+1). x Predicted concentration value, export NO x The concentration setting correction value and the predicted ammonia injection rate at time (t+2) are used as the output parameters, which are the predicted ammonia injection rate at time (t+3) and the outlet NO. x Predicted concentration values.

[0053] In this embodiment of the invention, the historical data used for training the neural network prediction model for inlet NOx concentration and the ammonia injection prediction model are preprocessed before training to eliminate the time lag between input parameters and output parameters caused by equipment or other reasons. The time lag correction range is estimated to be 1 minute to 3 minutes.

[0054] In this embodiment of the invention, the NO inlet is used x For the predictive neural network model, the absolute value of the deviation between the predicted and actual values ​​is no greater than 5%. For the neural network model used to predict ammonia injection rate and outlet NOx concentration, the absolute value of the deviation between the predicted and actual values ​​is no greater than 8%.

[0055] In this embodiment of the invention, the ammonia injection rate optimization control system sets upper and lower limits for the increase or decrease in the predicted ammonia injection rate between (t+1) and (t+2), between (t+1) and (t+3), and between (t+2) and (t+3). These limits are obtained through analysis of historical ammonia injection rate data and are used to prevent sudden changes in outlet NOx concentration caused by excessively high or low ammonia injection rates due to decreased accuracy of the predicted ammonia injection rate. The SCR denitrification system controls the ammonia injection rate of the SCR denitrification reactor based on the predicted ammonia injection rate signal.

[0056] The ammonia injection amount optimization control device based on the SCR denitration system provided by the embodiment of the application ensures that the NO x While meeting the export emission standard, the optimal ammonia injection amount is obtained, the ammonia utilization rate is improved, the ammonia escape is reduced, and the stability of the denitration system operation is improved.

[0057] Embodiment 2

[0058] The ammonia injection amount optimization control method provided by the embodiment of the application is based on the ammonia injection amount optimization control device of the SCR denitration system in the embodiment 1, and includes the following steps as shown in the figure. Figure 2 The optimization control method includes the following steps as shown in the figure.

[0059] Step S1: According to the historical data of the actual operation of the thermal power generating unit, a neural network model for predicting the inlet NO x concentration and three neural network models for predicting the ammonia injection amount are constructed.

[0060] Step S2: According to the real-time operation data at the current time (t), the inlet NO x concentration prediction model, the predicted value of the inlet NO x concentration at the time (t+1) is generated; according to the real-time operation data at the time (t), the outlet NO x concentration set value at the time (t+1), the ammonia injection amount and the outlet NO x concentration prediction model 1, the predicted value of the outlet NO x concentration at the time (t+1) and the predicted value of the ammonia injection amount are generated.

[0061] In the embodiment of the application, the outlet NO x concentration set value at the time (t+1) is calculated by the following formula in step 2:

[0062] date Std (t+1) = date Std -(date NOx -date Std )×α

[0063] Wherein, date Std (t+1) is the outlet NO x concentration set value at the time (t+1), date Std is the outlet NO x concentration standard set value, date Nox is the outlet NO x concentration actual value at the time (t), and α is a set parameter.

[0064] Step S3: According to the real-time operation data at the time (t), the inlet NOx concentration prediction value, outlet NO concentration at (t+2) time x concentration setting correction value and ammonia injection amount prediction value, the ammonia injection amount prediction model 2 is used to generate the ammonia injection amount prediction value at (t+2) time.

[0065] In the embodiment of the present application, the outlet NO concentration at (t+2) time in step 3 is calculated by the following formula x concentration setting correction value:

[0066] date Std (t+2) = date Std (t+1) - (date sim (t+1) - date Std (t+1)) x beta

[0067] wherein, date Std (t+2) is the outlet NO concentration at (t+2) time x concentration setting correction value, date Std (t+1) is the outlet NO concentration at (t+1) time x concentration setting value, date Sim (t+1) is the outlet NO concentration at (t+1) time x concentration prediction value, beta is the first preset parameter.

[0068] Step S4: according to the real-time running data at (t) time, the inlet NO concentration at (t+1) time x concentration prediction value, outlet NO concentration at (t+3) time x concentration setting correction value and ammonia injection amount prediction value, the ammonia injection amount prediction model 3 is used to generate the ammonia injection amount optimization value and the outlet NO concentration at (t+3) time x concentration prediction value.

