Boiler heating surface slagging condition monitoring method and system based on image calibration

Abstract

The invention provides a boiler heating surface slagging condition monitoring method and system based on image calibration, and the specific steps are as follows: obtaining boiler heating surface image data and heating surface calibration target image data, and respectively extracting pixel data; obtaining a pixel calibration coefficient based on the size of the heating surface calibration target and the obtained pixel data; a boundary line of a wall surface and a slagging layer in the heating surface pixel data is extracted, a slagging contour is determined according to the continuous boundary line, and the pixel thickness, the pixel number and the slagging parting dimension of a slagging part are obtained; based on the pixel calibration coefficient, the pixel thickness, the pixel number and the slagging parting dimension, the slagging quality is obtained, and the slagging condition is judged according to the slagging quality. The method solves the problems that traditional boiler heating surface slagging on-line monitoring has serious defects, only can qualitatively recognize a slagging area, cannot quantify the size and is difficult to meet actual requirements.

Description

Technical Field

[0001] This invention belongs to the field of boiler heating surface slagging monitoring technology, specifically to a method and system for monitoring boiler heating surface slagging based on image calibration. Background Technology

[0002] In the field of boiler heating surface slagging monitoring, traditional monitoring methods are inadequate in terms of real-time performance, often failing to provide timely feedback on slagging conditions, making it difficult for maintenance personnel to keep abreast of slagging dynamics. Furthermore, traditional methods largely remain at the qualitative analysis level, making it difficult to accurately quantify the specific dimensions of slagging, which results in a lack of precise basis for soot blowing decisions based on monitoring results.

[0003] With technological advancements, existing online monitoring systems are gradually being implemented, but they still have significant shortcomings. These systems typically only qualitatively determine slagging areas through contrast analysis, failing to provide quantitative feedback on slagging size. In practice, maintenance personnel can only estimate the slagging extent based on their own experience, a method with a large margin of error that easily leads to deviations in soot blowing operations, resulting in over-cleaning or insufficient slagging, thus affecting the normal operation and efficiency of the boiler.

[0004] Some studies have attempted to introduce image recognition schemes, bringing new ideas to monitoring technology. These schemes introduce the concept of calibration; however, there are unreasonable aspects in the placement of the calibration objects. Most calibration objects are placed outside the boiler observation port or based on existing equipment inside the furnace, resulting in spatial isolation between the calibration objects and the actual location of the heated surfaces. More importantly, the high-temperature flue gas inside the furnace can cause complex situations such as slag expansion, and these image recognition schemes completely ignore this crucial variable, thus rendering the pixel calibration coefficients ineffective and unable to accurately reflect the true situation of slag formation.

[0005] The lack of real-time capability is a prominent problem shared by traditional methods and existing online monitoring systems. The inability to obtain slagging information in a timely manner puts maintenance work in a passive state, hindering rapid responses to real-time changes in slagging and increasing the risk to boiler operation. The lack of quantitative capability severely restricts the practicality of monitoring technology. Neither traditional methods nor existing online monitoring systems can accurately quantify slagging size, resulting in a lack of scientific basis for soot blowing decisions, relying solely on the experience of maintenance personnel and increasing the uncertainty of decision-making. Calibration error problems are particularly evident in some image recognition solutions. Inappropriate setting of calibration object positions and the neglect of key variables prevent pixel calibration coefficients from accurately establishing the relationship between slagging size and monitoring data, significantly reducing the accuracy of monitoring results and failing to meet actual maintenance needs. The current reliance on experience-based estimations in maintenance makes it difficult to achieve ideal soot blowing results. Over-cleaning increases maintenance costs and equipment wear, while insufficient slagging affects boiler thermal efficiency and safety, posing a double risk to the stable operation of the boiler.

