Boiler combustion control method and system based on garbage visual data analysis

By using a boiler combustion control method based on visual data analysis of waste, image analysis technology is used to extract characteristic factors of the waste mixture and automatically adjust the boiler control strategy, which solves the shortcomings of traditional manual control and achieves a highly efficient and safe waste incineration process.

CN118794026BActive Publication Date: 2024-11-26HEBEI UNIVERSITY +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411288099.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-14
Publication Date
2024-11-26
Estimated Expiration
2044-09-14

AI Technical Summary

Technical Problem

Combustion control in traditional waste incinerators relies on manual experience, which presents problems such as high labor intensity, significant impact from individual ability, and high risk.

Method used

A boiler combustion control method based on waste visual data analysis is adopted. By capturing and visually analyzing images of the waste mixture, color distribution features and edge distribution features are extracted to form a real-time feature factor profile. This profile is then compared with a preset reference feature factor profile library to determine the optimal boiler control strategy.

Benefits of technology

It enables intelligent monitoring and optimized control of the waste incineration process, improving combustion efficiency and environmental performance while reducing health risks to operators.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118794026B_ABST
    Figure CN118794026B_ABST
Patent Text Reader

Abstract

The present invention discloses a boiler combustion control method and system based on garbage visual data analysis, which relates to the technical field of boiler control, and specifically discloses that the mixed components of pre-treated garbage mixture are determined, and a corresponding boiler control strategy judgment model is selected according to the mixed components; secondly, the garbage mixture is imaged, and color distribution characteristics and edge distribution characteristics are extracted through visual analysis; then, a real-time characteristic factor portrait of the garbage mixture is formed based on these characteristic factors, and the characteristic factor branches in the portrait are adjusted according to the color distribution characteristics; finally, the boiler control strategy judgment model is used to compare the real-time characteristic factor portrait with a preset reference characteristic factor portrait library, and the optimal boiler control strategy is determined, and applied to the actual control of the boiler. The above technical scheme realizes intelligent monitoring and optimization control of the garbage incineration process, and improves combustion efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of boiler control, and in particular to a boiler combustion control method and system based on garbage visual data analysis. Background Art

[0002] In the field of waste incineration, boiler combustion control is the core link to ensure efficient and environmentally friendly waste treatment, and plays a vital role in improving overall treatment efficiency and reducing environmental pollution. Traditionally, the operation of waste incinerators mainly depends on the experience and judgment of operators. They observe the state, color, height and other characteristics of the flame with their naked eyes to adjust the movement and combustion conditions of the grate to maintain a stable combustion process; however, this method has many limitations. First, it requires operators to have rich experience and keen insight to accurately judge the flame state and make corresponding adjustments. This not only increases the labor intensity requirements, but also makes the effect of combustion control affected by the personal ability and experience of the operator; secondly, there is a certain danger in observing the flame state with the naked eye; because waste incinerators are usually in a high temperature and high pressure working environment, operators need to be exposed to high temperature and toxic and harmful gases for a long time, which poses a potential threat to their health. Summary of the invention

[0003] The purpose of the present invention is to provide a boiler control method and system which can avoid manual regulation of garbage combustion.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] The boiler combustion control method based on garbage visual data analysis includes:

[0006] Determine the mixed components of the pre-treated garbage mixture, and based on the determined mixed components, determine the boiler control strategy judgment model to be applied in the preset mixed component-boiler control strategy judgment model correspondence relationship;

[0007] Taking an image of the garbage mixture to obtain a garbage mixture map, visually analyzing the garbage mixture map to determine the color distribution characteristics and edge distribution characteristics in the map;

[0008] Based on the color distribution characteristics of the garbage mixture image and the characteristic factors of the edge distribution characteristics, a real-time characteristic factor image of the garbage mixture image is formed, and based on the characteristic conspicuity judgment of the color distribution characteristics, the corresponding characteristic factor branches in the real-time characteristic factor image are adjusted;

[0009] The boiler control strategy judgment model is used to substitute the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval, and the matching reference characteristic factor portrait is determined. Based on the determined reference characteristic factor portrait, the optimal boiler control strategy is determined, and the boiler is controlled using the optimal boiler control strategy.

