CN113344909B - Method and device for identifying and displaying flame penetration height Wen Lvjing coking of thermal power boiler - Google Patents
Method and device for identifying and displaying flame penetration height Wen Lvjing coking of thermal power boiler Download PDFInfo
- Publication number
- CN113344909B CN113344909B CN202110747508.8A CN202110747508A CN113344909B CN 113344909 B CN113344909 B CN 113344909B CN 202110747508 A CN202110747508 A CN 202110747508A CN 113344909 B CN113344909 B CN 113344909B
- Authority
- CN
- China
- Prior art keywords
- coking
- image
- lvjing
- thermal power
- wen
- 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
- 238000004939 coking Methods 0.000 title claims abstract description 281
- 238000000034 method Methods 0.000 title claims abstract description 62
- 230000035515 penetration Effects 0.000 title claims abstract description 50
- 238000013528 artificial neural network Methods 0.000 claims abstract description 24
- 238000004445 quantitative analysis Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000012549 training Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000009499 grossing Methods 0.000 claims description 9
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 6
- 239000002245 particle Substances 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000001629 suppression Effects 0.000 claims description 2
- 238000013135 deep learning Methods 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 abstract description 5
- 230000008569 process Effects 0.000 description 10
- 238000012545 processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000003745 diagnosis Methods 0.000 description 4
- 239000000571 coke Substances 0.000 description 3
- 238000005235 decoking Methods 0.000 description 3
- 238000003708 edge detection Methods 0.000 description 2
- 230000035699 permeability Effects 0.000 description 2
- 238000010224 classification analysis Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06N3/045—Combinations of 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Radiation Pyrometers (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The application discloses a method and a device for identifying and displaying the high flame penetration Wen Lvjing coking of a thermal power boiler, wherein the method comprises the steps of preprocessing an acquired high flame penetration Wen Lvjing coking image of the thermal power boiler; performing coking recognition based on a neural network prediction algorithm; quantitatively analyzing the identified coking, and calculating the coking fraction; early warning is carried out when the coking score is larger than a first preset threshold value, and intervention measures are adopted when the coking score is larger than a second preset threshold value; showing the coking condition. The method can realize quantitative analysis of the high-flame-permeability Wen Lvjing coking of the thermal power boiler and classification and display of coking by utilizing coking image recognition based on deep learning, and improves detection precision and efficiency. The application discloses a recognition display device for the high Wen Lvjing coking of the flame passing height of a thermal power boiler, which has the same advantages as the method.
Description
Technical Field
The application belongs to the technical field of boiler coking monitoring of thermal power plants, and particularly relates to a method and a device for identifying and displaying the high Wen Lvjing coking of a flame passing height of a thermal power boiler.
Background
Coking of a boiler is a common problem in a coal-fired thermal power plant, and can damage normal combustion conditions, reduce boiler efficiency, damage normal water circulation, even cause pipe explosion accidents, and cause the outlet of a hearth to be blocked to be forced to stop the boiler when serious, so that the economical efficiency and the safety of the operation of the boiler are affected. At present, when a boiler furnace of a thermal power plant is decoking, a professional technician needs to observe and pre-judge the coking state through a fire observation port, and then the decoking is carried out by experience.
The existing boiler coking early warning method based on the convolutional neural network acquires temperature data of a plurality of measuring points in the same time period, obtains image characteristics of the temperature data of the measuring points through the neural network, judges whether the boiler is coked or not, but cannot judge the coking type and directly observe the coking state, and the indirect measurement mode belongs to prediction of small sample data, and is inaccurate in predicting the coking state and degree and cannot be intuitively displayed.
Disclosure of Invention
In order to solve the problems, the application provides a method and a device for identifying and displaying the high flame penetration Wen Lvjing coking of a thermal power boiler, which can realize quantitative analysis of coking and classification and display of coking by utilizing coking image identification based on deep learning, and improve detection precision and efficiency.
The application provides a thermal power boiler flame penetration height Wen Lvjing coking identification display method, which comprises the following steps:
preprocessing an acquired coking image with the flame penetration height Wen Lvjing of the thermal power boiler;
performing coking recognition based on a neural network prediction algorithm;
quantitatively analyzing the identified coking, and calculating the coking fraction;
early warning is carried out when the coking score is larger than a first preset threshold value, and intervention measures are adopted when the coking score is larger than a second preset threshold value;
showing the coking condition.
Preferably, in the method for identifying and displaying the flame penetration height Wen Lvjing coking of the thermal power boiler, the preprocessing the collected coking image includes:
reducing the amount of raw image data based on a weighted average method;
image noise generated by the influence of the environment in the furnace is inhibited based on a median filtering method;
drawing a coking target object based on wavelet transformation;
dividing the original image into coking and background objects based on an iteration method;
and performing corrosion-before-expansion operation on the coking image, eliminating small particle noise, and smoothing the coking boundary.
