CN114708924A - Model construction method and device for predicting soot blowing interval time of soot blower in SCR system - Google Patents
Model construction method and device for predicting soot blowing interval time of soot blower in SCR system Download PDFInfo
- Publication number
- CN114708924A CN114708924A CN202210314076.6A CN202210314076A CN114708924A CN 114708924 A CN114708924 A CN 114708924A CN 202210314076 A CN202210314076 A CN 202210314076A CN 114708924 A CN114708924 A CN 114708924A
- Authority
- CN
- China
- Prior art keywords
- soot blowing
- model
- soot
- blowing time
- neural network
- 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.)
- Granted
Links
- 239000004071 soot Substances 0.000 title claims abstract description 133
- 238000007664 blowing Methods 0.000 title claims abstract description 87
- 238000010276 construction Methods 0.000 title claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 44
- 238000012549 training Methods 0.000 claims abstract description 32
- 238000012360 testing method Methods 0.000 claims abstract description 25
- 238000005516 engineering process Methods 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 20
- 230000005284 excitation Effects 0.000 claims abstract description 12
- 238000010606 normalization Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 6
- 230000006870 function Effects 0.000 claims description 23
- 239000002245 particle Substances 0.000 claims description 20
- 238000009434 installation Methods 0.000 claims description 16
- 238000004140 cleaning Methods 0.000 claims description 15
- 230000001902 propagating effect Effects 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 5
- 238000003860 storage Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 5
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 239000002956 ash Substances 0.000 description 25
- 238000010586 diagram Methods 0.000 description 5
- 239000003054 catalyst Substances 0.000 description 3
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 238000010531 catalytic reduction reaction Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 239000003546 flue gas Substances 0.000 description 1
- 239000010881 fly ash Substances 0.000 description 1
- 239000008187 granular material Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
-
- 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
- G06N3/048—Activation functions
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Crystallography & Structural Chemistry (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Analytical Chemistry (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides a model construction method and a device for predicting soot blowing interval time of a soot blower in an SCR system based on BP neural network technology, wherein the method comprises the following steps: collecting structural parameters, operation data and historical soot blowing time intervals of an SCR system to construct an initial database; carrying out normalization processing on an initial database, and dividing the initial database into a training set and a test set; selecting an excitation function of a BP neural network, determining an input layer node, an output layer node and a hidden layer node of the BP neural network, and constructing a soot blowing time interval model; training a soot blowing time interval model by adopting a training set; and optimizing the weight and the threshold of the soot blowing time interval model through the test set until the weight and the threshold are converged to obtain the trained soot blowing time interval model. The invention determines the optimal soot blowing time quickly, conveniently and accurately according to the existing structural parameters and the operation data of the SCR system, avoids the debugging process in production and greatly saves time and economic cost.
Description
Technical Field
The document relates to the technical field of machine learning, in particular to a model construction method and a model construction device for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology.
Background
Selective Catalytic Reduction (SCR) denitration technology is the most important coal-fired boiler denitration method at home and abroad due to its mature process and high denitration efficiency. The denitration catalyst is the most important functional material in the SCR system, and the service life of the denitration catalyst has great influence on the denitration efficiency and the economic cost of the SCR technology. However, high concentrations of fly ash particles in flue gas flowing through SCR systems can cause catalyst plugging and attrition, which can significantly reduce denitration device performance, lead to increased ammonia slip and plugging of air preheaters, and in severe cases can even lead to unplanned shutdowns. Therefore, effectively separate the large granule ash, guarantee that SCR deNOx systems safe high-efficient operation is the urgent problem that needs to solve of each power plant.