[0010] In some embodiments disclosed in the present invention, the method for constructing the corresponding relationship between the mixed component and the boiler control strategy judgment model includes:

[0011] Based on the major categories of garbage components appearing in the mixed components of the burned garbage mixture, several major category groups of garbage components are formed, and based on the garbage component subcategories that can be determined in each major category of garbage components, secondary subdivision is performed, and the subdivided garbage component subcategories are used to supplement the data of the major category groups to which they belong;

[0012] The boiler combustion conditions and boiler control strategies corresponding to each garbage component group are collected and analyzed to determine the combustion characteristics in the boiler at different time nodes and the boiler control methods at different time nodes. Based on the combustion characteristics in the boiler, the boiler control methods at different time nodes are corrected, and the corrected boiler control methods are associated with the garbage component groups to obtain several corresponding relationships between component groups and boiler control strategies.

[0013] A visual analysis is performed on the garbage mixture map corresponding to each major component group~boiler control strategy, and the color distribution characteristics and edge distribution characteristics in the map are determined. Based on the color distribution characteristics and edge distribution characteristics, a reference characteristic factor portrait is constructed, and a correlation is established between the reference characteristic factor portrait and the corresponding relationship between the major component group~boiler control strategy.

[0014] In some embodiments disclosed in the present invention, a method for forming a real-time characteristic factor portrait of a garbage mixture map includes:

[0015] A characteristic factor portrait template is set, and the characteristic factor portrait template includes a plurality of characteristic factor expression lines, one end of the characteristic factor expression lines overlaps with each other, and the characteristic factor expression lines are evenly dispersed in a circular manner, and the characteristic factor expression lines correspond to a color collection point ratio of a specific color parameter interval, or a specific edge distribution density parameter, or a specific average edge length;

[0016] Based on the feature conspicuity judgment of the color distribution feature, the first amplification factor of the feature factor expression line corresponding to the color distribution feature is determined, and the length of the feature factor expression line is determined in combination with the parameters corresponding to each feature factor.

[0017] In some embodiments disclosed in the present invention, the method for judging the feature prominence of each feature factor includes:

[0018] The color collection point ratios of the color parameter intervals corresponding to all garbage mixed material images are clustered, and the clustering analysis method includes: identifying the color parameter intervals with color collection point ratios greater than or equal to 0 as valid color parameter intervals, marking and counting the valid color parameter intervals with color collection point ratios greater than or equal to a preset value, and recording them as the number of valid color parameter intervals;

[0019] The effective distinguishable number of mixed components in the garbage mixture in the historical records is counted, and the method for determining the effective distinguishable number includes classifying the mixed components, and performing a combustion test on the mixed components of each category, recording the mixed components with the same combustion characteristics and belonging to the same category as effective distinguishable components, and counting the total number of components of the effective distinguishable components;

[0020] Based on the effective color distinction difference between the number of effective color parameter intervals and the total number of components, the first amplification factor of the characteristic factor expression line corresponding to the color distribution feature is determined.

[0021] In some embodiments disclosed in the present invention, the expression for determining the first amplification factor is: ;

[0022] in, is the first amplification factor, is the amplification factor conversion factor, The difference between the preset effective color distinction standards, is the effective color difference, The adjustment coefficient for the difference in effective color distinction. Adjusts the constant for the amount of difference in effective color distinction.

[0023] In some embodiments disclosed in the present invention, a method for determining a matching reference feature factor portrait includes:

[0024] Determining the number of characteristic factor expression lines in the real-time characteristic factor portrait, and performing a first screening of reference characteristic factor portraits in the reference characteristic factor portrait library based on the number of characteristic factor expression lines;

[0025] The average length of the characteristic factor expression line in the real-time characteristic factor portrait is determined, and based on the average length of the characteristic factor expression line, the reference characteristic factor portrait after the first screening is screened for the second time to obtain several reference characteristic factor portraits participating in the comparison;

[0026] The reference characteristic factor portraits involved in the comparison are compared with the real-time characteristic factor portraits respectively, and the degree of matching between the two is calculated. If the degree of matching between the two is greater than or equal to the preset value, the corresponding reference characteristic factor portrait is deemed to match the real-time characteristic factor portrait.

[0027] In some embodiments disclosed in the present invention, the method for calculating the degree of matching between the reference feature factor portrait and the real-time feature factor portrait includes: ;

[0028] in, For the matching degree, is the first matching degree weight coefficient, is the second matching degree weight coefficient, is the third matching degree weight coefficient, is the color distribution feature corresponding to The characteristic factor expresses the difference in line length, is the total number of characteristic factor expression lines corresponding to the color distribution characteristics, is the length difference of the characteristic factor expression line corresponding to the edge distribution density parameter, The characteristic factor corresponding to the average edge length expresses the length difference of the line.