Preferably, in the method for identifying and displaying the flame penetration height Wen Lvjing coking of the thermal power boiler, the identifying coking based on the neural network prediction algorithm includes:
creating a coking image database with labels, wherein the labels comprise coking blocks, high-temperature coking, others and normal;
dividing the coking image database, taking 60% of training set and 40% of testing set;
creating an enhanced coking image database by setting the image size and immediately overturning, loading an Alex network, and modifying the output classification number of the full-connection layer;
setting learning rate and circulation parameters, and starting training;
reading coking images in the test set, and carrying out prediction classification to obtain accuracy information;
the neural network is optimized by optimizing the learning rate and enhancing the data set distribution.
Preferably, in the method for identifying and displaying the flame penetration height Wen Lvjing coking of the thermal power boiler, the quantitative analysis of the identified coking includes:
selecting a calibration object with a known actual size d;
measuring the pixel numbers D1 and D2 of the calibration material and the coking material in the image, and determining a calibration coefficient D/D1;
and (5) quantitatively calculating the coking area in the target image.
Preferably, in the method for identifying and displaying the coking of the flame penetration height Wen Lvjing of the thermal power boiler, the displaying the coking condition includes:
displaying coking images, coking classification results, coking quantitative analysis results and coking treatment suggestions.
The application provides a thermal power boiler flame penetration height Wen Lvjing coking identification display device, which comprises:
the pretreatment component is used for pretreating the collected coking image with the flame penetration height Wen Lvjing of the thermal power boiler;
the identification component is used for carrying out coking identification based on a neural network prediction algorithm;
an analysis and calculation component for quantitatively analyzing the identified coking and calculating the coking fraction;
the early warning component is used for carrying out early warning when the coking score is larger than a first preset threshold value and taking intervention measures when the coking score is larger than a second preset threshold value;
and the display component is used for displaying the coking condition.
Preferably, in the above-mentioned thermal power boiler flame penetration height Wen Lvjing coking recognition display device, the pretreatment component comprises:
a data amount reducing unit for reducing an original image data amount based on a weighted average method;
the noise suppression unit is used for suppressing image noise generated by the influence of the environment in the furnace based on a median filtering method;
the sketching unit is used for sketching a coking target object based on wavelet transformation;
the background object distinguishing unit is used for dividing the original image into coking objects and background objects based on an iteration method;
and the boundary smoothing unit is used for carrying out corrosion-before-expansion operation on the coking image, eliminating small particle noise and smoothing the coking boundary.
Preferably, in the above-mentioned thermal power boiler flame penetration height Wen Lvjing coking recognition display device, the recognition means includes:
the database creation unit is used for creating a coking image database with labels, wherein the labels comprise coking blocks, high-temperature coking, others and normal;
the segmentation unit is used for segmenting the coking image database, 60% of training sets and 40% of testing sets;
the enhancement database creation unit is used for creating an enhancement coking image database by setting the image size and immediately turning over, loading an Alex network and modifying the output classification number of the full-connection layer;
the training unit is used for setting learning rate and circulation parameters and starting training;
the prediction classification unit is used for reading the coking images in the test set, performing prediction classification and obtaining accuracy information;
and the neural network optimization unit is used for optimizing the neural network by optimizing the learning rate and enhancing the data set distribution.
Preferably, in the above-mentioned thermal power boiler flame penetration height Wen Lvjing coking recognition display device, the analysis and calculation means includes:
a calibration object selecting unit for selecting a calibration object with a known actual size d;
the pixel number measuring unit is used for measuring the pixel numbers D1 and D2 of the calibration object and the coking object in the image and determining a calibration coefficient D/D1;
and the coking area calculation unit is used for quantitatively calculating the coking area in the target image.
Preferably, in the device for identifying and displaying the flame penetration height Wen Lvjing coking of the thermal power boiler, the display component is specifically used for displaying coking images, coking classification results, coking quantitative analysis results and coking treatment suggestions.
As can be seen from the description, the method for identifying and displaying the coking of the flame penetration height Wen Lvjing of the thermal power boiler provided by the application comprises the steps of preprocessing the collected coking image of the flame penetration height Wen Lvjing of the thermal power boiler; performing coking recognition based on a neural network prediction algorithm; quantitatively analyzing the identified coking, and calculating the coking fraction; early warning is carried out when the coking score is larger than a first preset threshold value, and intervention measures are adopted when the coking score is larger than a second preset threshold value; the coking condition is displayed, so that quantitative analysis of coking and coking image recognition based on deep learning can be realized, classification and display of coking are realized, and detection precision and efficiency are improved. The identification display device for the high flame penetration Wen Lvjing coking of the thermal power boiler has the same advantages as the method.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of an embodiment of a method and apparatus for identifying and displaying a high Wen Lvjing coke flame penetration of a thermal power boiler;
FIG. 2 is a schematic diagram of a visual software module adopted by a thermal power boiler flame penetration height Wen Lvjing coking identification display method;
fig. 3 is a schematic diagram of an embodiment of a thermal power boiler flame penetration height Wen Lvjing coking identification display device provided by the application.
