CN113488111B - Ammonia injection amount optimization modeling method for SCR denitration system - Google Patents
Ammonia injection amount optimization modeling method for SCR denitration system Download PDFInfo
- Publication number
- CN113488111B CN113488111B CN202110760346.1A CN202110760346A CN113488111B CN 113488111 B CN113488111 B CN 113488111B CN 202110760346 A CN202110760346 A CN 202110760346A CN 113488111 B CN113488111 B CN 113488111B
- Authority
- CN
- China
- Prior art keywords
- scr
- injection amount
- outlet
- ammonia injection
- ammonia
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/10—Analysis or design of chemical reactions, syntheses or processes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/346—Controlling the process
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/46—Removing components of defined structure
- B01D53/54—Nitrogen compounds
- B01D53/56—Nitrogen oxides
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
- B01D53/34—Chemical or biological purification of waste gases
- B01D53/74—General processes for purification of waste gases; Apparatus or devices specially adapted therefor
- B01D53/86—Catalytic processes
- B01D53/8621—Removing nitrogen compounds
- B01D53/8625—Nitrogen oxides
- B01D53/8628—Processes characterised by a specific catalyst
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/30—Prediction of properties of chemical compounds, compositions or mixtures
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D2251/00—Reactants
- B01D2251/20—Reductants
- B01D2251/206—Ammonium compounds
- B01D2251/2062—Ammonia
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D2258/00—Sources of waste gases
- B01D2258/02—Other waste gases
- B01D2258/0283—Flue gases
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to an optimization modeling method for ammonia injection amount of an SCR (selective catalytic reduction) denitration system, which is a dynamic modeling method for outlet parameters of an SCR reactor based on deep learning. The method is characterized in that a gated cyclic neural network is adopted to establish an accurate SCR reactor characteristic identification model, on the basis, an ammonia injection amount target value optimization model based on a deep neural network is established according to main interference parameters of outlet NOx concentration, so that accurate regulation of SCR outlet NOx concentration is realized, the control quality of a control system is improved from the aspects of suppressing external interference and balancing ammonia injection economy, the regulation precision and robustness of an SCR denitration process are finally improved, and the technical problems of hysteresis and poor adaptability of the traditional PID control are solved.
Description
Technical Field
The invention relates to the technical field of control modeling based on deep learning, in particular to an ammonia injection amount optimization modeling method for an SCR denitration system.
Background
The nitrogen oxide is one of main pollutants in the atmosphere and has great harm to the ecological environment and human health, and the emission of the nitrogen oxide caused by the production of the power industry and the heat industry accounts for about 45 percent of the total emission. Therefore, china requires that the nitrogen oxide emission of all thermal power generating units meet 50mg/Nm 3 Ultra-low emission standards.
The content of NOx in smoke is reduced by a Selective Catalytic Reduction (SCR) technology widely adopted by a coal-fired power plant unit, the mechanism process of chemical reaction in an SCR denitration system is complex, the concentration of NOx is influenced by interference parameters such as smoke flow, smoke oxygen content and the like, and the SCR denitration system has the characteristics of nonlinearity, large delay, multivariable coupling and the like. The traditional denitration control system adopts cascade control logic and controls the concentration of NOx at an SCR outlet by adjusting the ammonia injection amount, wherein a NOx concentration signal is sampled and measured by CEMS, and the measurement process has large lag, so that the measurement signal of the NOx concentration cannot accurately reflect the operation state of the SCR system, and the precision and the accuracy of the denitration control system are difficult to guarantee.
In order to weaken the adverse effect of the interference parameters on the SCR denitration system, the adjustment precision and robustness of the denitration control system are improved. The methods mainly adopted at present comprise optimization of ammonia injection amount, controller parameters and the like. The traditional ammonia injection amount optimization calculation method does not consider the influence of input parameter disturbance, and the calculated ammonia injection amount cannot respond to the NOx concentration change at the SCR inlet in time, namely the traditional PID control has the technical problems of lag and poor adaptability, and the economical efficiency and the environmental protection performance of a unit cannot be ensured.
How to establish an accurate dynamic ammonia injection amount calculation model on the basis of the characteristics of a denitration system is a problem which needs to be solved urgently.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an ammonia injection amount optimization modeling method for an SCR denitration system, and aims to solve the technical problems of lag and poor adaptability of the traditional PID control.