[0029] In some embodiments disclosed in the present invention, there is also disclosed a boiler combustion control system based on garbage visual data analysis, comprising:

[0030] The first module is used to determine the mixed components of the pre-treated garbage mixture, and based on the determined mixed components, determine the boiler control strategy judgment model to be applied in the preset mixed component-boiler control strategy judgment model correspondence relationship;

[0031] The second module is used to take an image of the garbage mixture to obtain a garbage mixture map, perform visual analysis on the garbage mixture map, and determine the color distribution characteristics and edge distribution characteristics in the map;

[0032] The third module is used to form a real-time characteristic factor portrait of the garbage mixture image based on the color distribution characteristics and the characteristic factors of the edge distribution characteristics of the garbage mixture image, and adjust the corresponding characteristic factor branches in the real-time characteristic factor portrait based on the characteristic conspicuity judgment of the color distribution characteristics;

[0033] The fourth module is used to use the boiler control strategy judgment model to substitute the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval, determine the matching reference characteristic factor portrait, and determine the optimal boiler control strategy based on the determined reference characteristic factor portrait, and use the optimal boiler control strategy to control the boiler.

[0034] The present invention discloses a boiler combustion control method and system based on garbage visual data analysis, which relates to the technical field of boiler control, and specifically discloses that the mixed components of pre-treated garbage mixture are determined, and a corresponding boiler control strategy judgment model is selected according to the mixed components; secondly, the garbage mixture is imaged, and color distribution characteristics and edge distribution characteristics are extracted through visual analysis; then, a real-time characteristic factor portrait of the garbage mixture is formed based on these characteristic factors, and the characteristic factor branches in the portrait are adjusted according to the color distribution characteristics; finally, the boiler control strategy judgment model is used to compare the real-time characteristic factor portrait with a preset reference characteristic factor portrait library, and the optimal boiler control strategy is determined, and applied to the actual control of the boiler. The above technical scheme realizes intelligent monitoring and optimization control of the garbage incineration process, and improves combustion efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a method step diagram of a boiler combustion control method based on garbage visual data analysis disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0037] The purpose of the present invention is to provide a boiler control method and system which can avoid manual regulation of garbage combustion.

[0038] In order to achieve the above object, the present invention adopts the following technical solutions:

[0039] Boiler combustion control method based on garbage visual data analysis, see Figure 1 ,include:

[0040] Step S100, determining the mixed components of the pre-treated garbage mixture, and based on the determined mixed components, determining the boiler control strategy judgment model to be applied in a preset mixed component-boiler control strategy judgment model correspondence relationship.

[0041] The core purpose of this step is to improve the accuracy of the boiler control strategy judgment model, and the key to achieving this goal is to accurately determine the mixed composition of the garbage mixture. Through a detailed analysis of the garbage mixture, its main components, such as plastics, rubber, waste paper, etc., can be clearly identified. The information of these main materials is crucial for the subsequent combustion process because they directly affect the combustion efficiency, calorific value release and environmental performance. Once these mixed components are accurately determined, the control strategy model that best suits the current garbage mixture characteristics can be found in the correspondence between the preset mixed components and the boiler control strategy judgment model. In this way, it can be ensured that the selected control strategy can be optimized for a specific garbage mixture, thereby achieving a more efficient and environmentally friendly combustion process.

[0042] In some embodiments disclosed in the present invention, the method for constructing the corresponding relationship between the mixed component and the boiler control strategy judgment model includes:

[0043] Step S101, based on the major categories of garbage components appearing in the mixed components of the burned garbage mixture, several major category groups of garbage components are formed, and based on the minor categories of garbage components that can be determined in each major category of garbage components, secondary subdivision is performed, and the subdivided garbage component minor categories are used to supplement the data of the major category groups to which they belong.

[0044] The core of step S101 is to carefully classify and sort the mixed components of the garbage mixture. First, identify the main categories of garbage components in the burned garbage mixture, and divide these categories into different groups. Then, further subdivide the garbage components in each major group, identify the subcategories that can be clearly distinguished, and add the data of these subcategories to the major groups to which they belong. The purpose of this step is to establish a detailed and orderly garbage component classification system to provide accurate garbage component data for subsequent boiler combustion analysis and control strategy formulation.

[0045] Step S102, collect and analyze the boiler combustion conditions and boiler control strategies corresponding to each major garbage component group, determine the combustion characteristics in the boiler at different time nodes and the boiler control methods at different time nodes, and based on the combustion characteristics in the boiler, correct the boiler control methods at different time nodes, and associate the corrected boiler control methods with the major garbage component groups and record them, to obtain several major component group~boiler control strategy correspondences.