Detailed Description
The application provides a method and a device for identifying and displaying the high flame penetration Wen Lvjing coking of a thermal power boiler, which can realize quantitative analysis of the high flame penetration Wen Lvjing coking of the thermal power boiler and classification and display of coking by utilizing coking image identification based on deep learning, and improve detection precision and efficiency.
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
An embodiment of a method for identifying and displaying a high flame penetration Wen Lvjing coking of a thermal power boiler is shown in fig. 1, and fig. 1 is a schematic diagram of an embodiment of a method for identifying and displaying a high flame penetration Wen Lvjing coking of a thermal power boiler, which comprises the following steps:
s1: preprocessing an acquired coking image with the flame penetration height Wen Lvjing of the thermal power boiler;
after pretreatment, the coking part with high flame permeability Wen Lvjing of the thermal power boiler can be more obviously protruded and easily identified, and the identification efficiency is improved.
S2: performing coking recognition based on a neural network prediction algorithm;
specifically, the neural network prediction algorithm can be trained based on an Alex network in MATLAB.
S3: quantitatively analyzing the identified coking, and calculating the coking fraction;
specifically, the coking state may be scored according to the coking type and coking area, for example, may be classified into red, orange, yellow, and green, where red indicates danger and green indicates safety.
S4: early warning is carried out when the coking score is larger than a first preset threshold value, and intervention measures are adopted when the coking score is larger than a second preset threshold value;
specifically, whether early warning is sent out can be judged according to the analyzed coking score, so that an operator can timely react to the coking condition, the situation is more serious due to overlong time is avoided, manual intervention is required to be performed at a time higher than 70 minutes, and intervention measures are required to be performed at a time higher than 90 minutes, so that severe conditions such as overlarge coking and the like are avoided.
S5: showing coking conditions.
The method can particularly but not exclusively display the coking condition by adopting an upper computer screen, and can display various parameters, images, analysis results and the like of the coking at the same time, so that operators can better master the real-time condition of the coking.
As can be seen from the above description, in the embodiment of the identification and display method for the thermal power boiler flame penetration height Wen Lvjing coking provided by the application, the method comprises the steps of preprocessing the collected thermal power boiler flame penetration height Wen Lvjing coking image; performing coking recognition based on a neural network prediction algorithm; quantitatively analyzing the identified coking, and calculating the coking fraction; early warning is carried out when the coking score is larger than a first preset threshold value, and intervention measures are adopted when the coking score is larger than a second preset threshold value; the coking condition is displayed, so that quantitative analysis of coking and coking image recognition based on deep learning can be realized, classification and display of coking are realized, and detection precision and efficiency are improved.
In a specific embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display method, preprocessing the collected thermal power boiler flame penetration height Wen Lvjing coking image may include:
the original image data amount is reduced based on a weighted average method, namely a graying process, so that the processing speed can be increased;
image noise generated by the influence of the environment in the furnace is restrained based on a median filtering method, namely a smooth denoising process, so that the image quality can be improved;
drawing a coking target object based on wavelet transformation to make the characteristic of the coking target object more obvious, namely an edge detection process;
dividing an original image into coking and background images based on an iteration method, namely a binarization process;
and (3) performing corrosion-before-expansion operation on the coking image, eliminating small particle noise, and smoothing the coking boundary, namely an open operation process.
In another specific embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display method, the coking identification based on the neural network prediction algorithm may include the following steps:
creating a coking image database with labels, wherein the labels comprise coking blocks, high-temperature coking, others and normal;
dividing a coking image database, taking 60% of training sets and 40% of testing sets;
creating an enhanced coking image database by setting the image size and immediately overturning, loading an Alex network, and modifying the output classification number of the full-connection layer;
setting learning rate and circulation parameters, starting training, and storing the trained network;
reading coking images in the test set, and carrying out prediction classification to obtain accuracy information;
the neural network is optimized by optimizing the learning rate and enhancing the data set distribution so as to continuously improve the accuracy.
In another specific embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display method, the quantitatively analyzing the identified coking may include:
selecting a calibration object with a known actual size d;
measuring the pixel numbers D1 and D2 of the calibration material and the coking material in the image, and determining a calibration coefficient D/D1;
and (5) quantitatively calculating the coking area in the target image.
That is, the specific size of the coking can be identified in the mode, and when a certain threshold value is reached, an early warning is sent out to make timely treatment, so that the greater risk caused by the coking is avoided.
In a preferred embodiment of the foregoing method for identifying and displaying the flame penetration height Wen Lvjing coking of a thermal power boiler, the displaying the coking may include:
displaying coking images, coking classification results, coking quantitative analysis results and coking treatment suggestions.