The technical scheme adopted by the invention is as follows:
an ammonia injection amount optimization modeling method for an SCR denitration system comprises the following steps:
s1: determining the characteristics of NOx and ammonia escape at the outlet of the SCR and the interference parameters and the influence parameters of a control system of the SCR denitration system according to the operating characteristics of the SCR denitration system; the characteristics of the SCR outlet NOx and ammonia slip refer to changes in SCR outlet NOx and ammonia slip caused by changes in the influencing parameters;
s2: according to the characteristics of NOx at the outlet of the SCR, the influence parameters and the physical structure of the SCR, training an SCR outlet parameter prediction model based on a GRU gating cyclic neural network; the input of the GRU gated cyclic neural network is the influence parameter, and the output is the NOx concentration at the SCR outlet and the ammonia escape amount;
s3: according to the characteristics of SCR outlet NOx, on the basis of the SCR outlet parameter prediction model, establishing an ammonia injection amount target value optimization model based on a DNN deep neural network;
the input of the input layer of the DNN deep neural network is the first-order difference value of the interference parameter and the feedback deviation Es of the control system, and the output of the output layer of the DNN deep neural network is the ammonia injection amount offset value D out ；
The feedback deviation Es is output by a controller of a control system to correspond to the ammonia injection amount, and the ammonia injection amount U and the ammonia injection amount offset value D are compared out Adding to obtain the optimized ammonia injection quantity Q in ＝U+D out If Q is in The variation limit Q of the ammonia injection amount is not exceeded max Then F will be in ＝[Q in ，F other ]Inputting into the SCR outlet parameter prediction model, otherwise inputting F in ＝[Q max ，F other ]Calculating the NOx concentration and ammonia escape at the outlet of the SCR, and utilizing a Loss function Loss DNN Training the DNN deep neural network until the accuracy of the model meets the requirement or the training reaches the maximum iteration number to obtain the ammonia injection amount target value optimization model; f other Inputting other SCR outlet parameter prediction models except for ammonia injection amount; limit value Q for variation of ammonia injection amount max The limit value is determined according to the variation rate and the upper and lower limits of the ammonia injection amount;
s4: and dynamically optimizing the NOx at the outlet of the SCR according to the ammonia injection amount target value optimization model.
The feedback deviation Es is the difference value between the NOx concentration calculated by the SCR outlet parameter prediction model and the NOx set value at the SCR outlet after the NOx concentration is calculated by a delay link of a control system; the delay element of the control system is that CEMS measures the pure delay time of SCR outlet NOx.
Loss function Loss DNN The expression of (a) is as follows:
wherein, y p The output value is the output value after the SCR outlet NOx concentration is predicted and is subjected to a delay link; y is r Is a set value of NOx concentration at an SCR outlet; w is the weight coefficient of the average ammonia consumption; alpha is an ammonia escape penalty coefficient; epsilon is ammonia escape at the outlet of the SCR reactor; epsilon max Standard limits for ammonia slip; beta is an NOx concentration small-mean value overrun penalty coefficient; y is e Is a rated SCR outlet NOx concentration set value; reLU represents a linear rectification function, n represents the number of samples, and i is the sample number.
The influencing parameters comprise the SCR inlet flue gas flow, the SCR inlet NOx concentration, the SCR inlet oxygen quantity, the ammonia spraying quantity and the temperature of a catalyst layer.
The interference parameters comprise load, NOx concentration at an SCR inlet, flue gas flow at the SCR inlet, oxygen content of a boiler, coal quantity, primary air quantity and secondary air quantity.
In the application, "SCR outlet" and "SCR inlet" refer to outlet and inlet of SCR denitration reactor respectively. In this application "SCR reactor" refers to SCR denitration reactor.