[0046] The focus of step S102 is to analyze the boiler combustion conditions corresponding to each major group of garbage components and formulate appropriate control strategies accordingly. For each major group, the combustion characteristics in the boiler at different time nodes, such as temperature fluctuations, pressure changes, combustion rate, etc., are collected and analyzed. At the same time, the boiler control methods at different time nodes are also recorded, such as adjustment of fuel supply, control of air flow, etc. By comparing the combustion characteristics and control methods, the correlation and regularity can be found. Furthermore, according to the changes in the combustion characteristics in the boiler, the original boiler control method is corrected and optimized to ensure the stability and efficiency of the combustion process. Finally, the corrected boiler control method is associated with the corresponding major group of garbage components to form a series of corresponding relationships between the major component groups and the boiler control strategies.

[0047] Step S103, visually analyze the garbage mixture map corresponding to each major component group~boiler control strategy, and determine the color distribution characteristics and edge distribution characteristics in the map, and construct a reference characteristic factor portrait based on the color distribution characteristics and edge distribution characteristics, and establish an association relationship between the reference characteristic factor portrait and the corresponding relationship between the major component group~boiler control strategy.

[0048] Step S103 is to further visualize and extract features of the above correspondence. For each pair of major component groups and boiler control strategies, a corresponding garbage mixture image is generated and visually analyzed. Through professional image analysis technology, the color distribution characteristics and edge distribution characteristics in the image are extracted. These characteristics can intuitively reflect the physical state and combustion characteristics of the garbage mixture. Then, based on these characteristics, a reference characteristic factor portrait is constructed. The portrait is a comprehensive data model that can comprehensively describe the visual characteristics of the garbage mixture. Finally, the reference characteristic factor portrait is associated with the corresponding major component group and boiler control strategy to form a complete correspondence library of the mixed component and boiler control strategy judgment model. The implementation of this step provides intuitive and accurate data support and judgment basis for subsequent boiler combustion control.

[0049] Step S200, photographing the garbage mixture to obtain a garbage mixture map, performing visual analysis on the garbage mixture map to determine color distribution features and edge distribution features in the map.

[0050] In this step, the focus is on the state of the garbage mixture before combustion. Through advanced image capture technology, a clear image of the garbage mixture is captured, namely the garbage mixture map. This image provides an intuitive view of the garbage mixture, enabling further analysis of its characteristics. In order to judge the composition of the garbage mixture more carefully, professional visual analysis software is used to deeply analyze the image and extract color distribution characteristics and edge distribution characteristics. The color distribution characteristics can reveal the color composition of different materials in the garbage mixture, while the edge distribution characteristics help to identify the boundaries and distribution between materials. This information is crucial to understanding the composition and structure of the garbage mixture, and they provide more refined data support for the subsequent formulation of boiler control strategies.

[0051] In some embodiments disclosed in the present invention, a method for visually analyzing a garbage mixture map includes:

[0052] Step S201 , preprocessing the garbage mixture image, including denoising and color space conversion, to obtain a corresponding first processed garbage mixture image.

[0053] Step S201 is the preprocessing stage, the main purpose of which is to optimize the image quality of the garbage mixture map for subsequent visual analysis. This step includes two key operations: denoising and color space conversion. Denoising is to eliminate irrelevant information or interference in the image and improve the clarity of the image; while color space conversion is to convert the image from one color representation to another in order to better analyze and process the color information in the image. After preprocessing, the first processed garbage mixture map is obtained, which lays the foundation for subsequent color acquisition and analysis.

[0054] Step S202, randomly setting a number of color collection points for the first processed garbage mixture image, and collecting color parameters corresponding to each color collection point.

[0055] Step S202 is the color collection stage, in which a number of color collection points are randomly set on the first processed garbage mixture map, and the color parameters corresponding to each collection point are accurately collected. These color parameters may include hue, saturation, brightness, etc., which can fully describe the color characteristics at the collection point.

[0056] Step S203, setting a number of color parameter intervals, counting the color parameter intervals to which the color acquisition points correspond, determining the number of first color acquisition points corresponding to each color parameter interval, and calculating the ratio of the number of first color acquisition points to the total number of color acquisition points.

[0057] Step S203 is the color statistics and analysis stage, in which a number of color parameter intervals are pre-set, and these intervals cover all possible color ranges in the garbage mixture map. Then, the color parameters corresponding to the color collection points are judged as intervals, and the number of color collection points belonging to each color parameter interval is counted, that is, the number of first color collection points. Next, the ratio of the number of first color collection points to the total number of color collection points is calculated, and this ratio reflects the relative distribution of the color parameter interval in the garbage mixture map.