Specifically, two interfaces may be displayed in this step: the image processing interface comprises various image processing algorithms to be selected, focal area calculation, original image display, processed image display and other contents; the fault diagnosis interface comprises the display of identification coking images, the display of classification results of coking images, the display of diagnosis results, the display of scores, the display of alarm indicator lamps, the display of treatment method suggestions and the like.
Referring to fig. 2, fig. 2 is a schematic diagram of a visualization software module adopted by the high-flame-permeability Wen Lvjing coking identification display method of the thermal power boiler, which comprises a coking image reading module, a coking image processing module, a coking image classification identification module, a coking image display module, a coking image quantitative analysis module, an alarm module, a result display module and a suggestion module, wherein after the coking image is analyzed and processed by the coking image reading module and the image processing module, the coking image display module respectively displays an original image and the processed image in real time and intuitively, and simultaneously, the coking image is subjected to classification identification and quantitative analysis by the coking image classification identification module and the coking quantitative analysis module, the classification result display module and the quantitative analysis result display module are used for displaying the results, the alarm module is used for carrying out threshold judgment and scoring according to the classification result and the quantitative analysis result, if the score exceeds a set threshold, a red lamp is lighted for alarm, a picture window is also started at the moment, a first page of the coking single-frame image at the moment is amplified and automatically stored in a coking image file, if the score does not exceed the threshold, and the suggestion module does not respond later, and the suggestion module is also used for carrying out corresponding operation and prompting analysis according to the classification result.
In summary, by using the method for identifying and displaying the high flame penetration Wen Lvjing coking of the thermal power boiler, coking identification classification and quantitative analysis can be performed, a basis is provided for formulating a decoking strategy, human factors are overcome, and safe and economic operation of the boiler is ensured.
An embodiment of a thermal power boiler flame penetration height Wen Lvjing coking identification display device provided by the application is shown in fig. 3, fig. 3 is a schematic diagram of an embodiment of a thermal power boiler flame penetration height Wen Lvjing coking identification display device provided by the application, which can be integrated in an upper computer and comprises:
the preprocessing component 301 is used for preprocessing the collected coking image with high flame permeability Wen Lvjing of the thermal power boiler, and it is noted that after preprocessing, the coking part can be more obviously protruded and easily identified, so that the identification efficiency is improved;
the identifying component 302 is configured to perform coking identification based on a neural network prediction algorithm, and specifically, the neural network prediction algorithm may be trained based on an Alex network in MATLAB;
an analysis and calculation unit 303, configured to quantitatively analyze the identified coke, calculate a coking score, specifically, score a coking status according to a coking type and a coking area, for example, may be classified into red, orange, yellow, and green, where red indicates danger, and green indicates safety;
the early warning component 304 is configured to perform early warning when the coking score is greater than a first preset threshold value, and take intervention measures when the coking score is greater than a second preset threshold value, specifically, whether to send out early warning can be judged according to the analyzed coking score, so that an operator can timely respond to the coking condition, and serious conditions caused by overlong time are avoided, wherein the situation can be but is not limited to manual intervention when the coking score is greater than 70, and intervention measures are taken when the coking score is greater than 90, so that severe conditions such as overlarge coking condition are avoided;
the display unit 305 is used for displaying the coking condition, and can particularly but not exclusively display the coking condition by using an upper computer screen, and can display various parameters, images, analysis results and the like of the coking at the same time, so that operators can better master the real-time condition of the coking.
In a specific embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display device, the pretreatment component may include:
a data amount reducing unit for reducing the amount of original image data based on a weighted average method, that is, a graying process, which can increase the processing speed;
noise suppressing unit for suppressing image noise generated by environmental influence in furnace based on median filtering, that is, a smooth denoising process, so that image quality can be improved
A sketching unit, configured to sketch a coking target object based on wavelet transformation, which is an edge detection process;
the background object distinguishing unit is used for dividing the original image into coking and background objects based on an iteration method, namely a binarization process;
and the boundary smoothing unit is used for performing corrosion and expansion operation on the coking image, eliminating small particle noise and smoothing the coking boundary, namely an open operation process.
In another specific embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display device, the identification component may include:
the database creation unit is used for creating a coking image database with labels, wherein the labels comprise coke blocks, high-temperature coking, others and normal;
the segmentation unit is used for segmenting the coking image database, 60% of training sets and 40% of testing sets;
the enhancement database creation unit is used for creating an enhancement coking image database by setting the image size and immediately turning over, loading an Alex network and modifying the output classification number of the full-connection layer;
the training unit is used for setting learning rate and circulation parameters and starting training;
the prediction classification unit is used for reading coking images in the test set, performing prediction classification and obtaining accuracy information;
and the neural network optimization unit is used for optimizing the neural network by optimizing the learning rate and enhancing the data set distribution.