The large-particle ash blocking device arranged at the proper position of the flue is the main technology for separating and trapping large-particle ash at present, and the technology has the advantages of low design requirement on a flow field, high trapping efficiency, convenience in modification and installation and the like. At present, the large-particle ash intercepting device is successfully applied to a million domestic unit selective catalytic reduction denitration systems. In order to prevent the blocking of the interception net, a self-ash-cleaning structure is required to be arranged at the interception device, for example, an air-gun ash-blowing device is arranged on the lee side of the interception net. On the premise of determining the type and the installation position of the soot blower, the soot blowing interval is the most important factor influencing the interception and removal efficiency of large particle soot. Although the SCR system of the coal-fired power plant is provided with the soot blower for soot blowing, most power plants have insufficient on-time quantitative soot blowing modes established according to experience due to lack of real-time visual data of soot blower blockage, or the pressure drop loss and the efficiency are not effectively reduced due to insufficient soot blowing, or the energy consumption and the cost are increased due to excessive soot blowing times. Therefore, how to predict soot blower blockage in real time and formulate a reasonable soot blowing scheme has important significance for the stable operation of the SCR system. The existing large particle ash interception technology of the SCR denitration system is still in a starting stage, the ash blowing interval time setting in the ash blowing control scheme is still obtained by experience, and the applicability is limited under the complex operation working condition of a power plant.
Disclosure of Invention
The invention aims to provide a model construction method and a model construction device for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology, and aims to solve the problems in the prior art.
The invention provides a model construction method for predicting soot blowing interval time of a soot blower in an SCR system based on BP neural network technology, which comprises the following steps:
collecting the structural parameters, the operation data and the historical soot blowing time intervals of the SCR system, and constructing an initial database based on the structural parameters, the operation data and the historical soot blowing time intervals of the SCR system;
carrying out normalization processing on the initial database, and dividing the normalized initial database into a training set and a test set according to a specific proportion;
selecting an excitation function of the BP neural network, determining an input layer node, an output layer node and a hidden layer node of the BP neural network, and constructing a soot blowing time interval model;
training a soot blowing time interval model by adopting a training set;
and optimizing the weight and the threshold of the soot blowing time interval model through the test set until the weight and the threshold are converged to obtain the trained soot blowing time interval model.
The invention provides a model construction device for predicting soot blowing interval time of a soot blower in an SCR system based on BP neural network technology, which comprises the following steps:
the data acquisition module is used for acquiring the structural parameters, the operating data and the historical soot blowing time intervals of the SCR system and constructing an initial database based on the structural parameters, the operating data and the historical soot blowing time intervals of the SCR system;
the normalization module is used for performing normalization processing on the initial database and dividing the normalized initial database into a training set and a test set according to a specific proportion;
the model building module is used for building a soot blowing time interval model by selecting an excitation function of the BP neural network and determining an input layer node, an output layer node and a hidden layer node of the BP neural network;
the model training module is used for training the soot blowing time interval model through a training set;
and the model optimization module is used for optimizing the weight and the threshold of the soot blowing time interval model through the test set until the weight and the threshold are converged to obtain the trained soot blowing time interval model.
The present invention provides an electronic device, including: a processor;
and a memory arranged to store computer executable instructions that when executed cause the processor to perform the steps of a model building method for predicting soot blowing interval times of soot blowers in an SCR system based on a BP neural network technique as described above
The present invention provides a storage medium for storing computer-executable instructions which, when executed, implement the steps of the model building method for predicting soot blowing interval time of soot blowers in an SCR system based on BP neural network technology as described above
By adopting the embodiment of the invention, the soot blowing time interval model is constructed by collecting the structure parameters, the operation data and the historical soot blowing time interval of the SCR system, and the weight and the threshold of the soot blowing time interval model are optimized to obtain the optimal soot blowing time interval model, so that the optimal soot blowing time is quickly, conveniently and accurately determined, the debugging process in production is avoided, and the time and the economic cost are greatly saved.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and that other drawings can be obtained by those skilled in the art without inventive exercise.
FIG. 1 is a flowchart of a model construction method for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a BP neural network structure according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a model building device for predicting the soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in one or more embodiments of the present disclosure, the technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in one or more embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from one or more of the embodiments described herein without making any inventive step shall fall within the scope of protection of this document.