[0058] Step S204, identifying the color collection point ratio corresponding to each color parameter interval as a characteristic factor of the color distribution characteristics of the garbage mixture map.

[0059] Step S204 is the characteristic factor extraction stage, in which the ratio of the color collection points corresponding to each color parameter interval is identified as the characteristic factor of the color distribution characteristics of the garbage mixture map. These characteristic factors can intuitively reflect the distribution and characteristics of the colors in the garbage mixture map, and provide important data support for the subsequent boiler control strategy formulation. By extracting and analyzing these characteristic factors, the combustion characteristics and control requirements of the garbage mixture can be more accurately judged, thereby achieving efficient and environmentally friendly combustion of the boiler.

[0060] In some embodiments disclosed in the present invention, the method for visually analyzing a garbage mixture map further includes:

[0061] Step S205 , preprocessing the garbage mixture image, including denoising and graying, to obtain a corresponding second processed garbage mixture image.

[0062] Step S205 is one of the key steps for preprocessing the garbage mixture image. In this step, the image is mainly subjected to denoising and grayscale processing. The purpose of denoising is to eliminate irrelevant information or interference in the image and improve the clarity of the image; while grayscale is to convert the color image into a grayscale image, which can simplify the image data and make subsequent edge detection easier and more accurate. After this preprocessing, the second processed garbage mixture image is obtained, which provides a good foundation for subsequent edge detection and analysis.

[0063] Step S206, using edge detection technology to mark the edges in the second processed garbage mixture map, randomly select an edge parameter analysis block in the second processed garbage mixture map, count the number of edges in the edge parameter analysis block and the edge length of each edge, and calculate the ratio of the edge number to the block area of ​​the edge parameter analysis block, record it as the edge distribution density parameter, and calculate the average edge length of all edges.

[0064] Step S207, identifying the edge distribution density parameter and the average edge length as characteristic factors of the edge distribution characteristics of the garbage mixture map.

[0065] Step S300, based on the color distribution characteristics of the garbage mixture image and the characteristic factors of the edge distribution characteristics, a real-time characteristic factor portrait of the garbage mixture image is formed, and based on the feature conspicuity judgment of the color distribution characteristics, the corresponding characteristic factor branches in the real-time characteristic factor portrait are adjusted.

[0066] Based on the color distribution features and edge distribution features extracted in step S200, a real-time feature factor portrait of the garbage mixture is formed. This portrait is a comprehensive data model that fully reflects the real-time status of the garbage mixture. At the same time, according to the feature prominence judgment of the color distribution features, the corresponding feature factor branches in the real-time feature factor portrait are adjusted to ensure the accuracy and reliability of the portrait. This step provides a solid foundation for the subsequent formulation of boiler control strategies.

[0067] In some embodiments disclosed in the present invention, a method for forming a real-time characteristic factor portrait of a garbage mixture map includes:

[0068] Step S301, setting a characteristic factor portrait template, the characteristic factor portrait template includes a plurality of characteristic factor expression lines, one end of the characteristic factor expression lines overlap each other, and the characteristic factor expression lines are evenly dispersed in a circle, and the characteristic factor expression lines correspond to a color acquisition point ratio of a specific color parameter interval, or a specific edge distribution density parameter, or a specific average edge length.

[0069] Step S302, based on the feature conspicuity judgment of the color distribution feature, determine the first amplification factor of the feature factor expression line corresponding to the color distribution feature, and determine the length of the feature factor expression line in combination with the parameters corresponding to each feature factor.

[0070] In some embodiments disclosed in the present invention, the method for judging the feature prominence of each feature factor includes:

[0071] Step S3021, clustering the color collection point ratios of the color parameter intervals corresponding to all garbage mixture maps. The method of clustering analysis includes: identifying the color parameter intervals with color collection point ratios greater than or equal to 0 as valid color parameter intervals, marking and counting the valid color parameter intervals with color collection point ratios greater than or equal to a preset value, and recording them as the number of valid color parameter intervals.

[0072] Step S3022, counting the effective distinguishing number of mixed components in the garbage mixture in the historical records, the method for determining the effective distinguishing number includes classifying the mixed components, and performing a combustion test on the mixed components of each category, recording the mixed components that match the combustion characteristics and belong to the same category as effective distinguishing components, and counting the total number of components of the effective distinguishing components.

[0073] Step S3023, determining the first amplification factor of the characteristic factor expression line corresponding to the color distribution feature based on the effective color distinction difference between the number of effective color parameter intervals and the total number of components.