[0006] To address the lack of real-time and quantitative deficiencies in boiler heating surface slagging monitoring, it is necessary to overcome the bottleneck that traditional online monitoring can only qualitatively identify slagging areas but cannot quantify their dimensions. This requires considering spatial isolation errors caused by the calibration object being placed outside the boiler observation hole or as a non-in-situ reference object, as well as pixel calibration coefficient failure caused by high-temperature flue gas distortion and flame interference inside the furnace. The goal is to resolve the dilemma of blindly estimating soot blowing based on experience during operation and maintenance, and to provide a dynamic slagging dimension chain with millimeter-level accuracy for soot blowing decisions, thereby avoiding the dual risks of insufficient or excessive slagging. Summary of the Invention

[0007] This invention provides a method and system for monitoring slagging on boiler heating surfaces based on image calibration, which solves the serious defects of traditional online monitoring of slagging on boiler heating surfaces, which can only qualitatively identify slagging areas but cannot quantify their size, making it difficult to meet practical needs.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for monitoring slagging on boiler heating surfaces based on image calibration, the specific steps of which are as follows: Acquire image data of the boiler heating surface and image data of the heating surface calibration target, and extract pixel data respectively; The pixel calibration coefficients are obtained based on the target size of the heated surface and the acquired pixel data; Extract the boundary line between the wall and the slag layer from the pixel data of the heated surface, determine the slag outline based on the continuous boundary line, and obtain the pixel thickness, number of pixels and slag type dimension of the slag part. Slagging quality is obtained based on pixel calibration coefficient, pixel thickness, number of pixels, and slagging type dimension, and the slagging condition is judged based on the slagging quality.

[0009] Furthermore, industrial endoscopic imaging equipment was used to acquire image data of the boiler heating surface in areas prone to slagging, as well as image data of the heating surface calibration target.

[0010] Furthermore, the heating surface calibration target adopts a high-temperature resistant ceramic calibration target. Multiple concentric circles are etched on the surface of the calibration target, and a crosshair reference line is etched. The heating surface calibration target is fixed on the surface of the boiler heating surface. The calibration target is within ±5° of the imaging center line in the field of view. The center of the crosshair reference line is located on the center line of the industrial endoscopic imaging equipment. The pixel calibration coefficient is obtained based on the actual radius of the concentric circles and the acquired pixel data.

[0011] Furthermore, the pixel data of the heated surface is filtered, denoised, and the image edges are enhanced; the gradient value of the pixel color value change is calculated, and the points with large changes in RGB values ​​between two adjacent pixels are selected. The pixels with larger values ​​are taken as the edge points of the pipe wall and slag layer contour. The edge points are processed to obtain the boundary line between the wall and the slag layer in the image. The pixel thickness, pixel width, and pixel area of ​​the slag contour are statistically analyzed.

[0012] Furthermore, the slag classification dimension is determined based on surface roughness, which is represented by the average gradient of color values ​​along the boundary line, as follows:

[0013] In the formula, I represents surface roughness, C represents the normalized color value on the boundary line; n represents the number of points on the boundary line; i represents the starting point of the boundary line, and j represents the ending point of the boundary line. and These represent the gradients of color value C along the x and y directions, respectively; When the surface roughness is greater than the set threshold, the slag layer is newly deposited slag, and the slag classification dimension is set to 2.3. When the surface roughness is less than or equal to the set threshold, the slag layer is molten slag, and the slag classification dimension is set to 2.7.

[0014] Furthermore, the slag thickness is obtained using pixel thickness and pixel calibration coefficient, as shown below:

[0015] In the formula, d / px is defined as the pixel calibration coefficient K, where d is the diameter of the heated surface calibration target image, px is the pixel value, and P... L The pixel thickness of the slag outline; The slag area is obtained based on the pixel calibration coefficient and the number of pixels, as shown below:

[0016] In the formula, K is the pixel calibration coefficient, and n is the number of pixels with slag. Slagging quality is derived from slagging thickness, slagging area, slagging type dimension, and density, and is expressed as follows:

[0017] In the formula, ρ is density, and F d L represents the slagging type classification dimension, where L is the slagging thickness and S is the slagging area.