In another specific embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display device, the analysis and calculation component may include:
a calibration object selecting unit for selecting a calibration object with a known actual size d;
the pixel number measuring unit is used for measuring the pixel numbers D1 and D2 of the calibration object and the coking object in the image and determining a calibration coefficient D/D1;
and the coking area calculation unit is used for quantitatively calculating the coking area in the target image.
That is, the specific size of the coking can be identified in the mode, and when a certain threshold value is reached, an early warning is sent out to make timely treatment, so that the greater risk caused by the coking is avoided.
In a preferred embodiment of the foregoing thermal power boiler flame penetration height Wen Lvjing coking identification display device, the display component may be specifically configured to display a coking image, a coking classification result, a coking quantitative analysis result, and a coking treatment recommendation.
Two interfaces may be displayed: the image processing interface comprises various image processing algorithms to be selected, focal area calculation, original image display, processed image display and other contents; the fault diagnosis interface comprises the display of identification coking images, the display of classification results of coking images, the display of diagnosis results, the display of scores, the display of alarm indicator lamps, the display of treatment method suggestions and the like.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. The method for identifying and displaying the high flame penetration Wen Lvjing coking of the thermal power boiler is characterized by comprising the following steps of:
preprocessing an acquired coking image with the flame penetration height Wen Lvjing of the thermal power boiler;
performing coking recognition based on a neural network prediction algorithm;
quantitatively analyzing the identified coking, and calculating the coking fraction;
early warning is carried out when the coking score is larger than a first preset threshold value, and intervention measures are adopted when the coking score is larger than a second preset threshold value;
displaying the coking condition;
the preprocessing of the collected thermal power boiler flame penetration height Wen Lvjing coking image comprises the following steps:
reducing the amount of raw image data based on a weighted average method;
image noise generated by the influence of the environment in the furnace is inhibited based on a median filtering method;
drawing a coking target object based on wavelet transformation;
dividing the original image into coking and background objects based on an iteration method;
performing corrosion-before-expansion operation on the coking image, eliminating small particle noise, and smoothing a coking boundary;
the coking identification based on the neural network prediction algorithm comprises the following steps:
creating a coking image database with labels, wherein the labels comprise coking blocks, high-temperature coking, others and normal;
dividing the coking image database, taking 60% of training set and 40% of testing set;
creating an enhanced coking image database by setting the image size and immediately overturning, loading an Alex network, and modifying the output classification number of the full-connection layer;
setting learning rate and circulation parameters, and starting training;
reading coking images in the test set, and carrying out prediction classification to obtain accuracy information;
the neural network is optimized by optimizing the learning rate and enhancing the data set distribution.
2. The method for identifying and displaying the high flame penetration Wen Lvjing coking of the thermal power boiler according to claim 1, wherein the quantitative analysis of the identified coking comprises:
selecting a calibration object with a known actual size d;
measuring the pixel numbers D1 and D2 of the calibration material and the coking material in the image, and determining a calibration coefficient D/D1;
and (5) quantitatively calculating the coking area in the target image.
3. The method for identifying and displaying the coking of the high flame penetration Wen Lvjing of the thermal power boiler according to claim 2, wherein the displaying the coking condition comprises:
displaying coking images, coking classification results, coking quantitative analysis results and coking treatment suggestions.
4. The utility model provides a thermal power boiler passes through high Wen Lvjing coking discernment display device of flame which characterized in that includes:
the pretreatment component is used for pretreating the collected coking image with the flame penetration height Wen Lvjing of the thermal power boiler;
the identification component is used for carrying out coking identification based on a neural network prediction algorithm;
an analysis and calculation component for quantitatively analyzing the identified coking and calculating the coking fraction;
the early warning component is used for carrying out early warning when the coking score is larger than a first preset threshold value and taking intervention measures when the coking score is larger than a second preset threshold value;
a display unit for displaying the coking condition;
the preprocessing component includes:
a data amount reducing unit for reducing an original image data amount based on a weighted average method;
the noise suppression unit is used for suppressing image noise generated by the influence of the environment in the furnace based on a median filtering method;
the sketching unit is used for sketching a coking target object based on wavelet transformation;
the background object distinguishing unit is used for dividing the original image into coking objects and background objects based on an iteration method;
the boundary smoothing unit is used for carrying out corrosion-before-expansion operation on the coking image, eliminating small particle noise and smoothing the coking boundary;
the identification means includes:
the database creation unit is used for creating a coking image database with labels, wherein the labels comprise coking blocks, high-temperature coking, others and normal;
the segmentation unit is used for segmenting the coking image database, 60% of training sets and 40% of testing sets;
the enhancement database creation unit is used for creating an enhancement coking image database by setting the image size and immediately turning over, loading an Alex network and modifying the output classification number of the full-connection layer;
the training unit is used for setting learning rate and circulation parameters and starting training;
the prediction classification unit is used for reading the coking images in the test set, performing prediction classification and obtaining accuracy information;
and the neural network optimization unit is used for optimizing the neural network by optimizing the learning rate and enhancing the data set distribution.