Method embodiment
According to an embodiment of the present invention, a model construction method for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology is provided, fig. 1 is a flowchart of the model construction method for predicting soot blowing interval time of a soot blower in an SCR system based on the BP neural network technology according to an embodiment of the present invention, as shown in fig. 1, the model construction method for predicting soot blowing interval time of a soot blower in an SCR system based on the BP neural network technology according to an embodiment of the present invention specifically includes:
step S101, collecting SCR system structure parameters, operation data and historical soot blowing time intervals, and constructing an initial database based on the SCR system structure parameters, the operation data and the historical soot blowing time intervals; step S101 specifically includes: and (3) finishing a large number of simulations and tests which are completed before, collecting structural parameters and operation data of the SCR system, recording historical soot blowing time intervals, and constructing a corresponding database of the parameters and soot blowing time of a soot blower.
Step S102, carrying out normalization processing on an initial database, and dividing the normalized initial database into a training set and a test set according to a specific proportion; step S102 specifically includes: the normalization formula is adopted as follows:
wherein, XiFor the ith normalized input parameter, xiIs the ith input parameter, xmin，iIs the minimum value, x, of the ith input parametermax,iMaximum value in the ith input parameter. A 7:3 ratio is typically used to divide the training set and the test set.
Step S103, selecting an excitation function of the BP neural network, determining an input layer node, an output layer node and a hidden layer node of the BP neural network, and constructing a soot blowing time interval model;
adopting a sigmoid function as a stimulus function, wherein the sigmoid function is as follows:
where s is the argument of the excitation function and is the input variable transmitted between layers.
FIG. 2 is a schematic structural diagram of a BP neural network according to an embodiment of the present invention, and it can be known from FIG. 2 that an input parameter x1～x6Respectively comprises the installation height of the interception net, the air flow, the diameter of the large particle ash, the flow of the large particle ash, the number of soot blowers, the installation angle of the soot blowers and an output parameter y1The interval time for ash removal.
Step S104, training a soot blowing time interval model by adopting a training set; step S104 specifically includes:
and solving the output of each node of the hidden layer and the output layer, calculating the deviation between the target value and the actual output value, if the error is greater than the allowable error, reversely propagating the error to the input layer through the hidden layer, updating the weight value by the hidden layer according to the signal, and then, propagating the error forward again until the target value and the predicted value meet the allowable error, and finishing learning.
And S105, optimizing the weight and the threshold of the soot blowing time interval model through the test set until the weight and the threshold are converged to obtain the trained soot blowing time interval model. Step S105 specifically includes: and (3) verifying the precision of the prediction model by using the data of the test set, and successfully constructing the BP neural network prediction model when the precision reaches the standard. And when the precision does not reach the standard, repeating the training process until the precision reaches the standard. The evaluation indexes are a correlation coefficient and a variance, and the correlation coefficient is as follows:
the variance is:
wherein, yiIs the actual value of the jth ash cleaning interval time,
is the predicted value of the jth ash cleaning interval time,
is the average of the ash removal interval times, and N is the number of samples in the test set.
After the BP neural network prediction model is successfully constructed, in the actual operation of a power plant, firstly determining a large particle ash interception scheme (such as the position of an interception net, the opening shape of the interception net and the structural shape of the interception net) and a soot blowing scheme (such as the type of a soot blower, the installation position of the soot blower and the soot cleaning mode), and then inputting specific numerical values of six input parameters (the installation height of the interception net, the airflow, the diameter of the large particle ash, the flow of the large particle ash, the number of the soot blowers and the installation angle of the soot blower) into the constructed BP neural network to obtain the soot cleaning interval time meeting the optimal soot cleaning efficiency condition.
Apparatus embodiment one
According to an embodiment of the present invention, a model construction device for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology is provided, fig. 3 is a schematic diagram of a model construction device for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology according to an embodiment of the present invention, as shown in fig. 3, a model construction device for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology according to an embodiment of the present invention specifically includes:
the data acquisition module 30 is used for acquiring the structural parameters, the operation data and the historical soot blowing time intervals of the SCR system, and constructing an initial database based on the structural parameters, the operation data and the historical soot blowing time intervals of the SCR system; the data acquisition module is specifically configured to: and (3) finishing a large number of simulations and tests which are completed before, collecting structural parameters and operation data of the SCR system, recording historical soot blowing time intervals, and constructing a corresponding database of the parameters and soot blowing time of a soot blower.
