Heating furnace operation monitoring method and system based on infrared thermal imaging

Abstract

The invention discloses a heating furnace operation monitoring method and system based on infrared thermal imaging, and relates to the technical field of heating furnace operation monitoring, and the method comprises the following steps: erecting an infrared thermal imager, carrying out the comprehensive monitoring of a heating furnace, and obtaining a thermal imaging image; performing image preprocessing on the thermal imaging image, and converting the thermal imaging image into a high-quality image; performing image analysis on the high-quality image, and identifying different heat energy areas and heat energy transition areas in the heating furnace; analyzing the relationship between the heat energy transition area and the heat energy area to obtain a heat energy transition rule; judging whether the heating furnace operates normally based on the heat energy transition rule; the method is used for solving the problem that the abnormal state of the heating furnace cannot be timely and effectively recognized due to the fact that factors influencing the operation state of the heating furnace are not comprehensively considered in an existing heating furnace operation monitoring technology.

Description

Technical Field

[0001] This invention relates to the field of heating furnace operation monitoring technology, specifically to a heating furnace operation monitoring method and system based on infrared thermal imaging. Background Technology

[0002] Furnace operation monitoring technology refers to a technical system that uses sensor networks, data acquisition systems, and intelligent analysis methods to measure, evaluate, and provide early warnings of the operating status of a furnace in real time or periodically. Its core objectives are to ensure equipment safety, optimize operating efficiency, extend lifespan, and reduce maintenance costs.

[0003] Existing furnace operation monitoring technologies typically monitor whether the temperature within the furnace is normal. A furnace usually has three zones: a preheating zone, a heating zone, and a soaking zone, each with a different temperature. Current monitoring technologies generally determine normal operation by directly monitoring whether the temperature in different zones is within normal thresholds. However, because there are different temperature zones, there is a temperature transition zone between them. If this transition zone is abnormal, it indicates a temperature imbalance within the furnace, which can easily lead to material defects. Therefore, it is necessary to focus on monitoring this temperature transition zone, as illustrated in publication number CN119. Patent application 309428A discloses a "monitoring method, device, furnace, and storage medium for an aluminum rod heating furnace". This solution determines whether the furnace is abnormal by analyzing the correlation between the control system and temperature and judging whether the sensors are abnormal. Essentially, it analyzes whether the actual temperature in the furnace matches the expected temperature to determine whether the furnace is abnormal. However, this method cannot determine whether the temperature changes in different areas of the furnace are normal, so it is prone to misjudgment. Existing furnace operation monitoring technology also has the problem that it does not consider the factors affecting the furnace operation status in a comprehensive manner, resulting in the inability to identify abnormal states of the furnace in a timely and effective manner. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. By setting up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal imaging images, the thermal imaging images are then preprocessed to convert them into high-quality images. Image analysis is then performed on the high-quality images to identify different thermal energy regions and thermal energy transition regions in the heating furnace. At the same time, historical infrared data of the normally operating heating furnace is acquired and the boundary midpoints are analyzed. Based on the boundary midpoints, the thermal energy transition law is analyzed. Finally, based on the thermal energy transition law, it is determined whether the heating furnace is operating normally. This solves the problem that existing heating furnace operation monitoring technology does not consider the factors affecting the operating status of the heating furnace in a comprehensive manner, resulting in the inability to identify abnormal states of the heating furnace in a timely and effective manner.

[0005] To achieve the above objectives, in a first aspect, this application provides a method for monitoring the operation of a heating furnace based on infrared thermal imaging, comprising the following steps:

[0006] An infrared thermal imager was installed to comprehensively monitor the heating furnace and acquire thermal imaging images;

[0007] Perform image preprocessing on thermal imaging images to convert them into high-quality images;

[0008] Image analysis is performed on high-quality images to identify different thermal energy regions and thermal energy transition regions in the heating furnace;

[0009] By analyzing the relationship between the thermal energy transition region and the thermal energy region, the thermal energy transition law is obtained.

[0010] Determine whether the heating furnace is operating normally based on the thermal energy transition law.

[0011] Furthermore, an infrared thermal imager is set up to comprehensively monitor the heating furnace and acquire thermal imaging images, including the following sub-steps:

[0012] The heating furnace includes a preheating section, a heating section, and a soaking section;

[0013] An infrared thermal imager was set up to monitor the preheating section, heating section, and heat soaking section in a unified manner, and thermal images were obtained.

[0014] Furthermore, the image preprocessing of the thermal imaging image includes denoising, enhancement, and grayscale processing to ultimately obtain a high-quality image.

[0015] Furthermore, image analysis is performed on the high-quality images to identify different thermal energy regions and thermal energy transition regions in the heating furnace, including the following sub-steps:

[0016] The pixel in the nth row and mth column of the high-quality image is labeled as P(n,m), and the gray value of P(n,m) is labeled as H(n,m), where n and m are both positive integers and (n,m) is the index of P and H.

[0017] When analyzing any H(n,m), it is named the center gray level, and the H(n,m) adjacent to the center gray level is named the adjacent gray level.