[0018] The present invention also provides an image-calibrated monitoring system for slagging on boiler heating surfaces, comprising: The pixel acquisition module is used to acquire boiler heating surface image data and heating surface calibration target image data, and extract pixel data respectively. The pixel calibration module is used to obtain pixel calibration coefficients based on the target size of the heated surface and the acquired pixel data. The boundary line extraction module is used to extract the boundary lines between the wall and the slag layer in the pixel data of the heated surface, determine the slag outline based on the continuous boundary lines, and obtain the pixel thickness, number of pixels and slag type dimension of the slag part. The slagging monitoring module is used to obtain slagging quality based on pixel calibration coefficient, pixel thickness, number of pixels, and slagging type dimension, and to judge the slagging condition using the slagging quality.

[0019] The present invention also provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the above-described method for monitoring slagging on boiler heating surfaces based on image calibration.

[0020] The present invention also provides a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the above-described method for monitoring slagging on boiler heating surfaces based on image calibration.

[0021] The present invention also provides a computer program product, including a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the above-described method for monitoring slagging on boiler heating surfaces based on image calibration.

[0022] Compared with the prior art, the present invention has at least the following beneficial effects: This invention provides a method for monitoring slagging on boiler heating surfaces based on image calibration. Pixel calibration coefficients are obtained based on the target size and pixel data of the heating surface calibration, fully considering the size correspondence in the actual monitoring environment. This effectively solves the errors caused by calibration problems in traditional methods, making subsequent calculations more reliable. The boundary line between the wall surface and the slagging layer is extracted and the slagging contour is determined, thereby obtaining the pixel thickness, number of pixels, and slagging classification dimension of the slagging portion, which can comprehensively and meticulously describe the characteristics of the slagging. Finally, based on these parameters, the slagging quality is obtained and used to judge the slagging condition, realizing quantitative monitoring of slagging on boiler heating surfaces. This overcomes the limitations of traditional qualitative identification and provides a more accurate monitoring means for the safe and stable operation of boilers.

[0023] Furthermore, this invention conducts dimensional calibration experiments at actual monitoring locations on-site, using a high-temperature resistant ceramic calibration target to establish the correspondence between pixel values ​​and actual dimensions. This method fully considers the complex internal environment of the boiler, avoiding spatial isolation errors caused by the calibration object being placed outside the boiler observation hole or a non-in-situ reference object, ensuring the accuracy and reliability of the calibration results, and providing a solid foundation for the subsequent accurate calculation of slag size.

[0024] Furthermore, this invention performs refined feature value extraction and error assessment on the calibration target imaging image to obtain corrected pixel calibration coefficients. This step effectively solves the problem of pixel calibration coefficient failure caused by high-temperature flue gas distortion and flame interference inside the furnace, enabling the pixel calibration coefficients to better adapt to the harsh environment inside the boiler and ensuring the accuracy of subsequent calculations.

[0025] Furthermore, in the slag boundary pixel extraction stage, this invention uses slag edge recognition technology to filter the acquired heated surface image data, removing noise information based on texture features and enhancing image edges. By calculating the gradient values ​​of pixel color changes, the edge points of the tube wall and slag layer contour are accurately selected, thereby determining the slag contour and statistically analyzing relevant pixel parameters. This process effectively overcomes the interference of the complex furnace environment on image recognition, ensuring the accuracy of slag boundary recognition and providing a guarantee for the accurate calculation of subsequent slag parameters.

[0026] Furthermore, in the multi-dimensional slagging parameter inversion, this invention accurately calculates the slagging thickness, area, and mass based on the corrected pixel calibration coefficients. In particular, by calculating the fractal dimension through surface roughness and combining it with density test data, the nature of the slagging can be accurately determined, providing scientific and comprehensive data support for soot blowing decisions. This avoids the problem of blindly estimating soot blowing based on experience during operation and maintenance, and improves the efficiency and safety of boiler operation.