[0074] In some embodiments disclosed in the present invention, the expression for determining the first amplification factor is: .

[0075] in, is the first amplification factor, is the amplification factor conversion factor, The difference between the preset effective color distinction standards, is the effective color difference, The adjustment coefficient for the difference in effective color distinction. Adjusts the constant for the amount of difference in effective color distinction.

[0076] Step S400, using the boiler control strategy judgment model, substitutes the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval, determines the matching reference characteristic factor portrait, and based on the determined reference characteristic factor portrait, determines the optimal boiler control strategy, and uses the optimal boiler control strategy to control the boiler.

[0077] This step uses the boiler control strategy judgment model determined in step S100 to substitute the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval. Through comparison and matching, the reference characteristic factor portrait that is most similar to the real-time characteristic factor portrait is found, and the optimal boiler control strategy is determined based on the portrait. Finally, this preferred strategy is applied to the actual control of the boiler, and the operating parameters and combustion conditions of the boiler are adjusted to achieve efficient and environmentally friendly combustion of the garbage mixture. This step realizes the intelligent monitoring and optimization control of boiler combustion.

[0078] In some embodiments disclosed in the present invention, a method for determining a matching reference feature factor portrait includes:

[0079] Step S401, determining the number of characteristic factor expression lines in the real-time characteristic factor portrait, and performing a first screening of reference characteristic factor portraits in the reference characteristic factor portrait library based on the number of characteristic factor expression lines;

[0080] Step S402, determining the average length of the characteristic factor expression line in the real-time characteristic factor portrait, and based on the average length of the characteristic factor expression line, performing a second screening on the reference characteristic factor portrait after the first screening, to obtain a number of reference characteristic factor portraits participating in the comparison;

[0081] Step S403, compare the reference characteristic factor portraits involved in the comparison with the real-time characteristic factor portraits respectively, and calculate the degree of matching between the two. If the degree of matching between the two is greater than or equal to a preset value, the corresponding reference characteristic factor portrait is deemed to match the real-time characteristic factor portrait.

[0082] In some embodiments disclosed in the present invention, the method for calculating the degree of matching between the reference feature factor portrait and the real-time feature factor portrait includes: .

[0083] in, For the matching degree, is the first matching degree weight coefficient, is the second matching degree weight coefficient, is the third matching degree weight coefficient, is the color distribution feature corresponding to The characteristic factor expresses the difference in line length, is the total number of characteristic factor expression lines corresponding to the color distribution characteristics, is the length difference of the characteristic factor expression line corresponding to the edge distribution density parameter, The characteristic factor corresponding to the average edge length expresses the length difference of the line.

[0084] In some embodiments disclosed in the present invention, there is also disclosed a boiler combustion control system based on garbage visual data analysis, comprising:

[0085] The first module is used to determine the mixed components of the pre-treated garbage mixture, and based on the determined mixed components, determine the boiler control strategy judgment model to be applied in the preset mixed component-boiler control strategy judgment model correspondence relationship;

[0086] The second module is used to take an image of the garbage mixture to obtain a garbage mixture map, perform visual analysis on the garbage mixture map, and determine the color distribution characteristics and edge distribution characteristics in the map;

[0087] The third module is used to form a real-time characteristic factor portrait of the garbage mixture image based on the color distribution characteristics and the characteristic factors of the edge distribution characteristics of the garbage mixture image, and adjust the corresponding characteristic factor branches in the real-time characteristic factor portrait based on the characteristic conspicuity judgment of the color distribution characteristics;

[0088] The fourth module is used to use the boiler control strategy judgment model to substitute the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval, determine the matching reference characteristic factor portrait, and determine the optimal boiler control strategy based on the determined reference characteristic factor portrait, and use the optimal boiler control strategy to control the boiler.

[0089] The present invention discloses a boiler combustion control method and system based on garbage visual data analysis, which relates to the technical field of boiler control, and specifically discloses that the mixed components of pre-treated garbage mixture are determined, and a corresponding boiler control strategy judgment model is selected according to the mixed components; secondly, the garbage mixture is imaged, and color distribution characteristics and edge distribution characteristics are extracted through visual analysis; then, a real-time characteristic factor portrait of the garbage mixture is formed based on these characteristic factors, and the characteristic factor branches in the portrait are adjusted according to the color distribution characteristics; finally, the boiler control strategy judgment model is used to compare the real-time characteristic factor portrait with a preset reference characteristic factor portrait library, and the optimal boiler control strategy is determined, and applied to the actual control of the boiler. The above technical scheme realizes intelligent monitoring and optimization control of the garbage incineration process, and improves combustion efficiency.