The invention has the following beneficial effects:
according to the method, a GRU (generalized regression Unit) cyclic neural network is adopted to establish an accurate SCR outlet parameter prediction model, and on the basis, a DNN (dinitrogen-based) ammonia injection amount target value optimization model is established according to the interference parameters of the NOx at the SCR outlet, so that the control and regulation of the NOx at the SCR outlet are realized, the influence of the interference parameters on an SCR denitration system is inhibited, the control quality of the control system is improved, and the regulation precision and robustness of the NOx at the SCR outlet are finally improved. The ammonia injection amount target value optimization model established by the invention has high precision and strong adaptability, and improves the robustness, accuracy and economy of the ammonia injection process.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a schematic flow chart of an SCR outlet parameter prediction model obtained by GRU gated cyclic neural network training of the present invention.
FIG. 3 is a flowchart of an optimization model for obtaining the ammonia injection amount target value by DNN deep neural network training according to the present invention.
FIG. 4 is a comparison of SCR outlet NOx output results and SCR outlet NOx measurement point data after ammonia injection amount optimization by the control system according to the embodiment of the present invention.
Detailed Description
The following describes embodiments of the present invention with reference to the drawings.
According to the method for optimizing and modeling ammonia injection amount of SCR denitration system, a thermal power generating unit comprising the SCR denitration system and a control system is used as a physical system, SCR outlet NOx is outlet NOx concentration of SCR of the thermal power generating unit in a variable load running state, and the design value of the SCR outlet NOx concentration of the thermal power generating unit is 50mg/Nm 3 。
As shown in fig. 1, the ammonia injection amount optimization modeling method of the application comprises the following steps:
step S1: determining the characteristic of NOx at an outlet of an SCR according to the operating characteristic of a denitration system, and determining an interference parameter and an influence parameter of a control system; wherein the characteristic of SCR outlet NOx refers to the change of SCR outlet NOx caused by the change of the influencing parameter;
the influence parameters comprise the SCR inlet flue gas flow, the SCR inlet NOx concentration, the SCR inlet oxygen quantity, the ammonia spraying quantity and the catalyst layer temperature; the interference parameters comprise load, boiler oxygen amount, SCR inlet NOx concentration, SCR inlet flue gas flow, coal amount, primary air volume and secondary air volume.
Step S2: according to the characteristics of the SCR outlet NOx, the control system influence parameters and the physical structure of the SCR, training an SCR outlet parameter prediction model on the basis of a GRU gated cyclic neural network, wherein the SCR outlet parameter prediction model is used for obtaining the dynamic characteristics of an SCR denitration system;
referring to fig. 2, the specific process is as follows:
step S21: and inputting the influence parameters into a GRU network layer to obtain a calculated value of the concentration of NOx at the outlet of the SCR.
Wherein GRU input is F in ，F in Namely the SCR inlet flue gas flow, the SCR inlet NOx concentration, the SCR inlet oxygen quantity, the ammonia spraying quantity and the catalyst layer temperature, and outputs F out To SCR outlet NO x Concentration and ammonia escape amount, and the input and output expressions are as follows: f out ＝f GRU (F in )；
Step S22: training according to the step S21 until an SCR outlet parameter prediction model F determined by network parameters is obtained out ＝f model (F in )；
Wherein, GRU network layerThe calculation function is out GRU ＝f GRU (in GRU )，in GRU For the input of the GRU network layer, out GRU Is the output of the GRU network layer.
Specifically, the GRU network layer includes: if the number of the hidden units is h, inputting x in small batch with given time step length t t ∈R n ×d And last time step unit outputs system state h t-1 ∈R n×h R is an input parameter matrix, n represents the number of samples, and d represents the number of inputs;
the formulas of the reset gate, the update gate, the current memory state, the unit output system state and the calculation rule of the predicted value are as follows:
resetting a gate:
r t ＝σ(W r ·[h t-1 ，x t ])；
and (3) updating a door:
z t ＝σ(W z ·[h t-1 ，x t ])；
the current memory state:
the unit outputs the system state:
calculation rule of the predicted value:
y＝σ(W O ·h t )；
output system state h of each cyclic neural unit at current time step t ∈R n×h The current time step input and the last time step system state jointly determine.
In the above formulas, σ and tanh are Sigmoid activation function and hyperbolic tangent function, W r 、W z 、
And W o Is a corresponding weight matrix, wherein W r ＝W rx +W rh ，W rx And W rh Respectively represent input x t And unit output system state h t-1 Corresponding reset gate weight matrix, W z ＝W zx +W zh ，W zx And W zh Respectively represent input x t Sum unit output system state h t-1 The corresponding updated gate weight matrix is then updated,
and
respectively represent input x t And unit output system state h t-1 The corresponding currently memorized weight matrix.