[0027] Furthermore, this invention ultimately monitors the slagging condition of boiler heating surfaces based on slagging quality and provides clear criteria for judging the degree of slagging. By providing a dynamic slagging dimension chain with millimeter-level accuracy, maintenance personnel can clearly understand the actual slagging situation and thus formulate reasonable soot blowing strategies. This avoids problems such as decreased heating surface thermal efficiency and increased safety hazards caused by insufficient slagging, while also avoiding resource waste and equipment damage caused by excessive cleaning, resulting in good economic and social benefits. Taking the monitoring of the high-temperature heating surface of a 660MW coal-fired boiler in the embodiment as an example, the monitoring method of this invention accurately determines that the slagging condition is moderate, providing strong support for the safe, stable, and efficient operation of the boiler. Attached Figure Description

[0028] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0029] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0030] To address the quantitative monitoring needs of slagging on boiler heating surfaces, the core of this invention lies in constructing a technical closed loop of in-situ dynamic calibration, anti-interference identification and correction, and accurate inversion, providing a method for monitoring slagging on boiler heating surfaces based on image calibration, specifically including: (1) Size calibration experiment: A size calibration experiment was carried out at the actual monitoring location on site to obtain the image pixel values ​​of the calibration target inside the furnace, and to establish the correspondence between the pixel values ​​and the actual size. Specifically: S1: Prepare a high-temperature resistant ceramic calibration target, etch multiple concentric circles and crosshairs to obtain the diameter data of each concentric circle of the calibration target; S2: Install industrial endoscopic imaging equipment in the slag-prone area of ​​the boiler and keep it unchanged during the monitoring process. When the boiler is shut down, fix the calibration target on the surface of the heated surface. The calibration target is within ±5° of the imaging center line in the field of view. Prepare a light source in the furnace to make the concentric circle imaging on the calibration target clear and stable. S3: Analyze and process the imaging image of the calibration target, obtain the pixel value of each concentric circle and evaluate the error, establish a quantitative model of the diameter d of the concentric circle and the pixel value px, and obtain the corrected pixel calibration coefficient.

[0031] Preferably, feature values ​​are extracted from the imaging image of the calibration target, focusing on capturing the traces of multiple concentric circles and crosshairs etched on the target surface. Through repeated experiments, the x and y coordinates corresponding to the pixel values ​​are extracted, and the average value is used to compare the drift of the x and y coordinates to evaluate the pixel value error. A quantitative model of the concentric circle diameter d and the pixel value px is established to obtain the corrected pixel calibration coefficient.

[0032] (2) Slag Boundary Pixel Extraction: Based on slag edge recognition technology, the edge contours in the slag image are extracted to obtain the pixel values ​​occupied by the contours of the slag on the wall. Specifically: S1: Collect image data of each frame of the heated surface, extract image pixel information, filter the image information, filter out noise information based on texture features, and enhance image edges; S2: Calculate the gradient value of the color value change of the pixel, filter the points where the RGB values ​​of two adjacent pixels change significantly (the linear weighted change of the three R, G, and B values ​​is greater than 50), and take the pixel with the larger value as the edge point of the pipe wall and slag layer outline. Based on this, the boundary line of the wall and slag layer in the image is obtained.

[0033] S3: Filter the acquired image boundary lines, determine the slag contour based on continuous boundary lines, and calculate the pixel thickness P of the slag contour. L Number of pixels with slag inclusions. Pixel width P. W And the pixel area S.

[0034] (3) Multi-dimensional slag parameter inversion: Based on the corrected pixel calibration coefficients, the slag thickness, area, and mass are calculated, specifically: S1: Based on the corrected pixel calibration coefficients, calculate the slag thickness L using the following formula: (1) In the formula, d / px is defined as the pixel calibration coefficient K, where d is the diameter of the concentric circles and px is the pixel value.

[0035] S2: Calculate the slag-forming area S according to the following formula: (2) In the formula, n is the number of pixels with slag.