A normalization module 32, configured to perform normalization processing on the initial database, and divide the normalized initial database into a training set and a test set according to a specific ratio; the normalization module is specifically configured to: the normalization formula is adopted as follows:
wherein, XiFor the ith normalized input parameter, xiIs the ith input parameter, xmin，iIs the minimum value, x, of the ith input parametermax,iMaximum value in the ith input parameter. A 7:3 ratio is typically used to divide the training set and the test set.
The model building module 34 is used for building a soot blowing time interval model by selecting an excitation function of the BP neural network and determining an input layer node, an output layer node and a hidden layer node of the BP neural network; the model building module 34 is specifically configured to: adopting a sigmoid function as a stimulus function, wherein the sigmoid function is as follows:
where s is the argument of the excitation function and is the input variable transmitted between layers.
FIG. 2 is a schematic diagram of a BP neural network according to an embodiment of the present invention, and it can be known from FIG. 2 that an input parameter x1～x6Respectively comprises the installation height of the interception net, the air flow, the diameter of the large particle ash, the flow of the large particle ash, the number of soot blowers, the installation angle of the soot blowers and an output parameter y1The interval time for ash removal.
A model training module 36 for training the soot blowing time interval model through a training set; the model training module is specifically configured to: and solving the output of each node of the hidden layer and the output layer, calculating the deviation between the target value and the actual output value, if the error is greater than the allowable error, reversely propagating the error to the input layer through the hidden layer, updating the weight value by the hidden layer according to the signal, and then, propagating the error forward again until the target value and the predicted value meet the allowable error, and finishing learning.
A model optimization module 38, configured to optimize the weight and the threshold of the soot blowing time interval model through the test set until the weight and the threshold converge, to obtain a trained soot blowing time interval model, where the model optimization module is specifically configured to: and (3) verifying the precision of the prediction model by using the data of the test set, and successfully constructing the BP neural network prediction model when the precision reaches the standard. And when the precision does not reach the standard, repeating the training process until the precision reaches the standard. The evaluation indexes are a correlation coefficient and a variance, and the correlation coefficient is as follows:
the variance is:
wherein, yiIs the actual value of the jth ash cleaning interval time,
is the predicted value of the jth ash cleaning interval time,
is the average of the ash removal interval times, and N is the number of samples in the test set.
After the BP neural network prediction model is successfully constructed, in the actual operation of a power plant, firstly determining a large particle ash interception scheme (such as the position of an interception net, the opening shape of the interception net and the structural shape of the interception net) and a soot blowing scheme (such as the type of a soot blower, the installation position of the soot blower and the soot cleaning mode), and then inputting specific numerical values of six input parameters (the installation height of the interception net, the airflow, the diameter of the large particle ash, the flow of the large particle ash, the number of the soot blowers and the installation angle of the soot blower) into the constructed BP neural network to obtain the soot cleaning interval time meeting the optimal soot cleaning efficiency condition.
Device embodiment II
An embodiment of the present invention provides an electronic device, including: a processor;
and a memory arranged to store computer executable instructions that when executed cause the processor to perform the steps as described in the method embodiments.
Device embodiment III
An embodiment of the present invention provides a computer-readable storage medium, on which an implementation program for information transmission is stored, where the program, when executed by a processor, implements the steps described in the method embodiment.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A model construction method for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology is characterized by comprising the following steps:
collecting SCR system structure parameters, operation data and historical soot blowing time intervals, and constructing an initial database based on the SCR system structure parameters, the operation data and the historical soot blowing time intervals;
normalizing the initial database, and dividing the normalized initial database into a training set and a test set according to a specific proportion;
selecting an excitation function of the BP neural network, determining an input layer node, an output layer node and a hidden layer node of the BP neural network, and constructing a soot blowing time interval model;
training the soot blowing time interval model by adopting a training set;
and optimizing the weight and the threshold of the soot blowing time interval model through a test set until the weight and the threshold are converged to obtain a trained soot blowing time interval model.