And step S3: according to the characteristics of SCR outlet NOx, on the basis of an SCR outlet parameter prediction model, training an ammonia injection amount target value optimization model based on a DNN deep neural network;
referring to fig. 3, the specific process is as follows:
s31: inputting D to input layer of DNN deep neural network in Training to obtain the output ammonia spraying amount offset value D of the output layer of the DNN deep neural network out (ii) a Wherein Di n The method comprises the following steps of (1) carrying out interference on the first-order difference values of the interference parameters, namely load, boiler oxygen quantity, SCR inlet NOx concentration, SCR inlet flue gas flow, coal quantity, primary air quantity and secondary air quantity, and feedback deviation Es of a control system;
the feedback deviation Es of the control system is the difference value between the NOx concentration calculated by the SCR outlet parameter prediction model and the NOx set value at the SCR outlet after the NOx concentration is calculated by a delay link of the control system; the delay element of the control system is that the CEMS measures the pure delay time of the SCR outlet NOx.
In the control system, the feedback deviation Es is output through a controller to form an ammonia injection amount U; the ammonia injection amount U output by the controller and the output D of the optimization model of the ammonia injection amount target value out Adding to obtain the optimized ammonia injection quantity Q in ，Q in ＝U+D out ；
S32: if Q in The variation limit Q of the ammonia injection amount is not exceeded max Then [ Q ] will in ，F other ]As F in I.e. F in ＝[Q in ，F other ]Inputting into the SCR outlet parameter prediction model, otherwise inputting F in ＝[Q max ，F other ]Calculating the NOx concentration and ammonia escape at the outlet of the SCR, and utilizing a Loss function Loss DNN Training the DNN deep neural network until the precision of the model meets the requirement or the training reaches the maximum iteration number to obtain an ammonia injection amount target value optimization model f DNN 。
Wherein the limit value Q of variation of the ammonia injection amount max Determining according to the variation rate and the upper and lower limits of the ammonia injection amount; f other And inputting other SCR outlet parameter prediction models except for the ammonia injection amount.
Loss function Loss of DNN deep neural network DNN The expression is as follows:
wherein, y p The output value is the output value after the SCR outlet NOx concentration is predicted and is subjected to a delay link; y is r Is a set value of NOx concentration at an SCR outlet; a is the weight coefficient of the average ammonia consumption; alpha is an ammonia escape penalty coefficient; epsilon is ammonia escape at the outlet of the SCR reactor; epsilon max Is an ammonia slip standard limit; beta is an NOx concentration small-mean value overrun penalty coefficient; y is e Is a rated SCR outlet NOx concentration set value; reLU represents a linear rectification function, n represents the number of samples, and i is the sample number.
S4: optimizing model f according to ammonia injection amount target value DNN And dynamically optimizing the SCR outlet NOx concentration.
The denitration control system of a certain 660MW coal-fired unit is taken as an example, and the design value of the rated SCR outlet concentration is 50mg/Nm 3 . Standard limit of ammonia slip epsilon max Is 3ppm; the weight coefficient W of the average ammonia consumption is 0.1; the ammonia escape penalty coefficient alpha is 0.001; small concentration of NOxThe mean overrun penalty coefficient beta is 999; rated SCR outlet NOx concentration set value y e Is 50mg/Nm 3 (ii) a The controller is a PID controller; the measurement delay of the control system is 60 seconds.
FIG. 4 shows a comparison of SCR outlet NOx output results and SCR outlet NOx setpoint data over a period of time (00 on 10 days 08/10/2020. After further statistics, the Root Mean Square Error (RMSE) was 3.242mg/Nm for the entire test set before optimization 3 After optimization, the concentration is 0.903mg/Nm 3 . For the whole denitration process, the average ammonia injection amount is reduced by 1.252kg/Nm & lt 3 & gt, and the control precision and the economical efficiency of a control system are improved.