[0069] In the embodiment of the present application, the outlet NO concentration at (t+3) time in step 4 is calculated by the following formula x concentration setting correction value:

[0070] date Std (t+3) = date Std (t+2) - (date sim (t+2) - date Std (t+2)) x epsilon

[0071] wherein, date Std (t+3) is the outlet NO concentration at (t+3) time x concentration setting correction value, date Std (t+2) is the outlet NO concentration at (t+2) time x concentration setting correction value, dateSim (t+2) is the outlet NOx concentration at time (t+2) x The concentration prediction value is a second preset parameter.

[0072] Step S5: The SCR denitration system controls the ammonia injection amount of the SCR denitration reactor according to the ammonia injection amount prediction value provided by the prediction model.

[0073] The ammonia injection amount optimization control method provided in the embodiments of the present application ensures that the NOx concentration at the outlet of the SCR denitration system is less than 200 mg / m3, the ammonia injection amount is optimal, the ammonia utilization rate is improved, the ammonia escape is reduced, and the stability of the operation of the denitration system is improved. x The outlet emission is up to standard, the optimal ammonia injection amount is obtained, the ammonia utilization rate is improved, the ammonia escape is reduced, and the stability of the operation of the denitration system is improved.

[0074] Obviously, the above embodiments are merely examples for clear illustration, rather than limitation on the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments need not and cannot be exhausted. The obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. An ammonia injection amount optimization control device based on an SCR denitration system, characterized by, Comprise: Ammonia injection amount prediction system, protection logic, SCR denitration system; The ammonia injection amount prediction system comprises at least one neural network model, and the prediction result of the ammonia injection amount is calculated through mutual calling of the prediction result of each model; The protection logic is used for judging the prediction result of the ammonia injection amount, generating an ammonia injection amount prediction value signal, and preventing the ammonia injection amount from being substandard; The SCR denitration system controls the ammonia injection amount of the SCR denitration reactor according to the ammonia injection amount prediction value signal; The neural network model comprises 1 inlet NO x The prediction model and 3 ammonia injection amount prediction models The three ammonia injection amount prediction models are ammonia injection amount prediction model 1, ammonia injection amount prediction model 2 and ammonia injection amount prediction model 3; The ammonia injection amount prediction model 1, the ammonia injection amount prediction model 2 and the ammonia injection amount prediction model 3 are connected in series in sequence, and are connected in parallel with the inlet NO x prediction model are connected in parallel; The input parameters for the ammonia injection rate prediction model 1 are: the current time ( t Load, oxygen content, total fuel consumption, total air volume, and inlet NO x Concentration value, ( t +1) Exit NO at time x Concentration setpoint and ( t The ammonia injection rate at time () is given by the output parameter (). t +1) Predicted ammonia injection rate and outlet NO at time [time missing] x Concentration prediction; The input parameters for the ammonia injection rate prediction model 2 are: the current time ( t The load, oxygen content, total fuel quantity, and total air volume of ( ) t +1) Time Entry NO x Concentration prediction value, ( t +2) Exit NO at time NO x Concentration setting correction value and ( t The predicted ammonia injection rate at time +1 is given, and the output parameter is ( t +2) Predicted ammonia injection rate and outlet NO at time [time missing] x Concentration prediction; The input parameters for ammonia injection rate prediction model 3 are: the current time ( t The load, oxygen content, total fuel quantity, and total air volume of ( ) t +1) Time Entry NO x Concentration prediction value, ( t +3) Exit NO at time NO x Concentration setting correction value and ( t +2) Predicted ammonia injection rate at time 1, output parameter is ( t +3) Predicted ammonia injection rate and outlet NO at time [time missing] x Predicted concentration values. 2.The ammonia injection amount optimization control device based on an SCR denitration system according to claim 1, characterized in that, The inlet NO x The prediction model comprises: 1 input layer, a plurality of hidden layers and 1 output layer; wherein the input layer contains 5 neurons; each hidden layer contains a plurality of neurons; and the output layer contains 1 neuron. 3.The SCR-based denitration system ammonia injection amount optimization control device according to claim 1, characterized in that, Each of the ammonia injection amount prediction models comprises an input layer, a plurality of hidden layers and an output layer; wherein the input layer comprises 7 neurons; each hidden layer comprises a plurality of neurons; and the output layer comprises two neurons.