2. The method according to claim 1, wherein the selecting an excitation function of the BP neural network, and determining the input layer nodes, the output layer nodes, and the hidden layer nodes of the BP neural network to construct the soot blowing time interval model specifically comprises:
adopting a sigmoid function as a stimulus function, wherein the sigmoid function is obtained through a formula 1;
wherein s is an argument of the excitation function;
setting the input layer node parameter to x1～x6X is said1～x6The installation height of the interception net, the air flow, the diameter of large particle ash, the flow of the large particle ash, the number of soot blowers and the installation angle of the soot blowers are respectively set;
setting output layer node parameter y1The interval time of ash removal is;
solving the number l of hidden layer nodes according to a formula 2;
wherein m represents the number of nodes of the input layer, n represents the number of nodes of the output layer, and a is a constant and has a value range of [1,10 ].
3. The method of claim 1, wherein the training the soot blowing time interval model with a training set specifically comprises:
and calculating the deviation between the target value and the actual output value according to the output of the hidden layer node and the output layer node, if the deviation is greater than a preset error, reversely propagating the error to the input layer through the hidden layer, updating the weight value by the hidden layer according to the signal, and then, propagating the error in the forward direction again until the target value and the predicted value meet the preset error.
4. The method of claim 1, wherein the optimizing the weights and thresholds of the soot blowing time interval model through the test set specifically comprises:
optimizing the weight and the threshold of the soot blowing time interval model through a correlation coefficient and a mean square error, wherein the correlation coefficient is obtained through a formula 3, and the mean square error is obtained through a formula 4;
wherein, yiIs the actual value of the jth ash cleaning interval time,
is the predicted value of the jth ash cleaning interval time,
is the average of the ash removal interval times, and N is the number of samples in the test set.
5. A model building device for predicting soot blowing interval time of a soot blower in an SCR system based on a BP neural network technology is characterized by comprising the following components:
the data acquisition module is used for acquiring the structural parameters, the operating data and the historical soot blowing time intervals of the SCR system and constructing an initial database based on the structural parameters, the operating data and the historical soot blowing time intervals of the SCR system;
the normalization module is used for performing normalization processing on the initial database and dividing the normalized initial database into a training set and a test set according to a specific proportion;
the model building module is used for building a soot blowing time interval model by selecting an excitation function of the BP neural network and determining an input layer node, an output layer node and a hidden layer node of the BP neural network;
the model training module is used for training the soot blowing time interval model through a training set;
and the model optimization module is used for optimizing the weight and the threshold of the soot blowing time interval model through a test set until the weight and the threshold are converged to obtain a trained soot blowing time interval model.
6. The apparatus of claim 5, wherein the model building module is specifically configured to:
adopting a sigmoid function as a stimulus function, wherein the sigmoid function is obtained through a formula 1;
wherein s is an argument of the excitation function;
setting input layer node parameter as x1～x6X is said1～x6The installation height of the interception net, the air flow, the diameter of large particle ash, the flow of the large particle ash, the number of soot blowers and the installation angle of the soot blowers are respectively set;
setting an output layer node parameter y1The interval time of ash removal is;
solving the number l of hidden layer nodes according to a formula 2;
wherein m represents the number of nodes of the input layer, n represents the number of nodes of the output layer, a is a constant, and the value range is [1,10 ].
7. The apparatus of claim 5, wherein the model training module is specifically configured to:
and calculating the deviation between the target value and the actual output value according to the output of the hidden layer node and the output layer node, if the deviation is greater than a preset error, reversely propagating the error to the input layer through the hidden layer, updating the weight value by the hidden layer according to the signal, and then, propagating the error in the forward direction again until the target value and the predicted value meet the preset error.