[0036] S3: Calculate the slag mass m according to the following formula: (3) In the formula, ρ is density, and F d This refers to the slag formation classification dimension. Density can be obtained through sampling and testing, while the classification dimension can be calculated from surface roughness. When the surface roughness is high, it indicates that the slag is a loose and porous new sediment, and the fractal dimension is 2.3; when the surface roughness is low, it indicates that the slag is a dense molten slag, and the fractal dimension is 2.7.

[0037] The formula for calculating surface roughness I is: (4) In the formula, C is the normalized color value on the boundary line; n is the total number of points involved in the calculation (or the number of points on the boundary line); j represents the range from the starting point to the ending point of the boundary line. and These represent the rate of change (i.e., gradient) of the color value C in the x and y directions, respectively. The physical meaning of formula (4) is the normalized average gradient of the color value from the starting point i of the boundary line to the ending point j of the boundary line. Since the change of the image color value can reflect the surface condition of the layer, I can be used to quantitatively represent the surface roughness. When I is greater than the set threshold, the slag layer is newly deposited slag, and otherwise it is molten slag.

[0038] Preferably, the threshold can be 0.5. When I is greater than 0.5, the slag layer is newly deposited slag, and otherwise it is molten slag.

[0039] (4) Monitor the slagging status of the boiler heating surface based on the slagging quality.

[0040] The boiler heating surface slagging monitoring method proposed in this invention can not only accurately measure the size and quality of slagging, but also assess its classification characteristics, achieving comprehensive monitoring of the slagging state of the boiler heating surface. The use of a calibration target made of high-temperature resistant ceramic material ensures the accuracy and reliability of image data even in extreme environments. Furthermore, image preprocessing and advanced image processing algorithms improve the accuracy and efficiency of slagging detection, helping to prevent and control slagging problems, thereby significantly improving the safety and economy of boiler operation. The application of this method has significant practical implications for optimizing boiler operating conditions, reducing maintenance costs, and extending equipment lifespan.

[0041] Example 1 like Figure 1 As shown in the figure, this embodiment provides a method for monitoring slagging on boiler heating surfaces based on image calibration. Specifically, it includes: first, acquiring image data of the boiler heating surface using an industrial endoscope; then, performing dimensional calibration using a high-temperature resistant ceramic calibration target and calculating pixel calibration coefficients; next, preprocessing the image data, including filtering, noise reduction, and edge enhancement; subsequently, identifying the boundary between the wall surface and the slagging layer through pixel color gradient changes, and calculating the image thickness and number of pixels of the slagging; simultaneously, determining the slagging layer type through surface roughness calculation; finally, calculating the actual quality of the slagging by combining the pixel calibration coefficients, pixel features of the slagging, type dimension, and density. The entire process is closely integrated with the usage process, achieving real-time monitoring and quantitative evaluation of slagging on boiler heating surfaces through the combination of image processing technology and physical parameter calculation. In practical applications, this technical solution can provide boiler maintenance personnel with accurate slagging monitoring data, helping them to take timely measures such as soot blowing, thereby improving the efficiency and safety of boiler operation.

[0042] This embodiment takes the monitoring of the high-temperature heating surface of a 660MW coal-fired boiler as an example, and is divided into the following steps: implementation of size calibration experiment, development of slagging boundary identification algorithm, and development of multi-parameter slagging inversion algorithm.

[0043] Size calibration experiment implementation: A high-temperature resistant magnesium oxide calibration target was selected. Three sets of concentric circles with radii of 50mm, 100mm, and 150mm were etched on its surface, along with a crosshair reference line with a line width of 0.5mm. During boiler shutdown and maintenance, the calibration target was fixed to the high-temperature reheater, ensuring the center of the crosshair reference line was aligned with the center line of the monitoring detector, with a deviation not exceeding 3 degrees. A light source was provided at the calibration location inside the furnace. A visible light imaging industrial endoscope with a resolution of 1920*1200, high temperature resistance, and a frame rate of 30fps was fixed at the furnace opening. Clear and stable images of the calibration target were acquired. For example, the image pixel value corresponding to a concentric circle diameter of 100mm was measured to be 480px, and the pixel calibration coefficient was calculated. K = d / px = 100 / 480 = 0.2083 mm / px.

[0044] Development of a slagging boundary recognition algorithm: Under stable boiler load, images of the high-temperature reheater heating surface are continuously acquired using an industrial endoscopic imaging device. After filtering and enhancement, the RGB color values ​​of each pixel are extracted and normalized to grayscale values ​​(C=0.3R+0.6G+0.1B). The gradient vector of the grayscale value change of each pixel compared to its neighboring pixels is calculated, where the calculation formula in the x-direction is as follows: (5) y direction: (6) The same applies to gradient changes in other directions, when the following condition is met: Points selected as valid edge points are identified. Non-maximum suppression is applied to these valid edge points, retaining local maxima along the gradient direction to form an initial boundary line set. Geometric constraint filtering is applied to the initial boundary lines, using Hough transform to detect pipe array baselines with angles between 80° and 90° to the horizontal direction, excluding edge points within 5 pixels of the baseline. The slag contour is determined based on continuous boundary lines, and the pixel thickness PL of the slag contour is calculated to be 62 pixels, the pixel width PW to be 2497 pixels, and the pixel area S.

[0045] Development of a multi-dimensional slagging parameter inversion algorithm: Based on the results of step (2), the slagging layer thickness and slagging area were inverted: L=12.91mm, S=108.34mm 2 Surface roughness is calculated using the following formula: (6) The calculated surface roughness was 0.48. The threshold was determined by sampling the slag layer and using the same method under ordinary light in the laboratory, with a result of 0.4. Therefore, the slag sample was identified as loose, porous, newly deposited slag, with a classification dimension of 2.3. Using a density analyzer, the density of the loose slag was calculated to be 950 kg / m³. 3 The density of the dense slag is 2050 kg / m³. 3 .

[0046] according to The mass of the slag was calculated to be 1.329g.

[0047] According to the following criteria: if the value of m is greater than 1.5g, the slagging is heavy; if the value of m is less than 1.5g but greater than 0.5g, the slagging is moderate; if m is less than 0.5g, the slagging is light.

[0048] It can be seen that the current slagging situation is moderate.

[0049] The following are embodiments of the apparatus of the present invention, which can be used to execute embodiments of the method of the present invention. For details not disclosed in the apparatus embodiments, please refer to the embodiments of the method of the present invention.

[0050] In another embodiment of the present invention, a monitoring system for slagging on boiler heating surfaces based on image calibration is also provided, comprising: The pixel acquisition module is used to acquire boiler heating surface image data and heating surface calibration target image data, and extract pixel data respectively. The pixel calibration module is used to obtain pixel calibration coefficients based on the target size of the heated surface and the acquired pixel data. The boundary line extraction module is used to extract the boundary lines between the wall and the slag layer in the pixel data of the heated surface, determine the slag outline based on the continuous boundary lines, and obtain the pixel thickness, number of pixels and slag type dimension of the slag part. The slagging monitoring module is used to obtain slagging quality based on pixel calibration coefficient, pixel thickness, number of pixels, and slagging type dimension, and to judge the slagging condition using the slagging quality.

[0051] In another embodiment of the present invention, a terminal device is also provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can implement the operation of a method for monitoring slagging on boiler heating surfaces based on image calibration.

[0052] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and extended storage media supported by the terminal device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the image-calibrated boiler heating surface slagging monitoring method described in the above embodiments.

[0053] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0054] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0055] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for monitoring slagging on boiler heating surfaces based on image calibration, characterized in that, The specific steps are as follows: Acquire image data of the boiler heating surface and image data of the heating surface calibration target, and extract pixel data respectively; The pixel calibration coefficients are obtained based on the target size of the heated surface and the acquired pixel data; Extract the boundary line between the wall and the slag layer from the pixel data of the heated surface, determine the slag outline based on the continuous boundary line, and obtain the pixel thickness, number of pixels and slag type dimension of the slag part. Slagging quality is obtained based on pixel calibration coefficient, pixel thickness, number of pixels, and slagging type dimension, and the slagging condition is judged based on the slagging quality.

2. The method for monitoring slagging on boiler heating surfaces based on image calibration according to claim 1, characterized in that, Industrial endoscopic imaging equipment was used to acquire image data of the boiler heating surface in areas prone to slagging, as well as image data of the heating surface calibration target.

3. The method for monitoring slagging on boiler heating surfaces based on image calibration according to claim 1, characterized in that, The calibration target for the heated surface is made of high-temperature resistant ceramic. Multiple concentric circles are etched on the surface of the calibration target, and a crosshair reference line is also etched. The calibration target is fixed on the surface of the heated surface of the boiler. The calibration target is within ±5° of the imaging center line in the field of view. The center of the crosshair reference line is located on the center line of the industrial endoscopic imaging equipment. The pixel calibration coefficient is obtained based on the actual radius of the concentric circles and the acquired pixel data.

4. The method for monitoring slagging on boiler heating surfaces based on image calibration according to claim 1, characterized in that, The pixel data of the heated surface is filtered, denoised, and the image edges are enhanced. Calculate the gradient value of the color value change of each pixel, filter the pixels with large RGB value changes between two adjacent pixels, and take the pixels with larger values ​​as the edge points of the pipe wall and slag layer contour. Perform continuous processing on the edge points to obtain the boundary line of the wall and slag layer in the image, and count the pixel thickness, pixel width and pixel area of ​​the slag contour.

5. The method for monitoring slagging on boiler heating surfaces based on image calibration according to claim 4, characterized in that, The slag formation classification dimension is determined based on surface roughness, which is represented by the average gradient of color values ​​along the boundary line, as follows: In the formula, I is the surface roughness, C is the normalized color value on the boundary line, and n is the number of points on the boundary line. i is the starting point of the boundary line, and j is the ending point of the boundary line. and These represent the gradients of color value C along the x and y directions, respectively; When the surface roughness is greater than the set threshold, the slag layer is newly deposited slag, and the slag classification dimension is set to 2.

3. When the surface roughness is less than or equal to the set threshold, the slag layer is molten slag, and the slag classification dimension is set to 2.

7.

6. The method for monitoring slagging on boiler heating surfaces based on image calibration according to claim 4, characterized in that, The slag thickness is obtained using pixel thickness and pixel calibration coefficient, as shown below: In the formula, d / px is defined as the pixel calibration coefficient K, where d is the diameter of the heated surface calibration target image, px is the pixel value, and P... L The pixel thickness of the slag outline; The slag area is obtained based on the pixel calibration coefficient and the number of pixels, as shown below: In the formula, K is the pixel calibration coefficient, and n is the number of pixels with slag. Slagging quality is derived from slagging thickness, slagging area, slagging type dimension, and density, and is expressed as follows: In the formula, ρ is density, and F d L represents the slagging type classification dimension, where L is the slagging thickness and S is the slagging area.

7. The present invention also provides a monitoring system for slagging on boiler heating surfaces based on image calibration, characterized in that, include: The pixel acquisition module is used to acquire boiler heating surface image data and heating surface calibration target image data, and extract pixel data respectively. The pixel calibration module is used to obtain pixel calibration coefficients based on the target size of the heated surface and the acquired pixel data. The boundary line extraction module is used to extract the boundary lines between the wall and the slag layer in the pixel data of the heated surface, determine the slag outline based on the continuous boundary lines, and obtain the pixel thickness, number of pixels and slag type dimension of the slag part. The slagging monitoring module is used to obtain slagging quality based on pixel calibration coefficient, pixel thickness, number of pixels, and slagging type dimension, and to judge the slagging condition using the slagging quality.

8. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the image calibration-based method for monitoring slagging on boiler heating surfaces as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the image calibration-based method for monitoring slagging on boiler heating surfaces as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the image calibration-based method for monitoring slagging on boiler heating surfaces as described in any one of claims 1 to 6.

Images