Claims (5)
1. An ammonia injection amount optimization modeling method for an SCR denitration system is characterized by comprising the following steps:
s1: determining the characteristics of NOx and ammonia escape at an SCR outlet and the interference parameters and the influence parameters of a control system of the SCR denitration system according to the operating characteristics of the SCR denitration system; the characteristics of the SCR outlet NOx and ammonia slip refer to changes in SCR outlet NOx and ammonia slip caused by changes in the influencing parameters;
s2: according to the characteristics of NOx at the outlet of the SCR, the influence parameters and the physical structure of the SCR, training an SCR outlet parameter prediction model based on a GRU gated cyclic neural network; the input of the GRU gated cyclic neural network is the influence parameter, and the output is the NOx concentration at the SCR outlet and the ammonia escape amount;
s3: according to the characteristics of SCR outlet NOx, on the basis of the SCR outlet parameter prediction model, establishing an ammonia injection amount target value optimization model based on a DNN deep neural network;
the input of the input layer of the DNN deep neural network is the first-order difference value of the interference parameter and the feedback deviation Es of the control system, and the output of the output layer of the DNN deep neural network is the ammonia injection amount offset value D out ；
The feedback deviation Es is output by a controller of a control system to correspond to the ammonia injection amount U, and the ammonia injection amount U and the ammonia injection amount offset value D are compared out Adding to obtain the optimized ammonia injection quantity Q in ＝U+D out If Q is in The variation limit Q of the ammonia injection amount is not exceeded max Then F will be in ＝[Q in ，F other ]Inputting into the SCR outlet parameter prediction model, otherwise inputting F in ＝[Q max ，F other ]Calculating the NOx concentration and ammonia escape at the outlet of the SCR, and utilizing a Loss function Loss DNN Training the DNN deep neural network until the precision of the model meets the requirement or the training reaches the maximum iteration number to obtain the ammonia injection amount target value optimization model; wherein, F other Inputting other SCR outlet parameter prediction models except for ammonia injection amount;
s4: and dynamically optimizing the NOx at the outlet of the SCR according to the ammonia injection amount target value optimization model.
2. The method for optimizing and modeling the ammonia injection amount of the SCR denitration system according to claim 1, wherein the feedback deviation Es is a difference value between a NOx concentration calculated by the SCR outlet parameter prediction model and a NOx set value at an SCR outlet after the NOx concentration is calculated by a delay link of a control system; the delay element of the control system is that CEMS measures the pure delay time of the SCR outlet NOx.
3. The method for optimizing and modeling ammonia injection amount of SCR denitration system according to claim 2, wherein the Loss function Loss is less DNN The expression of (a) is as follows:
wherein, y p After the NOx concentration at the SCR outlet is predicted, the output value is subjected to a delay link; y is r Is a set value of NOx concentration at an SCR outlet; w is the weight coefficient of the average ammonia consumption; alpha is an ammonia escape penalty coefficient; epsilon is ammonia escape at the outlet of the SCR reactor; epsilon max Is an ammonia slip standard limit; beta is an NOx concentration small-mean value overrun penalty coefficient; y is e Is a rated SCR outlet NOx concentration set value; reLU denotes a linear rectification function, n tableThe number of samples is shown, and i is the sample number.
4. The SCR denitration system ammonia injection amount optimization modeling method of claim 1, wherein the influencing parameters comprise SCR inlet flue gas flow, SCR inlet NOx concentration, SCR inlet oxygen amount, ammonia injection amount and catalyst layer temperature.