8. The apparatus of claim 5, wherein the model optimization module is specifically configured to:
optimizing the weight and the threshold of the soot blowing time interval model through a correlation coefficient and a mean square error, wherein the correlation coefficient is obtained through a formula 3, and the mean square error is obtained through a formula 4;
wherein, yiIs the actual value of the jth ash cleaning interval time,
is the predicted value of the jth ash cleaning interval time,
is the average of the ash removal interval times, and N is the number of samples in the test set.
9. An electronic device, comprising: a processor;
and a memory arranged to store computer executable instructions that when executed cause the processor to implement the steps of a model building method of predicting sootblower interval times in an SCR system based on BP neural network techniques as claimed in any one of claims 1 to 4.
10. A storage medium storing computer executable instructions which, when executed, implement the steps of a model building method of predicting soot blowing interval times of soot blowers in an SCR system based on a BP neural network technique as claimed in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210314076.6A CN114708924B (en) | 2022-03-28 | 2022-03-28 | Model construction method and device for predicting soot blower soot blowing interval time in SCR system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210314076.6A CN114708924B (en) | 2022-03-28 | 2022-03-28 | Model construction method and device for predicting soot blower soot blowing interval time in SCR system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114708924A true CN114708924A (en) | 2022-07-05 |
| CN114708924B CN114708924B (en) | 2024-06-18 |
Family
ID=82171566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210314076.6A Active CN114708924B (en) | 2022-03-28 | 2022-03-28 | Model construction method and device for predicting soot blower soot blowing interval time in SCR system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114708924B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069185A (en) * | 2015-07-14 | 2015-11-18 | 东南大学 | Method for establishing air pre-heater clean factor calculation model by using smoke pressure difference, and application |
| CN106599586A (en) * | 2016-12-19 | 2017-04-26 | 北京国能中电节能环保技术股份有限公司 | Neural network-based SCR intelligent ammonia-spraying optimization method and apparatus |
| US20180024512A1 (en) * | 2016-07-25 | 2018-01-25 | General Electric Company | System modeling, control and optimization |
| KR20180076918A (en) * | 2016-12-28 | 2018-07-06 | 대우조선해양 주식회사 | Soot blowing apparatus and method of exhaust system for vessel |
| CN109603545A (en) * | 2018-12-28 | 2019-04-12 | 浙江大学 | A kind of SCR denitration method and device for combining soot blowing |
| CN109654517A (en) * | 2018-12-05 | 2019-04-19 | 中北大学 | A kind of boiler soot-blowing optimization method based on the prediction of heating surface health status |
| CN111068516A (en) * | 2020-01-18 | 2020-04-28 | 浙江大学 | System and method for preventing high-viscosity ash from depositing on surface of catalyst through multi-element reinforced coupling intelligent regulation |
| US20200319631A1 (en) * | 2019-04-06 | 2020-10-08 | Avanseus Holdings Pte. Ltd. | Method and system for accelerating convergence of recurrent neural network for machine failure prediction |
| CN112833409A (en) * | 2021-01-18 | 2021-05-25 | 江苏方天电力技术有限公司 | Hearth soot blowing optimization method based on dynamic loss prediction |
| CN114225662A (en) * | 2021-12-07 | 2022-03-25 | 国网河北能源技术服务有限公司 | Flue gas desulfurization and denitrification optimization control method based on hysteresis model |
-
2022
- 2022-03-28 CN CN202210314076.6A patent/CN114708924B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069185A (en) * | 2015-07-14 | 2015-11-18 | 东南大学 | Method for establishing air pre-heater clean factor calculation model by using smoke pressure difference, and application |
| US20180024512A1 (en) * | 2016-07-25 | 2018-01-25 | General Electric Company | System modeling, control and optimization |
| CN106599586A (en) * | 2016-12-19 | 2017-04-26 | 北京国能中电节能环保技术股份有限公司 | Neural network-based SCR intelligent ammonia-spraying optimization method and apparatus |