5. The thermal power boiler flame penetration height Wen Lvjing coking identification display device of claim 4, wherein the analysis and calculation means comprises:
a calibration object selecting unit for selecting a calibration object with a known actual size d;
the pixel number measuring unit is used for measuring the pixel numbers D1 and D2 of the calibration object and the coking object in the image and determining a calibration coefficient D/D1;
and the coking area calculation unit is used for quantitatively calculating the coking area in the target image.
6. The thermal power boiler flame penetration height Wen Lvjing coking recognition display device of claim 5, wherein the display component is specifically used for displaying coking images, coking classification results, coking quantitative analysis results and coking treatment suggestions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110747508.8A CN113344909B (en) | 2021-07-01 | 2021-07-01 | Method and device for identifying and displaying flame penetration height Wen Lvjing coking of thermal power boiler |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110747508.8A CN113344909B (en) | 2021-07-01 | 2021-07-01 | Method and device for identifying and displaying flame penetration height Wen Lvjing coking of thermal power boiler |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113344909A CN113344909A (en) | 2021-09-03 |
| CN113344909B true CN113344909B (en) | 2023-12-08 |
Family
ID=77482194
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110747508.8A Active CN113344909B (en) | 2021-07-01 | 2021-07-01 | Method and device for identifying and displaying flame penetration height Wen Lvjing coking of thermal power boiler |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344909B (en) |
Citations (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000132080A (en) * | 1998-10-22 | 2000-05-12 | Kawasaki Heavy Ind Ltd | Fire simulation device for fire extinguishing training |
| JP2009236391A (en) * | 2008-03-27 | 2009-10-15 | Japan Atom Power Co Ltd:The | High temperature atmosphere furnace observation device |
| CN101655700A (en) * | 2009-09-17 | 2010-02-24 | 同济大学 | Topographic control system and method of waste pyrolysis |
| CN101688127A (en) * | 2007-06-04 | 2010-03-31 | 埃克森美孚化学专利公司 | A process for pyrolyzing a hydrocarbon feedstock pyrolysis reactor system |
| CN101799661A (en) * | 2007-01-10 | 2010-08-11 | 株式会社日立制作所 | The training operator device of the control device of boiler plant and boiler plant |
| CN102425807A (en) * | 2011-11-23 | 2012-04-25 | 华北电力大学（保定） | Combustion feedforward and feedback composite optimization controlling method for pulverized coal fired boiler |
| CN102707762A (en) * | 2012-06-04 | 2012-10-03 | 同济大学 | Heating power control method for waste high polymer material pyrolysis |
| CN202692120U (en) * | 2012-07-09 | 2013-01-23 | 大唐双鸭山热电有限公司 | Visual monitoring control module in boiler hearth |
| CN105046868A (en) * | 2015-06-16 | 2015-11-11 | 苏州华启智能科技股份有限公司 | Fire early warning method based on infrared thermal imager in narrow environment |
| CN105423273A (en) * | 2015-12-15 | 2016-03-23 | 天津鹰麟节能科技发展有限公司 | Spectroscopic boiler anti-coking system and control method |
| CN107559878A (en) * | 2017-09-25 | 2018-01-09 | 华能国际电力股份有限公司日照电厂 | The method and device of fire coal management |
| CN109063412A (en) * | 2018-08-27 | 2018-12-21 | 浙江大学 | Source Data Fusion System and method for the assessment of coal pyrolysis in plasma producing acetylene reactiveness |
| CN109145689A (en) * | 2017-06-28 | 2019-01-04 | 南京理工大学 | A kind of robot fire detection method |
| CN208421626U (en) * | 2018-07-23 | 2019-01-22 | 合肥金星机电科技发展有限公司 | Ethane cracking furnace burning process monitors system |
| CN109934417A (en) * | 2019-03-26 | 2019-06-25 | 国电民权发电有限公司 | Boiler coke method for early warning based on convolutional neural networks |
| CN109977838A (en) * | 2019-03-20 | 2019-07-05 | 西安理工大学 | A kind of flame combustion state detection method |
| CN110197199A (en) * | 2019-04-17 | 2019-09-03 | 广东石油化工学院 | Embedded DCNN and the weight tube temperature degree recognition methods of the pyrolysis furnace of edge calculations |
| CN110222814A (en) * | 2019-04-25 | 2019-09-10 | 广东石油化工学院 | The pipe recognition methods again of Ethylene Cracking Furnace Tubes based on embedded DCNN |
| CN110222633A (en) * | 2019-06-04 | 2019-09-10 | 北京工业大学 | City solid waste burning process combusts operating mode's switch method based on flame image color feature extracted |