[0018] Calculate the absolute value of the difference between each adjacent gray level and the center gray level, and name the calculation result as gray level difference. Calculate all gray level differences with each H(n,m) as the center gray level.

[0019] Further analysis of the grayscale difference was conducted to identify different thermal energy zones and thermal energy transition zones in the heating furnace.

[0020] Further analysis of the grayscale difference and identification of different thermal energy zones and thermal energy transition zones in the heating furnace includes the following sub-steps:

[0021] Construct a one-dimensional coordinate system with the gray-level difference as the X-axis, name it gray-level distribution map, and record the gray-level difference values ​​in the gray-level distribution map;

[0022] Cluster analysis was performed on the grayscale distribution image to obtain grayscale difference clusters. These grayscale difference clusters were then numbered and labeled with the symbol S. i This indicates that i is a positive integer and i is the index of S;

[0023] The gray-scale difference clustering population is divided into two clustering types: non-transitional clustering and transitional clustering. The statistical S... i The number of grayscale differences in the data is named the difference quantity. The two grayscale difference clusters with the largest difference quantity are marked as the initial parent clusters. The initial parent cluster on the left is marked as the non-transitional parent cluster, the initial parent cluster on the right is marked as the transitional parent cluster, and the remaining grayscale difference clusters are marked as the initial sub-clusters.

[0024] Calculate the distance between the initial sub-cluster and the initial parent cluster, named the cluster distance. The cluster distance is the difference between the nearest grayscale difference between the initial sub-cluster and the initial parent cluster. The cluster distance includes non-transition distance and transition distance.

[0025] If the non-transition distance is less than the transition distance, the initial sub-cluster is incorporated into the non-transition mother cluster; otherwise, the initial sub-cluster is incorporated into the transition mother cluster. After all the initial sub-clusters have been divided, the non-transition mother cluster is the non-transition cluster, and the transition mother cluster is the transition cluster.

[0026] The grayscale difference calculated with any H(n,m) as the center grayscale is named the judgment difference. Non-transition parameter and transition parameter are set. The non-transition parameter and transition parameter are initially 0. If the judgment difference is in the non-transition cluster, the non-transition parameter is incremented by one. If the judgment difference is in the transition cluster, the transition parameter is incremented by one.

[0027] For each judgment difference, analyze it. After the analysis is completed, if the non-transition parameter is greater than the transition parameter, mark P(n,m) corresponding to the center gray level at this time as a non-transition pixel; otherwise, mark P(n,m) corresponding to the center gray level at this time as a transition pixel.

[0028] A non-transition region is formed by consecutive adjacent non-transition pixels, and a transition region is formed by consecutive adjacent transition pixels. The average value of H(n,m) in the non-transition region is calculated and named the non-transition average gray level. The average value of H(n,m) in the transition region is calculated and named the transition average gray level.

[0029] The non-transition region with the highest average gray value is marked as the preheating region, the non-transition region with the lowest average gray value is marked as the heating region, and the remaining non-transition regions are marked as the homogenization region. The transition region between the preheating region and the heating region is marked as the preheating-heating transition region, and the remaining transition regions are marked as the heating-homogenization transition region. The preheating region, the heating region, and the homogenization region constitute the thermal energy region, and the preheating-heating transition region and the heating-homogenization transition region constitute the thermal energy transition region.

[0030] Furthermore, analyzing the relationship between the thermal energy transition region and the thermal energy region, the thermal energy transition law is obtained, which includes the following sub-steps:

[0031] Acquire historical infrared data of a normally operating heating furnace and analyze the midpoint of the boundary;

[0032] The thermal energy transition law is analyzed based on the midpoint of the boundary.

[0033] Furthermore, acquiring historical infrared data of a normally operating heating furnace and analyzing the boundary midpoint includes the following sub-steps:

[0034] The historical infrared data refers to the historical data of thermal energy region, thermal energy transition region and high-quality image. When the heating temperature in the heating furnace changes, the gray value, thermal energy region and thermal energy transition region division of the heating furnace in the high-quality image will also change accordingly.

[0035] Using the side length of a pixel in the high-quality image as a unit length, draw the horizontal and vertical axes to construct a two-dimensional coordinate system, named the boundary line confirmation map. Delineate the thermal energy region and the thermal energy transition region within the high-quality image to obtain the region division map. Align the lower left vertex of the region division map with the origin and ensure that the entire region division map is within the first quadrant.

[0036] Name the rectangular square corresponding to the pixel in the region division map as the first square, name the first square adjacent to the thermal energy region and the thermal energy transition region as the second square, and remove the first square.

[0037] The intersection line between two adjacent second squares in the same row is named the boundary intersection line, and the midpoint of the boundary intersection line is marked as the boundary intersection point. The boundary intersection point includes the first boundary intersection point, the second boundary intersection point, the third boundary intersection point, and the fourth boundary intersection point, which correspond to the intersection point between the preheating area and the preheating heating transition area, the intersection point between the preheating heating transition area and the heating area, the intersection point between the heating area and the heating homogenization transition area, and the intersection point between the heating homogenization transition area and the homogenization area, respectively.