| KR20180076918A (en) * | 2016-12-28 | 2018-07-06 | 대우조선해양 주식회사 | Soot blowing apparatus and method of exhaust system for vessel |
| CN109654517A (en) * | 2018-12-05 | 2019-04-19 | 中北大学 | A kind of boiler soot-blowing optimization method based on the prediction of heating surface health status |
| CN109603545A (en) * | 2018-12-28 | 2019-04-12 | 浙江大学 | A kind of SCR denitration method and device for combining soot blowing |
| US20200319631A1 (en) * | 2019-04-06 | 2020-10-08 | Avanseus Holdings Pte. Ltd. | Method and system for accelerating convergence of recurrent neural network for machine failure prediction |
| CN111068516A (en) * | 2020-01-18 | 2020-04-28 | 浙江大学 | System and method for preventing high-viscosity ash from depositing on surface of catalyst through multi-element reinforced coupling intelligent regulation |
| CN112833409A (en) * | 2021-01-18 | 2021-05-25 | 江苏方天电力技术有限公司 | Hearth soot blowing optimization method based on dynamic loss prediction |
| CN114225662A (en) * | 2021-12-07 | 2022-03-25 | 国网河北能源技术服务有限公司 | Flue gas desulfurization and denitrification optimization control method based on hysteresis model |
Non-Patent Citations (1)
| Title |
|---|
| 孙大瑞: "SCR脱硝过程神经网络建模及控制", 《中国优秀硕士学位论文全文数据库信息科技辑》, 15 March 2018 (2018-03-15), pages 140 - 118 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114708924B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111624876B (en) | Intelligent ammonia injection optimization control system | |
| CN110263395A (en) | The power plant's denitration running optimizatin method and system analyzed based on numerical simulation and data | |
| CN111068518B (en) | Non-uniform ammonia spraying system and method for SCR denitration device | |
| CN107694337A (en) | Coal unit SCR denitrating flue gas control methods based on network response surface | |
| KR102094288B1 (en) | System and method for optimizing boiler combustion | |
| CN110188383B (en) | Selective integration model-based power station SCR denitration modeling method | |
| CN112651166B (en) | Denitration system inlet nitrogen oxide concentration prediction method and device and denitration system | |
| CN113433911B (en) | Accurate control system and method for ammonia spraying of denitration device based on accurate concentration prediction | |
| CN110716512A (en) | Environmental protection equipment performance prediction method based on coal-fired power plant operation data | |
| CN112364562B (en) | Flue gas environment-friendly island cooperative control method and system | |
| CN111522290A (en) | Denitration control method and system based on deep learning method | |
| CN110299188A (en) | SCR flue gas denitrification system GRNN modeling method based on GA variables choice | |
| CN112613237A (en) | CFB unit NOx emission concentration prediction method based on LSTM | |
| CN115145152A (en) | Boiler combustion and denitration process collaborative optimization control method | |
| CN112667613A (en) | Flue gas NOx prediction method and system based on multi-delay characteristic multivariable correction | |
| CN116128136A (en) | LSO-Catboost-based coal-fired power plant boiler NO X Emission prediction method | |
| CN112183872A (en) | Blast furnace gas generation amount prediction method combining generation of countermeasure network and neural network | |
| CN111489605A (en) | Ammonia spraying optimization control simulation system based on Simulink and WinCC | |
| CN117829342A (en) | Coal-fired power plant nitrogen oxide emission prediction method based on improved random forest algorithm | |
| CN114708924A (en) | Model construction method and device for predicting soot blowing interval time of soot blower in SCR system | |
| CN115113519A (en) | Coal-gas co-combustion boiler denitration system outlet NO x Concentration early warning method | |
| CN109933884B (en) | Neural network inverse control method for SCR denitration system of coal-fired unit | |
| CN116341974A (en) | Comprehensive evaluation method for SCR system performance of coal-fired power plant based on big data | |
| CN111219733A (en) | Apparatus for managing combustion optimization and method thereof | |
| CN112308397B (en) | Denitration comprehensive performance evaluation method based on thermal power unit |
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 |