| CN110419018A (en) * | 2016-12-29 | 2019-11-05 | 奇跃公司 | The automatic control of wearable display device based on external condition |
| CN110633675A (en) * | 2019-09-18 | 2019-12-31 | 东北大学 | System and method for identifying fire in video based on convolutional neural network |
| CN111062293A (en) * | 2019-12-10 | 2020-04-24 | 太原理工大学 | Unmanned aerial vehicle forest flame identification method based on deep learning |
| CN111095261A (en) * | 2017-04-27 | 2020-05-01 | 视网膜病答案有限公司 | Automatic analysis system and method for fundus images |
| CN111368771A (en) * | 2020-03-11 | 2020-07-03 | 四川路桥建设集团交通工程有限公司 | Tunnel fire early warning method and device based on image processing, computer equipment and computer readable storage medium |
| CN111473358A (en) * | 2020-05-09 | 2020-07-31 | 中国华能集团有限公司 | Hearth flame television decoking device and control method thereof |
| CN111882810A (en) * | 2020-07-31 | 2020-11-03 | 广州市微智联科技有限公司 | Fire identification and early warning method and system |
| CN112080745A (en) * | 2020-09-02 | 2020-12-15 | 新疆金泰非晶科技有限公司 | Composite coating containing amorphous alloy identification layer and preparation method and application thereof |
| CN112521955A (en) * | 2020-11-04 | 2021-03-19 | 中南大学 | Coke cake center temperature detection method and system |
| CN112784405A (en) * | 2021-01-06 | 2021-05-11 | 润电能源科学技术有限公司 | Boiler slagging prediction method based on numerical simulation and related device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8375032B2 (en) * | 2009-06-25 | 2013-02-12 | University Of Tennessee Research Foundation | Method and apparatus for predicting object properties and events using similarity-based information retrieval and modeling |
-
2021
- 2021-07-01 CN CN202110747508.8A patent/CN113344909B/en active Active
Patent Citations (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000132080A (en) * | 1998-10-22 | 2000-05-12 | Kawasaki Heavy Ind Ltd | Fire simulation device for fire extinguishing training |
| CN101799661A (en) * | 2007-01-10 | 2010-08-11 | 株式会社日立制作所 | The training operator device of the control device of boiler plant and boiler plant |
| CN101688127A (en) * | 2007-06-04 | 2010-03-31 | 埃克森美孚化学专利公司 | A process for pyrolyzing a hydrocarbon feedstock pyrolysis reactor system |
| JP2009236391A (en) * | 2008-03-27 | 2009-10-15 | Japan Atom Power Co Ltd:The | High temperature atmosphere furnace observation device |
| CN101655700A (en) * | 2009-09-17 | 2010-02-24 | 同济大学 | Topographic control system and method of waste pyrolysis |
| CN102425807A (en) * | 2011-11-23 | 2012-04-25 | 华北电力大学（保定） | Combustion feedforward and feedback composite optimization controlling method for pulverized coal fired boiler |
| CN102707762A (en) * | 2012-06-04 | 2012-10-03 | 同济大学 | Heating power control method for waste high polymer material pyrolysis |
| CN202692120U (en) * | 2012-07-09 | 2013-01-23 | 大唐双鸭山热电有限公司 | Visual monitoring control module in boiler hearth |
| CN105046868A (en) * | 2015-06-16 | 2015-11-11 | 苏州华启智能科技股份有限公司 | Fire early warning method based on infrared thermal imager in narrow environment |
| CN105423273A (en) * | 2015-12-15 | 2016-03-23 | 天津鹰麟节能科技发展有限公司 | Spectroscopic boiler anti-coking system and control method |
| CN110419018A (en) * | 2016-12-29 | 2019-11-05 | 奇跃公司 | The automatic control of wearable display device based on external condition |
| CN111095261A (en) * | 2017-04-27 | 2020-05-01 | 视网膜病答案有限公司 | Automatic analysis system and method for fundus images |
| CN109145689A (en) * | 2017-06-28 | 2019-01-04 | 南京理工大学 | A kind of robot fire detection method |
| CN107559878A (en) * | 2017-09-25 | 2018-01-09 | 华能国际电力股份有限公司日照电厂 | The method and device of fire coal management |
| CN208421626U (en) * | 2018-07-23 | 2019-01-22 | 合肥金星机电科技发展有限公司 | Ethane cracking furnace burning process monitors system |
| CN109063412A (en) * | 2018-08-27 | 2018-12-21 | 浙江大学 | Source Data Fusion System and method for the assessment of coal pyrolysis in plasma producing acetylene reactiveness |
| CN109977838A (en) * | 2019-03-20 | 2019-07-05 | 西安理工大学 | A kind of flame combustion state detection method |
| CN109934417A (en) * | 2019-03-26 | 2019-06-25 | 国电民权发电有限公司 | Boiler coke method for early warning based on convolutional neural networks |
| CN110197199A (en) * | 2019-04-17 | 2019-09-03 | 广东石油化工学院 | Embedded DCNN and the weight tube temperature degree recognition methods of the pyrolysis furnace of edge calculations |
| CN110222814A (en) * | 2019-04-25 | 2019-09-10 | 广东石油化工学院 | The pipe recognition methods again of Ethylene Cracking Furnace Tubes based on embedded DCNN |