[0090] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. Boiler combustion control method based on garbage visual data analysis, It is characterized in that include: Determine the mixed components of the pre-treated garbage mixture, and based on the determined mixed components, determine the boiler control strategy judgment model to be applied in the preset mixed component-boiler control strategy judgment model correspondence relationship; Taking an image of the garbage mixture to obtain a garbage mixture map, visually analyzing the garbage mixture map to determine the color distribution characteristics and edge distribution characteristics in the map; Based on the color distribution characteristics of the garbage mixture image and the characteristic factors of the edge distribution characteristics, a real-time characteristic factor image of the garbage mixture image is formed, and based on the characteristic conspicuity judgment of the color distribution characteristics, the corresponding characteristic factor branches in the real-time characteristic factor image are adjusted; The boiler control strategy judgment model is used to substitute the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval, and the matching reference characteristic factor portrait is determined. Based on the determined reference characteristic factor portrait, the optimal boiler control strategy is determined, and the boiler is controlled using the optimal boiler control strategy.

2. The boiler combustion control method based on garbage visual data analysis according to claim 1, It is characterized in that Methods for constructing the corresponding relationship between the mixed component and boiler control strategy judgment model include: Based on the major categories of garbage components appearing in the mixed components of the burned garbage mixture, several major category groups of garbage components are formed, and based on the garbage component subcategories that can be determined in each major category of garbage components, secondary subdivision is performed, and the subdivided garbage component subcategories are used to supplement the data of the major category groups to which they belong; The boiler combustion conditions and boiler control strategies corresponding to each garbage component group are collected and analyzed to determine the combustion characteristics in the boiler at different time nodes and the boiler control methods at different time nodes. Based on the combustion characteristics in the boiler, the boiler control methods at different time nodes are corrected, and the corrected boiler control methods are associated with the garbage component groups to obtain several corresponding relationships between component groups and boiler control strategies. A visual analysis is performed on the garbage mixture map corresponding to each major component group~boiler control strategy, and the color distribution characteristics and edge distribution characteristics in the map are determined. Based on the color distribution characteristics and edge distribution characteristics, a reference characteristic factor portrait is constructed, and a correlation is established between the reference characteristic factor portrait and the corresponding relationship between the major component group~boiler control strategy.

3. The boiler combustion control method based on garbage visual data analysis according to claim 1, It is characterized in that Methods for visual analysis of waste mix maps include: Preprocessing the garbage mixture image, including denoising and color space conversion, to obtain a corresponding first processed garbage mixture image; A plurality of color collection points are randomly set for the first processed garbage mixture image, and the color parameters corresponding to each color collection point are collected; A number of color parameter intervals are set, and the color parameter intervals to which the color acquisition points correspond are counted, the number of first color acquisition points of the color acquisition points corresponding to each color parameter interval is determined, and the ratio of the number of first color acquisition points to the total number of color acquisition points is calculated; The ratio of color collection points corresponding to each color parameter interval is identified as the characteristic factor of the color distribution characteristics of the garbage mixture map.

4. The boiler combustion control method based on garbage visual data analysis according to claim 3, It is characterized in that Methods for visual analysis of waste mixture maps also include: Preprocessing the garbage mixture image, including denoising and graying, to obtain a corresponding second processed garbage mixture image; Using edge detection technology, the edges in the second processed garbage mixture image are marked, edge parameter analysis blocks are randomly selected in the second processed garbage mixture image, the number of edges in the edge parameter analysis block and the edge length of each edge are counted, and the ratio of the edge number to the block area of ​​the edge parameter analysis block is calculated and recorded as the edge distribution density parameter, and the average edge length of all edges is calculated; The edge distribution density parameter and the average edge length are identified as the characteristic factors of the edge distribution characteristics of the garbage mixture map.

5. The boiler combustion control method based on garbage visual data analysis according to claim 1, It is characterized in that The method of forming a real-time characteristic factor profile of the garbage mixture map includes: A characteristic factor portrait template is set, and the characteristic factor portrait template includes a plurality of characteristic factor expression lines, one end of the characteristic factor expression lines overlaps with each other, and the characteristic factor expression lines are evenly dispersed in a circular manner, and the characteristic factor expression lines correspond to a color collection point ratio of a specific color parameter interval, or a specific edge distribution density parameter, or a specific average edge length; Based on the feature conspicuity judgment of the color distribution feature, the first amplification factor of the feature factor expression line corresponding to the color distribution feature is determined, and the length of the feature factor expression line is determined in combination with the parameters corresponding to each feature factor.