4. The ammonia injection amount optimization control device for an SCR denitration system according to claim 3, characterized by, For inlet NO x The historical data for the prediction model and the ammonia injection amount prediction model are preprocessed before training.

5. The ammonia injection amount optimization control device for an SCR denitration system according to claim 4, characterized by, Inlet NO x a prediction model whose absolute value of deviation between predicted value and actual value is not greater than 5%; The absolute value of the deviation between the prediction value and the actual value of the ammonia injection amount prediction model 1, the ammonia injection amount prediction model 2 and the ammonia injection amount prediction model 3 is not greater than 8%.

6. An ammonia injection amount optimization control method characterized by, The ammonia injection amount optimization control device based on the SCR denitration system according to any one of claims 1-5, the optimization control method comprises: Step 1: According to the historical data of the actual operation of the thermal power generating unit, a prediction model for predicting the inlet NO x concentration of the inlet NO x prediction model and three ammonia injection amount prediction models for predicting the ammonia injection amount; Step 2: Based on the current time ( t Real-time running data, entry point NO x Concentration prediction model, generate ( t +1) Entry point NO at time x Concentration prediction; based on ( t Real-time running data, t +1) Exit NO at time x The concentration setpoint is generated using ammonia injection rate prediction model 1. t +1) Exit NO at time NO x Predicted concentration and predicted ammonia injection rate; Step 3: According to ( t Real-time running data, t +1) Entry point NO at time x Concentration prediction value, ( t +2) Time Exit NO x Concentration setting correction value and ( t The predicted ammonia injection rate at time +1 is used to generate (using ammonia injection rate prediction model 2). t +2) Predicted ammonia injection rate and outlet NO at time [time missing] x Concentration prediction; Step 4: According to ( t Real-time running data, t +1) Entry point NO at time x Concentration prediction value, ( t +3) Time Exit NO x Concentration setting correction value and ( t The predicted ammonia injection rate at time +2 is used to generate (using ammonia injection rate prediction model 3). t +3) Predicted ammonia injection rate and outlet NO at time [time missing] x Concentration prediction; Step 5: The SCR denitration system controls the ammonia injection amount of the SCR denitration reactor according to the ammonia injection amount prediction value provided by the prediction model.

7. The ammonia injection amount optimization control method according to claim 6, characterized by, The outlet NOx concentration at the time of step 2 (t+1) is calculated by the following equation. t +1) time x Concentration set value: Where, date Std ( t +1) is ( t +1) Exit NO at time x Concentration setpoint, date Std For export NO x Concentration reaches the set value. date Nox for( t (Exit NO) x The actual concentration value, where α is a set parameter.

8. The ammonia injection amount optimization control method according to claim 6, characterized by, The exhaust NOx concentration at the time of step 3 (t+2) is calculated by the following equation t +2) when the engine is operated at the time of step 3 x Concentration setting correction value: wherein date Std t +2) is the NOx concentration at the time point (t t +2), and x the set correction value of the NOx concentration, date Std t +1) is the NOx concentration at the time point (t t +1), and x the set value of the NOx concentration, date Sim t +1) is the NOx concentration at the time point (t t +1), and x the predicted value of the NOx concentration, β is a first preset parameter.​​​ 9. The ammonia injection amount optimization control method according to claim 6, characterized by, The exhaust NOx concentration at the time of step 4 when the engine is operating at the engine speed of 3,000 rpm and the engine load of 0.3 is calculated by the following equation. t +3) in terms of the engine speed and the engine load. x Concentration setting correction value: wherein date Std t +3) is the outlet NO t +3) concentration setpoint correction value, x date Std t +2) is the outlet NO t +2) concentration setpoint correction value, x date Sim t +2) is the outlet NO t +2) concentration prediction value, x ε is a second predetermined parameter.​​​​​​

Images