| CN110222633A (en) * | 2019-06-04 | 2019-09-10 | 北京工业大学 | City solid waste burning process combusts operating mode's switch method based on flame image color feature extracted |
| CN110633675A (en) * | 2019-09-18 | 2019-12-31 | 东北大学 | System and method for identifying fire in video based on convolutional neural network |
| CN111062293A (en) * | 2019-12-10 | 2020-04-24 | 太原理工大学 | Unmanned aerial vehicle forest flame identification method based on deep learning |
| CN111368771A (en) * | 2020-03-11 | 2020-07-03 | 四川路桥建设集团交通工程有限公司 | Tunnel fire early warning method and device based on image processing, computer equipment and computer readable storage medium |
| CN111473358A (en) * | 2020-05-09 | 2020-07-31 | 中国华能集团有限公司 | Hearth flame television decoking device and control method thereof |
| CN111882810A (en) * | 2020-07-31 | 2020-11-03 | 广州市微智联科技有限公司 | Fire identification and early warning method and system |
| CN112080745A (en) * | 2020-09-02 | 2020-12-15 | 新疆金泰非晶科技有限公司 | Composite coating containing amorphous alloy identification layer and preparation method and application thereof |
| CN112521955A (en) * | 2020-11-04 | 2021-03-19 | 中南大学 | Coke cake center temperature detection method and system |
| CN112784405A (en) * | 2021-01-06 | 2021-05-11 | 润电能源科学技术有限公司 | Boiler slagging prediction method based on numerical simulation and related device |
Non-Patent Citations (5)
| Title |
|---|
| Guided Image Filtering;HE. K;《IEEE Transactions on Pattern Analysis & Machine Intelligence》;第1397-1409页 * |
| 基于形态学的快速图像搜索与识别算法;常锐;裴海龙;;计算机工程与设计(07);第93-95页 * |
| 基于模糊聚类的稳健支撑向量回归机及火焰图像处理;陈晓峰;王士同;曹苏群;崔运静;马培勇;仇性启;;中国图象图形学报(03);第92-99页 * |
| 大数据技术应用于火电机组深度调峰的研究;陈鑫;刘利;徐威;陈啸;;科技创新与应用(28);第156-157页 * |
| 锅炉性能优化***的应用试验研究;王秀林,孔令君,姜仕涛,张竟思,付全方,张世荣,杨启璋;中国电力(12);第49-52页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113344909A (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110097034B (en) | Intelligent face health degree identification and evaluation method | |
| US20220300809A1 (en) | Data generation system, learning apparatus, data generation apparatus, data generation method, and computer-readable storage medium storing a data generation program | |
| CN113592828B (en) | Nondestructive testing method and system based on industrial endoscope | |
| CN114581764B (en) | Underground structure crack disease discriminating method based on deep learning algorithm | |
| US20220133215A1 (en) | Method for evaluating skin lesions using artificial intelligence | |
| CN111680729A (en) | Airport passenger security inspection dangerous goods real-time high-recognition rate method based on X-ray machine image | |
| CN107576660B (en) | A kind of double yellow duck egg Automatic Visual Inspection method based on apart from contour | |
| CN113935666A (en) | Building decoration wall tile abnormity evaluation method based on image processing | |
| CN115995056A (en) | Automatic bridge disease identification method based on deep learning | |
| Nabilah et al. | Classification of ovarian cysts on ultrasound images using watershed segmentation and contour analysis | |
| CN113344909B (en) | Method and device for identifying and displaying flame penetration height Wen Lvjing coking of thermal power boiler | |
| CN116385953B (en) | Railway wagon door hinge breaking fault image identification method | |
| CN112818806A (en) | Transformer substation inspection robot auxiliary navigation method based on deep learning | |
| CN111860437A (en) | Method and device for judging fatigue degree based on facial expression | |
| CN115937085B (en) | Nuclear cataract image processing method based on neural network learning | |
| CN116542978A (en) | Quality detection method and device for FISH probe | |
| CN110751687A (en) | Apple size grading method based on computer vision minimum and maximum circles | |
| CN115601701A (en) | Insufficient sintering ignition strength judgment method and device based on machine vision | |
| RU2426069C1 (en) | Rail surface quality nondestructive examination | |
| KR101782366B1 (en) | Vision inspection method based on learning data using normalization of sample scale | |
| CN114841968A (en) | Fatigue state detection method based on YOLOv5 | |
| CN114973104A (en) | Dynamic flame detection algorithm and system based on video image | |
| CN113283322A (en) | Livestock trauma detection method, device, equipment and storage medium | |
| CN111340802A (en) | Intelligent cloth inspecting machine capable of automatically inspecting cloth by adopting artificial intelligence technology | |
| CN112767365A (en) | Flaw detection method |
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 |