5. The method for optimizing and modeling ammonia injection amount of the SCR denitration system according to claim 1, wherein the disturbance parameters comprise load, SCR inlet flue gas flow rate, SCR inlet NOx concentration, boiler oxygen amount, coal amount, primary air amount and secondary air amount.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110760346.1A CN113488111B (en) | 2021-07-05 | 2021-07-05 | Ammonia injection amount optimization modeling method for SCR denitration system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110760346.1A CN113488111B (en) | 2021-07-05 | 2021-07-05 | Ammonia injection amount optimization modeling method for SCR denitration system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113488111A CN113488111A (en) | 2021-10-08 |
| CN113488111B true CN113488111B (en) | 2022-11-22 |
Family
ID=77941091
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110760346.1A Active CN113488111B (en) | 2021-07-05 | 2021-07-05 | Ammonia injection amount optimization modeling method for SCR denitration system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113488111B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114664389B (en) * | 2022-03-24 | 2023-09-22 | 南方电网电力科技股份有限公司 | Prediction method and device for reaction conditions of urea hydrolysis ammonia production |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104826492A (en) * | 2015-04-23 | 2015-08-12 | 华北电力大学（保定） | Improvement method for selective catalytic reduction flue gas denitrification and ammonia injection control system |
| CN107694337A (en) * | 2017-11-03 | 2018-02-16 | 吉林省电力科学研究院有限公司 | Coal unit SCR denitrating flue gas control methods based on network response surface |
| CN108837698A (en) * | 2018-07-02 | 2018-11-20 | 大唐环境产业集团股份有限公司 | Based on advanced measuring instrumentss and the SCR denitration of advanced control algorithm spray ammonia optimization method and system |
-
2021
- 2021-07-05 CN CN202110760346.1A patent/CN113488111B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104826492A (en) * | 2015-04-23 | 2015-08-12 | 华北电力大学（保定） | Improvement method for selective catalytic reduction flue gas denitrification and ammonia injection control system |
| CN107694337A (en) * | 2017-11-03 | 2018-02-16 | 吉林省电力科学研究院有限公司 | Coal unit SCR denitrating flue gas control methods based on network response surface |
| CN108837698A (en) * | 2018-07-02 | 2018-11-20 | 大唐环境产业集团股份有限公司 | Based on advanced measuring instrumentss and the SCR denitration of advanced control algorithm spray ammonia optimization method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113488111A (en) | 2021-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109062053B (en) | Denitration ammonia injection control method based on multivariate correction | |
| CN109343349B (en) | SCR flue gas denitration optimal control system and method based on ammonia injection amount compensator | |
| CN109583585B (en) | Construction method of power station boiler wall temperature prediction neural network model | |
| CN110368808B (en) | Ammonia spraying amount control method and system for SCR flue gas denitration system | |
| CN112580250A (en) | Thermal power generating unit denitration system based on deep learning and optimization control method | |
| CN113433911B (en) | Accurate control system and method for ammonia spraying of denitration device based on accurate concentration prediction | |
| CN111897373A (en) | Model prediction-based ammonia injection flow adjusting method for SCR denitration device | |
| CN110597070B (en) | Method for identifying model parameters of thermal power generating unit system | |
| CN113488111B (en) | Ammonia injection amount optimization modeling method for SCR denitration system | |
| Xu et al. | Control of denitration system in cement calcination process: A Novel method of Deep Neural Network Model Predictive Control | |
| CN114721263B (en) | Intelligent regulation and control method for cement decomposing furnace based on machine learning and intelligent optimization algorithm | |
| CN112613237B (en) | CFB unit NOx emission concentration prediction method based on LSTM | |
| CN112016754A (en) | Power station boiler exhaust gas temperature advanced prediction system and method based on neural network | |
| CN115018158A (en) | SCR (Selective catalytic reduction) outlet NOx emission prediction method based on BWOA-BiGRU-LAM (lean-reactive inert gas) | |
| CN115049139A (en) | Multi-index model prediction control method for cement sintering denitration system | |
| CN109046021B (en) | SCR system accurate ammonia injection control method with strong self-adaptive capacity | |
| CN114186708A (en) | Circulating fluidized bed unit SO based on PSO-ELM2Concentration prediction method | |
| Tang et al. | Predictive control of SCR denitrification system in thermal power plants based on GA‐BP and PSO | |
| CN117270387A (en) | SCR denitration system low ammonia escape control method and system based on deep learning | |
| WO2022095534A1 (en) | Method for predicting ammonia escaping from thermal power plant | |
| CN112231978B (en) | Boiler flue gas acid dew point testing method based on artificial neural network | |
| CN115034486A (en) | Prediction control method for iSNCR denitration of cement kiln flue gas | |
| CN115113519A (en) | Coal-gas co-combustion boiler denitration system outlet NO x Concentration early warning method | |
| CN113485106B (en) | Method for controlling concentration of nitrogen oxide in thermal power plant | |
| CN117190173B (en) | Optimal control method and control system for flue gas recirculation and boiler coupling system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |