Method for cleaning accumulated coke on inner wall of boiler based on roller robot

Abstract

The invention relates to the technical field of boiler inner wall accumulated coke cleaning, in particular to a boiler inner wall accumulated coke cleaning method based on a roller robot. The method comprises the following steps: primarily cleaning a coke accumulation area by using a roller robot; judging whether a subsequent shot image is interfered or not according to the smoke, and if so, acquiring a thermal imaging image sequence of a focus accumulation area; the temperature value stability and the temperature value change consistency of the pixel points at the same position are obtained respectively, the stability degree of the pixel points at the position is obtained, and then stable pixel points and unstable pixel points are obtained; obtaining smoke concentration relative values of the unstable pixel points, obtaining a weighted stability degree of each smoke concentration relative value, obtaining a fitting curve, further obtaining a correction stability degree of the unstable pixel points and the stable pixel points, and finally obtaining a correction image; and evaluating the cleaning condition of the focus accumulation area based on the corrected image until the focus accumulation cleaning is completed. According to the invention, the accumulated coke cleaning efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler inner wall coke cleaning, and particularly relates to a method for cleaning boiler inner wall coke based on a drum robot. Background Art

[0002] The coke on the inner wall of the boiler is a solid deposit formed by the attachment, carbonization and gradual accumulation of combustible substances such as fuel residues on the inner wall of the boiler due to high temperature during the operation of the boiler. The coke has a large thermal resistance, which will affect the heat conduction efficiency of the boiler, resulting in local overheating, increased fuel consumption and emissions, and has certain potential safety hazards. Therefore, it is necessary to clean the coke on the inner wall of the boiler in time to maintain the normal operation of the boiler. Traditional coke cleaning work often relies on manual labor. However, due to the high temperature and enclosed space inside the boiler, manual coke cleaning is highly dangerous, and the manual coke cleaning speed is slow and the work efficiency is low. With the development of technology, robots are gradually replacing manual labor for coke cleaning work.

[0003] Because the coke has poor heat conductivity, the temperature at the coke position is significantly higher than that at the clean inner wall position. Therefore, robots often use thermal imaging images to identify the coke position on the inner wall of the boiler. During the coke cleaning process of the drum robot, a large amount of flue gas is often generated. The flue gas has properties such as absorption, scattering and occlusion of thermal infrared waves, resulting in blurred and inaccurate thermal imaging images generated, which affects the evaluation of the coke state. And in the positions with poor ventilation in the boiler, the flue gas usually lasts for a long time, affecting the efficiency of the robot's coke cleaning work. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a method for cleaning boiler inner wall coke based on a drum robot, and the specific technical solution adopted is as follows: An embodiment of the present invention provides a method for cleaning boiler inner wall coke based on a drum robot, and the method includes: Using the drum robot to take an initial thermal imaging image of the boiler, and determining the coke region according to the initial thermal imaging image; using the drum robot to conduct a preliminary cleaning of the coke region; Judging whether the subsequent captured images are interfered based on the water vapor concentration, carbon dioxide concentration and soot particle concentration in the boiler after the preliminary cleaning; if the subsequent captured images are interfered, continuously capture thermal imaging images for a set period of time in a coke region to obtain a thermal imaging image sequence; Based on the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence, respectively obtain the temperature value stability and temperature value change consistency of the pixel points at this position, and obtain the stability degree of the pixel points at this position; obtain stable pixel points and unstable pixel points according to the stability degree of the pixel points at each position; Obtain the relative value of the flue gas concentration of the unstable pixel point according to the distance between the unstable pixel point and each pixel point included in the in-focus area; perform weighted averaging on the stability degrees of each unstable pixel point under a certain relative value of the flue gas concentration to obtain the weighted stability degree of this relative value of the flue gas concentration, and obtain a fitting curve; Obtain the corrected stability degree of each unstable pixel point according to the fitting curve; obtain the corrected stability degree of the stable pixel points, and correct the temperature values of each unstable pixel point in the last image of the thermal imaging image sequence to obtain a corrected image; evaluate the cleaning condition of the in-focus area based on the corrected image.

[0005] Preferably, judge whether the subsequent captured images are interfered based on the water vapor concentration, carbon dioxide concentration, and soot particle concentration in the boiler after preliminary cleaning, including: Normalize and average the difference in water vapor concentration in the boiler before and after preliminary cleaning, the difference in carbon dioxide concentration in the boiler before and after preliminary cleaning, and the difference in soot particle concentration in the boiler before and after preliminary cleaning respectively to obtain the possibility of interference of the thermal imaging image; if the possibility of interference of the thermal imaging image is greater than or equal to the first threshold, the subsequent captured images are interfered.

[0006] Preferably, obtain the temperature value stability and temperature value change consistency of the pixel points at the same position on each image in the thermal imaging image sequence, including: Add the standard deviation of the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence to the first preset value and take the reciprocal to obtain the temperature value stability of the pixel points at this position; arrange the temperature values of the pixel points at the same position in chronological order to obtain a chronological temperature value sequence; calculate the difference between the latter temperature value and the former temperature value in each two adjacent temperature values in the chronological temperature value sequence, and sort the obtained differences in chronological order to obtain a difference sequence; obtain the temperature value change consistency of the pixel points at this position according to the difference sequence.

[0007] Preferably, obtain the temperature value change consistency of the pixel points at this position according to the difference sequence, including: Obtain the absolute value of the difference between the latter difference and the former difference in each two adjacent differences in the difference sequence, calculate the reciprocal of the average value of all absolute differences, and then normalize to obtain the temperature value change consistency.

[0008] Preferably, the method for obtaining the stability degree is specifically: Obtain the minimum and maximum values of the standard deviation of the temperature values of each pixel at each position on each image in the thermal imaging image sequence, and denote them as the minimum standard deviation and the maximum standard deviation respectively; obtain the weights corresponding to the temperature value stability and the temperature value change consistency based on the minimum standard deviation and the maximum standard deviation respectively; perform weighted summation on the temperature value stability and the temperature value change consistency of the pixel at a position based on the weights corresponding to the temperature value stability and the temperature value change consistency to obtain the stability degree of the pixel at the position.

[0009] Preferably, obtaining the weights corresponding to the temperature value stability and the temperature value change consistency based on the minimum standard deviation and the maximum standard deviation respectively includes: The weight corresponding to the temperature value change consistency is the ratio of the minimum standard deviation to the maximum standard deviation; the weight corresponding to the temperature value stability is the difference between the first preset value and the weight corresponding to the temperature value change consistency.

[0010] Preferably, obtaining stable pixels and unstable pixels according to the stability degree of the pixel at each position includes: Specifically, if the stability degree of the pixel at a position on each image in the thermal imaging image sequence is less than the second threshold, then all the pixels corresponding to that position are unstable pixels; if the stability degree is greater than or equal to the second threshold, then all the pixels corresponding to that position are stable pixels.

[0011] Preferably, obtaining the relative value of the flue gas concentration of an unstable pixel according to the distances between the unstable pixel and the pixels included in the in-focus area includes: Based on the coordinates of an unstable pixel in the image and the coordinates of each pixel in the in-focus area within the in-focus area, use the Euclidean distance calculation method to obtain the distances between the unstable pixel and each pixel in the in-focus area respectively, and sum them to obtain the sum of the distances; take the reciprocal of the sum of the distances and normalize it to obtain the relative flue gas concentration value of the unstable pixel.

[0012] Preferably, performing weighted averaging on the stability degrees of the unstable pixels under a certain relative value of the flue gas concentration to obtain the weighted stability degree of the certain relative value of the flue gas concentration includes: Respectively obtain the absolute value of the difference between the stability degree of an unstable pixel under a certain relative flue gas concentration value and the stability degrees of other unstable pixels under the same relative flue gas concentration value and sum them to obtain the summation result, take the reciprocal of the summation result and normalize it to obtain the possibility of being affected by the flue gas of the stability degree of the unstable pixel under the certain relative flue gas concentration value; The possibility of a non - stable pixel point being affected by flue gas at a relative flue gas concentration value is compared with the sum of the possibilities of all non - stable pixel points being affected by flue gas at the same relative flue gas concentration value to obtain the weight of the stability degree of the non - stable pixel point at the relative flue gas concentration value; the weighted average of the stability degrees of all non - stable pixel points at the relative flue gas concentration value is calculated using the weights of the stability degrees of the non - stable pixel points at the relative flue gas concentration value to obtain the weighted stability degree of the relative flue gas concentration value.

[0013] Preferably, obtaining the fitting curve includes: Fitting the weighted stability degree of each relative flue gas concentration value and each relative flue gas concentration value to obtain a fitting curve; the abscissa of the fitting curve is each relative flue gas concentration value, and the ordinate is the weighted stability degree of each relative flue gas concentration value.

[0014] Preferably, obtaining the corrected stability degree of each non - stable pixel point according to the fitting curve includes: Substituting the relative flue gas concentration value of the non - stable pixel point into the fitting curve to obtain the corrected stability degree of the non - stable pixel point.

[0015] Preferably, correcting the temperature value of each non - stable pixel point in the last image of the thermal imaging image sequence to obtain a corrected image includes: Obtaining the neighborhood pixel points of a non - stable pixel point in the last image; taking the ratio of the corrected stability degree of a neighborhood pixel point to the sum of the corrected stability degrees of all neighborhood pixel points of the non - stable pixel point as the weight corresponding to the neighborhood pixel point; using the weights corresponding to all neighborhood pixel points of the non - stable pixel point to perform weighted summation on the temperature values of the neighborhood pixel points to obtain the corrected value of the temperature value of the non - stable pixel point; obtaining the corrected values of the temperature values of all non - stable pixel points in the last image to obtain the corrected image.

[0016] Preferably, obtaining the corrected stability degree of the stable pixel points includes: Setting the first preset value as the corrected stability degree of the stable pixel points.

[0017] The embodiments of the present invention have at least the following beneficial effects: The present application first determines the coking area in the boiler through the initial thermal imaging image, and conducts preliminary cleaning on the coking area. Then, according to the water vapor concentration, carbon dioxide concentration, and soot particle concentration in the boiler after preliminary cleaning, it is judged whether the subsequent captured images will be interfered when analyzing the coking situation. If interfered, continuous shooting of a coking area is performed within a set time period to obtain a thermal imaging image sequence for subsequent analysis, so as to obtain a more accurate coking evaluation result; Furthermore, analyze the stability of the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence to obtain the stability degree of the pixel points at each position, thereby identifying stable pixel points and unstable pixel points. Then, obtain the relative smoke concentration value of the unstable pixel points and the weighted stability degree of each relative smoke concentration value, and obtain a fitting curve. Furthermore, based on the fitting curve, obtain the corrected stability degree of the unstable pixel points to achieve the purpose of correcting the stability degree of the unstable pixel points and improve the accuracy of the analysis. Furthermore, correct the temperature values of the unstable pixel points in the last image to obtain a corrected image, which can effectively avoid the problem of over-cleaning or incomplete cleaning caused by the interference of the smoke generated during the coke cleaning process on the judgment of the coke accumulation situation on the inner wall of the boiler, and improve the effectiveness of the robot coke cleaning work. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a method flow chart of a method for cleaning coke accumulation on the inner wall of a boiler based on a drum robot provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific implementation manner, structure, features, and effects of a method for cleaning coke accumulation on the inner wall of a boiler based on a drum robot proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0022] The following will specifically describe the specific solution of a method for cleaning coke accumulation on the inner wall of a boiler based on a drum robot provided by the present invention in conjunction with the accompanying drawings.

[0023] Embodiment: The main application scenario of the present invention is as follows: When a drum robot is performing coke cleaning work on the inner wall of a boiler, a large amount of flue gas may be generated, which affects the clarity and accuracy of the thermal imaging image, resulting in inaccurate evaluation results of the coke accumulation situation on the inner wall of the boiler and affecting the efficiency of the robot's coke cleaning work. This application corrects the affected thermal imaging image to facilitate coke cleaning.

[0024] Please refer to Figure 1 , which shows a flowchart of a method for cleaning coke accumulation on the inner wall of a boiler based on a drum robot provided by an embodiment of the present invention. The method includes the following steps: Step S1, use the drum robot to capture an initial thermal imaging image of the boiler interior, and determine the coke accumulation area based on the initial thermal imaging image; use the drum robot to perform preliminary cleaning on the coke accumulation area.

[0025] First, an infrared thermal imager needs to be installed at the front end of the drum robot to capture the thermal imaging image inside the boiler. Thus, the initial thermal imaging image of the inner wall of the boiler is captured by the infrared thermal imager installed at the front end of the drum robot. Since the normal inner wall of the boiler has good heat conduction performance and a smooth surface, the temperature transition is natural, while the temperature difference between the coke accumulation area and the non-coke accumulation area is obvious. Therefore, edge detection is performed on the initial thermal imaging image to obtain the coke accumulation area. Then, the position of the drum robot is adjusted so that the robot is facing the center of the coke accumulation area and preliminary cleaning is performed on the coke accumulation position.

[0026] Also, since it may be difficult to directly clean it completely in one cleaning, it is necessary to evaluate the situation of the currently cleaned coke accumulation area after preliminary cleaning to determine whether it is cleaned. Therefore, after preliminary cleaning, a thermal imaging image of the coke accumulation area is captured again to evaluate the cleaning situation.

[0027] However, more flue gas (including water vapor, carbon dioxide, soot particles, etc.) may be generated during the cleaning process. The thermal imaging image of a coke accumulation area captured again after preliminary cleaning may be interfered with and its accuracy may be reduced. To facilitate judging the degree of interference of the image by the flue gas, a flue gas analyzer and a light scattering dust meter (laser) also need to be installed on the robot to monitor the water vapor concentration H, carbon dioxide concentration CO, and soot particle concentration PM in the current environment respectively, and collect relevant flue gas concentration data while capturing thermal imaging images before and after cleaning.

[0028] Step S2, judge whether the subsequent captured image is interfered based on the water vapor concentration, carbon dioxide concentration, and soot particle concentration inside the boiler after preliminary cleaning; if the subsequent captured image is interfered, continuously capture thermal imaging images of a coke accumulation area within a set time period to obtain a thermal imaging image sequence.

[0029] Water vapor, carbon dioxide, and soot particles in the interfering flue gas are the main factors interfering with thermal imaging images. Therefore, the higher the concentrations of water vapor, carbon dioxide, and soot particles in the environment during thermal imaging image capture, the more likely the captured thermal imaging image will be interfered with. Since the concentrations of water vapor, carbon dioxide, and soot particles in the environment are mainly generated during the cleaning process, the relevant flue gas concentration before preliminary cleaning is used as the normal concentration in the boiler. Based on the difference between the relevant flue gas concentration after preliminary cleaning and the normal concentration, the possibility of the current thermal imaging image being interfered with is calculated.

[0030] Based on the water vapor concentration, carbon dioxide concentration, and soot particle concentration in the boiler after preliminary cleaning, it is determined whether the subsequent captured images are interfered with.

[0031] Specifically, the differences between the water vapor concentrations in the boiler after and before preliminary cleaning, the differences between the carbon dioxide concentrations in the boiler after and before preliminary cleaning, and the differences between the soot particle concentrations in the boiler after and before preliminary cleaning are respectively normalized and averaged to obtain the possibility of the thermal imaging image being interfered with.

[0032] The calculation model for the possibility of the thermal imaging image being interfered with is specifically: , Where, represents the possibility of the thermal imaging image being interfered with, , and respectively represent the water vapor concentration, carbon dioxide concentration, and soot particle concentration in the boiler after preliminary cleaning, , and respectively represent the water vapor concentration, carbon dioxide concentration, and soot particle concentration in the boiler before preliminary cleaning; represents the increase value of the water vapor concentration in the environment after cleaning compared to before. The larger this difference, the greater the possibility that the subsequent captured thermal imaging image will be interfered with, is used to normalize the increase value of the water vapor concentration, represents the possibility that the subsequent captured thermal imaging image is interfered with by water vapor; represents the maximum value of the historical water vapor concentration data in the boiler, represents the maximum value of the historical carbon dioxide concentration data in the boiler, represents the maximum value of the historical soot particle concentration data in the boiler; represents the possibility that the subsequent captured thermal imaging image is interfered with by carbon dioxide, represents the possibility that the subsequent captured thermal imaging image is interfered with by soot particles. The possibility of interference in capturing a thermal imaging image is represented by the average value of the possibility of interference in subsequent captured thermal imaging images caused by water vapor, carbon dioxide, and soot particles.

[0033] Furthermore, a first threshold is set. Preferably, the value of the first threshold is 0.4 (reference value, and the implementer can adjust its value according to the actual situation. If higher sensitivity is required, its value can be appropriately increased). If the possibility of interference in the thermal imaging image is greater than or equal to the first threshold, then the subsequent captured image is interfered. At this time, continuous shooting of a focusing area needs to be performed within a set time period to obtain a thermal imaging image sequence corresponding to the focusing area, and then subsequent analysis is carried out to facilitate image correction and obtain a more accurate evaluation of the focusing situation. Preferably, the set time period is 1 minute, and the shooting interval during continuous shooting is 10 seconds. The implementer can adjust it based on the actual situation.

[0034] If the possibility of interference in the thermal imaging image is less than the first threshold, it indicates that the accuracy of the evaluation image of the focusing area is relatively high, and then the thermal imaging image of the subsequent captured focusing area can be directly used for evaluating the focusing situation. Additionally, it should be noted that if it is determined that the subsequent captured image is not interfered after preliminary cleaning, then the thermal imaging image of the subsequent captured focusing area can be directly used for evaluating the focusing situation. After evaluation, if it is necessary to continue with the focusing cleaning of this focusing area, then after cleaning this focusing area, it is necessary to determine again whether the image is interfered, and then determine whether to directly use the thermal imaging image of the subsequent captured focusing area for judging the focusing situation, or whether to capture a continuous image to form a thermal imaging image sequence for subsequent analysis and then judge the focusing situation, until the evaluation result is that the focusing in this focusing area is completely cleaned. At this time, the focusing situation of the next focusing area is evaluated, cleaned, and analyzed according to the plan, following the same process as the first focusing area analyzed, until the focusing in the next focusing area is completely removed.

[0035] Step S3: Based on the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence, the temperature value stability and temperature value change consistency of the pixel points at this position are respectively obtained, and the stability degree of the pixel points at this position is obtained; stable pixel points and unstable pixel points are obtained according to the stability degree of the pixel points at each position.

[0036] In the case where the subsequent captured images of a focusing area are affected by water vapor and smoke concentration, it is necessary to analyze the thermal imaging image sequence of this focusing area to obtain an image that can ultimately be used for evaluating the focusing situation.

[0037] Water vapor, carbon dioxide, and soot particles affect thermal imaging images by absorbing, reflecting, and blocking thermal infrared rays. Although the cleaning process will increase the concentration of these soot and gases, they will gradually settle and recover after cleaning. That is, the water vapor, carbon dioxide, and soot particles in the environment after cleaning are dynamically changing. Therefore, the pixel points disturbed in the thermal imaging image will show unstable changes. The real boiler inner wall and coke deposits do not change over time. Therefore, by comparing the changes in multiple consecutive thermal imaging images, inaccurate thermal imaging areas (areas disturbed by flue gas) can be identified.

[0038] First, since the temperature values of pixel points not disturbed do not change significantly in each image of the thermal imaging image sequence corresponding to a coke deposit area, it is necessary to obtain the temperature values of the same position on each image in the thermal image sequence and analyze them. For example, for a pixel point with coordinates (a, b) on the first image, then search for pixel points with coordinates (a, b) on the remaining images. These pixel points are pixel points at the same position.

[0039] Based on the temperature values of pixel points at the same position on each image in the thermal imaging image sequence, the temperature value stability and temperature value change consistency of the pixel points at that position are obtained respectively.

[0040] Specifically, add the standard deviation of the temperature values of pixel points at the same position on each image in the thermal imaging image sequence to the first preset value and take the reciprocal to obtain the temperature value stability of the pixel points at that position; arrange the temperature values of the pixel points at the same position in chronological order to obtain a chronological temperature value sequence; calculate the difference between the latter temperature value and the former temperature value for every two adjacent temperature values in the chronological temperature value sequence, and sort the obtained differences in chronological order to obtain a difference sequence; obtain the temperature value change consistency of the pixel points at that position according to the difference sequence.

[0041] Among them, the calculation model of the temperature value stability is specifically: , Among them, represents the temperature value stability of the pixel points at the i-th position on each image in the thermal imaging image sequence, represents the standard deviation of the temperature values of the pixel points at the i-th position on each image in the thermal imaging image sequence, indicates that the smaller the standard deviation, the smaller the temperature change at the pixel point position, that is, the more stable the temperature value, is used to ensure that the fraction is meaningful and , and the first preset value is 1.

[0042] Moreover, without reheating, the temperature values of pixel points at the same position on each undisturbed image should change regularly, that is, the temperature may gradually decrease over time instead of fluctuating repeatedly. Therefore, based on the consistency of the temperature change of pixel points in multiple thermal imaging images.

[0043] Obtain the consistency of the temperature value change of the pixel point at this position according to the difference sequence. Specifically, calculate the absolute value of the difference between the latter difference and the former difference in every two adjacent differences in the difference sequence, then calculate the reciprocal of the average value of all absolute differences, and finally normalize it to obtain the consistency of the temperature value change.

[0044] The calculation model of the consistency of the temperature value change is specifically as follows: , where, represents the consistency of the temperature value change of the pixel point at the i-th position on each image in the thermal imaging image sequence; norm represents the normalization operation for normalization; represents the number of images in the thermal imaging image sequence, represents the difference in the temperature values of the pixel point at the i-th position between the (t + 1)-th image and the t-th image in the thermal imaging image sequence, that is, the difference between the (t + 1)-th temperature value and the t-th temperature value in the temperature value sequence; represents the absolute value of the difference in the temperature change of the pixel point at the i-th position between three adjacent thermal imaging images. The smaller the absolute value, the more stable the temperature change trend of the pixel point at the i-th position in the continuous image sequence.

[0045] Although the boiler has been shut down and cooled when the drum robot enters the boiler for cleaning, during the operation of the robot in the boiler, the inner wall of the boiler may still continue to cool down. Generally, at this time, the temperature change of the inner wall of the boiler is slow, and when the external environment is cold, the cooling inside the boiler may be aggravated. Therefore, if the normal cooling rate of the inner wall of the boiler is relatively fast, the temperature change at the position of the undisturbed pixel points will also be relatively obvious, that is, the stability of the temperature value of each pixel point cannot well distinguish between the disturbed pixel points and the undisturbed pixel points, and it is necessary to judge based on the stability of the temperature change trend. Then it is necessary to judge based on the stability of the temperature change trend. Thus, according to the minimum value of the standard deviation of the temperature values of each pixel point in multiple thermal imaging images, it represents the severity of the normal cooling of the current inner wall of the boiler, and this is used as the influence weight of the temperature value stability.

[0046] Thus, obtain the minimum and maximum values of the standard deviations of the temperature values of each pixel at each position on each image in the thermal imaging image sequence, denoted as the minimum standard deviation and the maximum standard deviation respectively; obtain the weights corresponding to the temperature value stability and the temperature value change consistency respectively based on the minimum standard deviation and the maximum standard deviation; perform weighted summation on the temperature value stability and the temperature value change consistency of the pixel at a position based on the weights corresponding to the temperature value stability and the temperature value change consistency to obtain the stability degree of the pixel at the position.

[0047] If the normal cooling rate of the inner wall of the boiler is relatively fast, the temperature values in the normal area are unstable. Therefore, it is not possible to distinguish the disturbed pixels and the undisturbed pixels well based on the temperature value stability. So, it is necessary to increase the weight of the temperature change consistency and decrease the weight of the temperature value stability, and thus obtain the weights corresponding to the temperature value stability and the temperature value change consistency respectively based on the minimum standard deviation and the maximum standard deviation. Specifically, the weight corresponding to the temperature value change consistency is the ratio of the minimum standard deviation to the maximum standard deviation; the weight corresponding to the temperature value stability is the difference between the first preset value and the weight corresponding to the temperature value change consistency.

[0048] The calculation model of the stability degree is specifically: , wherein, represents the stability degree of the pixel at the i-th position on each image in the thermal imaging image sequence, represents the temperature value change consistency of the pixel at the i-th position on each image in the thermal imaging image sequence, represents the temperature value stability of the pixel at the i-th position on each image in the thermal imaging image sequence; represents the minimum value of the standard deviation of the temperature values of the pixel in multiple frames of thermal imaging images, that is, the minimum standard deviation, represents the maximum value of the standard deviation of the temperature values of the pixel in multiple frames of thermal imaging images, that is, the maximum standard deviation. A standard deviation represents the standard deviation of the temperature values of the pixel at the same position in multiple frames of thermal imaging images, represents the severity of the normal cooling of the current inner wall of the boiler, is used to perform normalization processing on

[0049] Set a second threshold value. Preferably, the value of the second threshold is 0.7 (a reference empirical value, and the implementer can adjust the value of the second threshold according to the distribution of multiple tests or the stability of data statistics). Further, stable pixels and unstable pixels are obtained according to the stability of the pixel points at each position. Specifically, if the stability of the pixel points at a position on each image in the thermal imaging image sequence is less than the second threshold, all the corresponding pixel points at that position are unstable pixels; if the stability is greater than or equal to the second threshold, all the corresponding pixel points at that position are stable pixels. Subsequently, the unstable pixel points on the last image in the thermal imaging image sequence need to be analyzed and corrected.

[0050] Step S4: Obtain the relative flue gas concentration value of an unstable pixel point according to the distances between the unstable pixel point and the pixel points included in the defocusing area; perform weighted averaging on the stability of the unstable pixel points under a certain relative flue gas concentration value to obtain the weighted stability of this relative flue gas concentration value, and obtain a fitting curve.

[0051] Although the stability of the pixel points obtained in the above steps can reflect to a certain extent the situation of the thermal imaging image at the pixel point position being interfered by flue gas, during the defocusing process, cleaning methods such as heating may be used, which will also cause a large change in temperature. Therefore, only based on the stability of the pixel points at one position, the degree of interference at the pixel point position, that is, the accuracy of the temperature value at the pixel point position, cannot be accurately reflected.

[0052] Since water vapor, carbon dioxide, and soot particles that interfere with the authenticity of the thermal imaging image are generated during the defocusing process of the robot, the relative flue gas concentration at the defocusing position is relatively large, while the surrounding area is affected by the change in the flue gas concentration at the defocusing position, so the corresponding flue gas concentration is relatively small. Thus, according to the distances between the pixel points at different positions and the current defocusing area, the difference in the relative flue gas concentration of each pixel point position relative to the center point can be estimated. According to the flue gas concentration at different positions and the stability in multiple frames of images, the influence relationship of the flue gas concentration on the stability of the pixel points in the image is obtained, and then the inaccurate pixel points are corrected according to the temperature values of the pixel points less affected.

[0053] Obtain the relative flue gas concentration value of an unstable pixel point according to the distances between the unstable pixel point and the pixel points included in the defocusing area. For the pixel points at the same position, if the pixel points at that position are unstable pixels, since their positions are the same on each image, only one unstable pixel point needs to be selected for calculation.

[0054] Specifically, based on the coordinates of an unstable pixel in the image and the coordinates of each pixel in the focus area within the focus area, the distances between the unstable pixel and each pixel in the focus area are obtained using the Euclidean distance calculation method, and the sum of the distances is obtained; the reciprocal of the sum of the distances is taken and normalized to obtain the relative flue gas concentration value of the unstable pixel.

[0055] The calculation model of the relative flue gas concentration value is specifically as follows: , where, represents the relative flue gas concentration value of the i-th unstable pixel, represents the number of pixels in the focus area, and the focus area is the initially detected focus area; represents the coordinates of the i-th unstable pixel in the image, represents the coordinates of the j-th pixel in the focus area, represents the distance between the two obtained using the Euclidean distance calculation method based on the coordinates of the i-th unstable pixel and the coordinates of the j-th pixel in the focus area, indicates that the smaller the distance between the i-th unstable pixel and the pixels in the focus area, the greater its relative flue gas concentration, and norm represents the normalization operation. The relative flue gas concentration value refers to the relative flue gas concentration value at the location of the unstable pixel.

[0056] Next, project the stability levels and relative flue gas concentration values of the unstable pixels at all positions onto a two-dimensional rectangular coordinate system, with the abscissa being the relative flue gas concentration value and the ordinate being the stability level. The same relative flue gas concentration value may correspond to multiple stability levels. Some values may be affected by other factors rather than being unstable due to flue gas interference, and they differ significantly from the data values only affected by flue gas and do not show regularity. Therefore, at the same relative flue gas concentration value, according to the density of each stability level compared to other stability levels, calculate the possibility that each stability level is affected by flue gas. It should be noted that for each image in the thermal imaging image sequence, the same position means the same position on each image, the unstable pixels and stable pixels on each image are the same, and the stability levels and relative flue gas concentration values are also the same. Here, only the unstable pixels and stable pixels on one image need to be analyzed, with the aim of analyzing the characteristics of each position on the image.

[0057] Specifically, calculate the sum of the absolute values of the differences between the stability level of an unstable pixel at a certain relative flue gas concentration value and the stability levels of other unstable pixels at the same relative flue gas concentration value, and obtain the summation result. Take the reciprocal of the summation result and normalize it to obtain the possibility that the stability level of the unstable pixel at the relative flue gas concentration value is affected by flue gas.

[0058] Its specific calculation model is as follows: , where, represents the likelihood of being affected by flue gas for the stability degree of the v-th unstable pixel point under the r-th relative flue gas concentration value, represents the quantity of the stability degree of the unstable pixel points under the r-th relative flue gas concentration value, represents the stability degree of the v-th unstable pixel point under the r-th relative flue gas concentration value, represents the stability degree of the a-th unstable pixel point under the r-th relative flue gas concentration value. represents the sum of the differences between the stability degree of the v-th unstable pixel point and the stability degrees of other unstable pixel points under this relative flue gas concentration value. The smaller this sum value is, the greater the density of the stability degree of the v-th unstable pixel point relative to the stability degrees of other unstable pixel points, that is, the more likely it is to be affected by flue gas.

[0059] Next, the likelihood of being affected by flue gas for the stability degree of an unstable pixel point under a relative flue gas concentration value is compared with the sum of the likelihoods of being affected by flue gas for the stability degrees of all unstable pixel points under this relative flue gas concentration value to obtain the weight of the stability degree of this unstable pixel point under this relative flue gas concentration value; the weights of the stability degrees of all unstable pixel points under this relative flue gas concentration value are used to perform weighted averaging on the stability degrees of all unstable pixel points to obtain the weighted stability degree of this relative flue gas concentration value.

[0060] Its specific calculation model is as follows: , where, represents the weighted stability degree of the r-th relative flue gas concentration value, represents the number of the weighted stability degrees under the r-th relative flue gas concentration value, represents the likelihood of being affected by flue gas for the stability degree of the v-th unstable pixel point under the r-th relative flue gas concentration value, represents the likelihood of being affected by flue gas for the stability degree of the a-th unstable pixel point under the r-th relative flue gas concentration value, represents the sum; represents the weight of the stability degree of the v-th unstable pixel point under the r-th relative flue gas concentration value, represents the stability degree of the v-th unstable pixel point under the r-th relative flue gas concentration value.

[0061] Finally, the weighted stability degree of each relative flue gas concentration value and each relative flue gas concentration value are fitted to obtain a fitting curve; the abscissa of the fitting curve is each relative flue gas concentration value, and the ordinate is the weighted stability degree of each relative flue gas concentration value. The method used for fitting is the least squares method. This fitting curve represents the influence relationship of the flue gas concentration on the temperature of the pixel points in the thermal imaging image, and can further correct the stability degree of the unstable pixel points.

[0062] Step S5: Obtain the corrected stability degree of each unstable pixel point according to the fitting curve; obtain the corrected stability degree of the stable pixel points, and correct the temperature values of each unstable pixel point in the last image of the thermal imaging image sequence to obtain a corrected image; evaluate the cleaning condition of the coking area based on the corrected image.

[0063] In step S5, after obtaining the fitting curve, further, the stability degree of each unstable pixel point can be corrected according to the fitting curve. Specifically, substitute the relative flue gas concentration value of the unstable pixel point into the fitting curve to obtain the corrected stability degree of the unstable pixel point, which represents the degree of influence of the position where the unstable pixel point is located by the flue gas. The larger the value, the smaller the degree of influence by the flue gas. In addition, for the stable pixel points, it is considered that the corresponding temperature values are relatively real, so the first preset value 1 is set as its corrected stability degree.

[0064] Further, use the image in the thermal imaging image sequence closest to the current moment as the image to be corrected, and obtain the neighborhood pixel points of an unstable pixel point in the last image; use the ratio of the corrected stability degree of a neighborhood pixel point to the sum of the corrected stability degrees of all neighborhood pixel points of this unstable pixel point as the weight corresponding to this neighborhood pixel point; use the weights corresponding to the neighborhood pixel points of this unstable pixel point to perform weighted summation on the temperature values of each neighborhood pixel point to obtain the corrected value of the temperature value of this unstable pixel point; obtain the corrected values of the temperature values of all unstable pixel points in the last image to obtain a corrected image. The neighborhood pixel points of the unstable pixel point refer to the pixel points within its eight neighborhoods. The calculation model for the corrected value of the temperature value of an unstable pixel point in the last image is specifically: , where, represents the corrected value of the temperature value of the b-th unstable pixel point in the last image, represents the number of neighborhood pixel points of the b-th unstable pixel point, represents the corrected stability degree of the i-th neighborhood pixel point among the neighborhood pixel points of the b-th unstable pixel point, represents the temperature value of the i-th neighborhood pixel point among the neighborhood pixel points of the b-th unstable pixel point; denotes the sum of the correction stability degrees of the neighboring pixels of the b-th unstable pixel point. denotes the weight corresponding to the i-th neighboring pixel among the neighboring pixels of the b-th unstable pixel point. Thus, the temperature values of all the unstable pixel points in the last image are corrected to obtain a corrected image.

[0065] Furthermore, based on the corrected image, the cleaning condition of the fouling region is evaluated. Edge detection is performed on the corrected image to obtain the region with a higher temperature value, which is the fouling that has not been cleaned thoroughly and needs to be defouled again. Then, the image is taken again for evaluation until the fouling region is completely cleaned. If all regions in the corrected image have a stable temperature and a natural transition during the evaluation, it indicates that the fouling region has been cleaned thoroughly, and the defouling work for the next fouling region can be carried out according to the cleaning route. The acquisition of the fouling region is a prior art and can be obtained through edge detection, so it will not be elaborated in detail here.

[0066] In summary, by analyzing the stability degrees of different pixel points in multiple thermal imaging images, this application determines the degree of interference of the temperature value of each pixel point in the image by the flue gas, combines the change characteristics of the interfered pixel points and their adjacent pixel points in multiple images, corrects the temperature values of the interfered pixel points in the image, and then evaluates the fouling condition after defouling based on the corrected image, which can effectively improve the efficiency of the defouling work of the drum robot.

[0067] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0068] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

[0069] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for cleaning the coking on the inner wall of a boiler based on a drum robot, characterized in that, The method includes: Using a drum robot to capture an initial thermal imaging image inside the boiler, determining the fouling area based on the initial thermal imaging image; using the drum robot to conduct a preliminary cleaning of the fouling area; Judging whether the subsequent captured images are interfered with based on the water vapor concentration, carbon dioxide concentration, and soot particle concentration inside the boiler after the preliminary cleaning; if the subsequent captured images are interfered with, continuously capture thermal imaging images of a fouling area within a set time period to obtain a thermal imaging image sequence; Based on the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence, respectively obtain the temperature value stability and temperature value change consistency of the pixel points at this position, and obtain the stability degree of the pixel points at this position; obtain stable pixel points and unstable pixel points according to the stability degree of the pixel points at each position; Obtain the relative value of the flue gas concentration of an unstable pixel point according to the distance between the unstable pixel point and each pixel point included in the fouling area; perform a weighted average on the stability degrees of the unstable pixel points under a certain relative value of the flue gas concentration to obtain the weighted stability degree of this relative value of the flue gas concentration, and obtain a fitting curve; Obtain the corrected stability degree of each unstable pixel point according to the fitting curve; obtain the corrected stability degree of the stable pixel points, and correct the temperature values of the unstable pixel points in the last image of the thermal imaging image sequence to obtain a corrected image; evaluate the cleaning situation of the fouling area based on the corrected image.

2. The method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, wherein, The judging whether the subsequent captured images are interfered with based on the water vapor concentration, carbon dioxide concentration, and soot particle concentration inside the boiler after the preliminary cleaning includes: Normalize and average the difference in water vapor concentration inside the boiler before and after the preliminary cleaning, the difference in carbon dioxide concentration inside the boiler before and after the preliminary cleaning, and the difference in soot particle concentration inside the boiler before and after the preliminary cleaning to obtain the possibility of interference of the thermal imaging image; if the possibility of interference of the thermal imaging image is greater than or equal to the first threshold, the subsequent captured images are interfered with.

3. A method for cleaning the coking on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, The obtaining the temperature value stability and temperature value change consistency of the pixel points at a certain position based on the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence includes: Add the standard deviation of the temperature values of the pixel points at the same position on each image in the thermal imaging image sequence to the first preset value and take the reciprocal to obtain the temperature value stability of the pixel points at this position; arrange the temperature values of the pixel points at the same position in chronological order to obtain a chronological temperature value sequence; obtain the difference between the latter temperature value and the former temperature value in each two adjacent temperature values in the chronological temperature value sequence, and sort the obtained differences in chronological order to obtain a difference sequence; obtain the temperature value change consistency of the pixel points at this position according to the difference sequence.

4. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 3, characterized in that, The obtaining the temperature value change consistency of the pixel points at this position according to the difference sequence includes: Obtain the absolute value of the difference between the latter difference and the former difference in each two adjacent differences in the difference sequence, calculate the reciprocal of the average value of all the absolute values of the differences, and then normalize it to obtain the temperature value change consistency.

5. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, The specific method for obtaining the stability degree is: The minimum and maximum values of the standard deviations of the temperature values of the pixels at each position in each image in the thermal imaging image sequence are obtained, which are recorded as the minimum standard deviation and the maximum standard deviation, respectively. The weights corresponding to the temperature value stability and the temperature value change consistency are obtained based on the minimum standard deviation and the maximum standard deviation, respectively. Based on the weights corresponding to the temperature value stability and the temperature value change consistency, the temperature value stability and the temperature value change consistency of the pixel point at a position are weightedly summed to obtain the stability of the pixel point at the position.

6. A method for cleaning the coking on the inner wall of a boiler based on a drum robot according to claim 5, characterized in that, The step of respectively obtaining weights corresponding to the temperature value stability and the temperature value change consistency based on the minimum standard deviation and the maximum standard deviation includes: The weight corresponding to the consistency of temperature value change is the ratio of the minimum standard deviation to the maximum standard deviation; the weight corresponding to the temperature value stability is the difference between the first preset value and the weight corresponding to the consistency of temperature value change.

7. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, The obtaining of stable pixels and unstable pixels according to the stability of the pixels at each position includes: Specifically, if the stability of a pixel point at a position on each image in the thermal imaging image sequence is less than a second threshold, all the corresponding pixel points at that position are unstable pixel points; if the stability is greater than or equal to the second threshold, all the corresponding pixel points at that position are stable pixel points.

8. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, The step of obtaining the relative value of the smoke concentration of an unstable pixel point based on the distance between the unstable pixel point and each pixel point included in the focal accumulation area includes: Based on the coordinates of an unstable pixel point in the image and the coordinates of each pixel point in the focused area, the Euclidean distance calculation method is used to obtain the distances between the unstable pixel point and each pixel point in the focused area, and the sum of the distances is obtained. The sum of the distances is inverted and normalized to obtain the relative smoke concentration value of the unstable pixel point.

9. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, The weighted average of the stability levels of unstable pixels at a relative value of smoke concentration to obtain the weighted stability level of the relative value of smoke concentration includes: The absolute values of the differences between the stability of an unstable pixel under a certain relative smoke concentration value and the stability of other unstable pixels under the same relative smoke concentration value are calculated and summed to obtain a summation result. The summation result is inverted and normalized to obtain the possibility that the stability of the unstable pixel under the relative smoke concentration value is affected by smoke; The possibility that the stability of an unstable pixel point under a relative smoke concentration value is affected by smoke is compared with the sum of the possibilities that the stability of each unstable pixel point under this relative smoke concentration value is affected by smoke to obtain the weight of the stability of the unstable pixel point under this relative smoke concentration value; the weight of the stability of each unstable pixel point under this relative smoke concentration value is used to weightedly average the stability of each unstable pixel point to obtain the weighted stability of this relative smoke concentration value.

10. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, The obtaining of the fitting curve comprises: The weighted stability of each relative smoke concentration value and each relative smoke concentration value are fitted to obtain a fitting curve; the abscissa of the fitting curve is each relative smoke concentration value, and the ordinate is the weighted stability of each relative smoke concentration value.

11. A method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, Obtaining the correction stability degree of each unstable pixel point according to the fitting curve includes: Substituting the relative flue gas concentration value of the unstable pixel point into the fitting curve to obtain the correction stability degree of the unstable pixel point.

12. The method for cleaning the coke deposits on the inner wall of a boiler based on a drum robot according to claim 1, wherein Correcting the temperature values of each unstable pixel point in the last image of the thermal imaging image sequence to obtain a corrected image includes: Obtaining the neighborhood pixel points of an unstable pixel point in the last image; taking the ratio of the correction stability degree of a neighborhood pixel point to the sum of the correction stability degrees of all neighborhood pixel points of the unstable pixel point as the weight corresponding to the neighborhood pixel point; using the weights corresponding to the neighborhood pixel points of the unstable pixel point to perform weighted summation on the temperature values of the neighborhood pixel points to obtain the corrected value of the temperature value of the unstable pixel point; obtaining the corrected values of the temperature values of all unstable pixel points in the last image to obtain a corrected image.

13. A method for cleaning coking on the inner wall of a boiler based on a drum robot according to claim 1, characterized in that, Obtaining the correction stability degree of the stable pixel point includes: Setting the first preset value as the correction stability degree of the stable pixel point.

Images