6. The boiler combustion control method based on garbage visual data analysis according to claim 5, It is characterized in that The method for judging the characteristic significance of each characteristic factor includes: The color collection point ratios of the color parameter intervals corresponding to all garbage mixed material images are clustered, and the clustering analysis method includes: identifying the color parameter intervals with color collection point ratios greater than or equal to 0 as valid color parameter intervals, marking and counting the valid color parameter intervals with color collection point ratios greater than or equal to a preset value, and recording them as the number of valid color parameter intervals; The effective distinguishable number of mixed components in the garbage mixture in the historical records is counted, and the method for determining the effective distinguishable number includes classifying the mixed components, and performing a combustion test on the mixed components of each category, recording the mixed components with the same combustion characteristics and belonging to the same category as effective distinguishable components, and counting the total number of components of the effective distinguishable components; Based on the effective color distinction difference between the number of effective color parameter intervals and the total number of components, the first amplification factor of the characteristic factor expression line corresponding to the color distribution feature is determined.

7. The boiler combustion control method based on garbage visual data analysis according to claim 6, It is characterized in that The expression for determining the first amplification factor is: ; in, is the first amplification factor, is the amplification factor conversion factor, The difference between the preset effective color distinction standards, is the effective color difference, The adjustment coefficient for the difference in effective color distinction. Adjusts the constant for the amount of difference in effective color distinction.

8. The boiler combustion control method based on garbage visual data analysis according to claim 5, It is characterized in that The method of determining the matching reference feature factor portrait includes: Determining the number of characteristic factor expression lines in the real-time characteristic factor portrait, and performing a first screening of reference characteristic factor portraits in the reference characteristic factor portrait library based on the number of characteristic factor expression lines; The average length of the characteristic factor expression line in the real-time characteristic factor portrait is determined, and based on the average length of the characteristic factor expression line, the reference characteristic factor portrait after the first screening is screened for the second time to obtain several reference characteristic factor portraits participating in the comparison; The reference characteristic factor portraits involved in the comparison are compared with the real-time characteristic factor portraits respectively, and the degree of matching between the two is calculated. If the degree of matching between the two is greater than or equal to the preset value, the corresponding reference characteristic factor portrait is deemed to match the real-time characteristic factor portrait.

9. The boiler combustion control method based on garbage visual data analysis according to claim 8, It is characterized in that The method for calculating the matching degree between the reference feature factor portrait and the real-time feature factor portrait includes: ; in, For the matching degree, is the first matching degree weight coefficient, is the second matching degree weight coefficient, is the third matching degree weight coefficient, is the color distribution feature corresponding to The characteristic factor expresses the difference in line length, is the total number of characteristic factor expression lines corresponding to the color distribution characteristics, is the length difference of the characteristic factor expression line corresponding to the edge distribution density parameter, The characteristic factor corresponding to the average edge length expresses the length difference of the line.

10. Boiler combustion control system based on garbage visual data analysis, It is characterized in that The boiler combustion control method according to claim 1 is applied, and the boiler combustion control system comprises: The first module is used to determine the mixed components of the pre-treated garbage mixture, and based on the determined mixed components, determine the boiler control strategy judgment model to be applied in the preset mixed component-boiler control strategy judgment model correspondence relationship; The second module is used to take an image of the garbage mixture to obtain a garbage mixture map, perform visual analysis on the garbage mixture map, and determine the color distribution characteristics and edge distribution characteristics in the map; The third module is used to form a real-time characteristic factor portrait of the garbage mixture image based on the color distribution characteristics and the characteristic factors of the edge distribution characteristics of the garbage mixture image, and adjust the corresponding characteristic factor branches in the real-time characteristic factor portrait based on the characteristic conspicuity judgment of the color distribution characteristics; The fourth module is used to use the boiler control strategy judgment model to substitute the real-time characteristic factor portrait into the preset reference characteristic factor portrait library for retrieval, determine the matching reference characteristic factor portrait, and determine the optimal boiler control strategy based on the determined reference characteristic factor portrait, and use the optimal boiler control strategy to control the boiler.

Citation Information

Patent Citations

  • Automatic feeding control method and system for waste incineration feeding

    CN112097272A

  • Waste incineration feeding method and system, terminal and storage medium

    CN116447603A

  • Device for estimating supply amount of garbage for garbage incinerator

    JP2005090774A

  • Apparatus for detecting agitated state in refuse pit and method of detecting agitated state in refuse pit

    JP2015143139A

  • Method and system for diagnosing combustion state of garbage incinerator on basis of image processing

    WO2024179215A1