[0038] Linear regression was performed on the first, second, third, and fourth boundary intersection points, respectively. The resulting line segments were named the first boundary line segment, the second boundary line segment, the third boundary line segment, and the fourth boundary line segment, respectively. The midpoints of the first, second, third, and fourth boundary line segments were obtained and named the first boundary midpoint, the second boundary midpoint, the third boundary midpoint, and the fourth boundary midpoint, respectively.

[0039] The midpoints of the first, second, third, and fourth boundaries are the boundary midpoints. Different boundary midpoints are obtained by analyzing each set of historical infrared data.

[0040] Furthermore, the analysis of thermal energy transition based on the midpoint of the boundary includes the following sub-steps:

[0041] Obtain the heating temperature corresponding to the historical infrared data. With the heating temperature as the X-axis, establish four two-dimensional coordinate systems with the midpoints of the first, second, third, and fourth boundaries as the Y-axis. Name them the First Boundary Pattern Map, Second Boundary Pattern Map, Third Boundary Pattern Map, and Fourth Boundary Pattern Map respectively, and collectively name them Boundary Pattern Map.

[0042] Discrete regression analysis was performed on the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram to obtain the first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve. The first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve were collectively named the boundary pattern curve.

[0043] Increase the line width of the boundary law curve so that the boundary law curve exactly covers all the coordinate points in the boundary law diagram. The final boundary law curve is the thermal energy transition law, which includes the first transition law, the second transition law, the third transition law, and the fourth transition law.

[0044] Furthermore, determining whether the heating furnace is operating normally based on the thermal energy transition law includes the following sub-steps:

[0045] Obtain the current heating temperature of the furnace, name it the real-time temperature, and label it T;

[0046] The midpoints of the first, second, third, and fourth boundaries of the heating furnace are analyzed and calculated, and labeled as Q1, Q2, Q3, and Q4, respectively.

[0047] Enter the coordinates (T,Q1), (T,Q2), (T,Q3), and (T,Q4) into the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram, respectively. Determine whether they are in the first transition pattern, the second transition pattern, the third transition pattern, and the fourth transition pattern, respectively. If they are, output a normal operation signal; otherwise, output an abnormal operation signal.

[0048] Secondly, this application provides a heating furnace operation monitoring system based on infrared thermal imaging, including an infrared monitoring module, an image preprocessing module, an area recognition module, a transition law analysis module, and an operation status judgment module; the infrared monitoring module, the image preprocessing module, the area recognition module, and the transition law analysis module are respectively connected to the operation status judgment module for data transmission.

[0049] The infrared monitoring module is used to set up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal imaging images.

[0050] The image preprocessing module is used to preprocess the thermal imaging image and convert it into a high-quality image.

[0051] The region recognition module is used to perform image analysis on high-quality images and identify different thermal energy regions and thermal energy transition regions in the heating furnace.

[0052] The transition law analysis module is used to analyze the relationship between the thermal energy transition region and the thermal energy region to obtain the thermal energy transition law;

[0053] The operating status judgment module is used to determine whether the heating furnace is operating normally based on the thermal energy transition law.

[0054] The beneficial effects of this invention are as follows: This invention uses an infrared thermal imager to comprehensively monitor the heating furnace, acquire thermal imaging images, preprocess the thermal imaging images to convert them into high-quality images, and then perform image analysis on the high-quality images to identify different thermal energy regions and thermal energy transition regions in the heating furnace. The advantage is that it can accurately identify the preheating section, heating section and soaking section in the heating furnace, and at the same time identify the transition regions between them, so as to accurately analyze the pattern of thermal energy transition regions, thereby improving the effectiveness and rationality of heating furnace operation monitoring.

[0055] This invention acquires historical infrared data of a normally operating heating furnace, analyzes the boundary midpoint, and analyzes the thermal energy transition law based on the boundary midpoint. Finally, it determines whether the heating furnace is operating normally based on the thermal energy transition law. The advantage is that changes in heating temperature will also lead to changes in the thermal energy transition area, and there is a certain relationship between them, namely the thermal energy transition law. This allows analysis of whether the thermal energy transition area between different thermal energy areas is within the normal range. If the thermal energy transition area exceeds the thermal energy transition law, it means that the temperature between the two thermal energy areas is out of balance. When the material passes through the thermal energy transition area, it is easily affected, leading to material defects. Therefore, analyzing the thermal energy transition law ensures the normality of the thermal energy transition area, improving the accuracy and effectiveness of heating furnace operation monitoring. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the system of the present invention;

[0057] Figure 2 This is the grayscale distribution diagram of the present invention;

[0058] Figure 3 This is a schematic diagram of grayscale difference clustering according to the present invention;

[0059] Figure 4 This is a cross-sectional view of the heating furnace of the present invention;

[0060] Figure 5 This is a schematic diagram of the intersection lines between the second squares of the present invention;

[0061] Figure 6 This is a schematic diagram of the boundary intersection points of the present invention;

[0062] Figure 7 This is a schematic diagram of the first boundary intersection point, the first boundary line segment, and the midpoint of the first boundary according to the present invention.

[0063] Figure 8 This is the first boundary law diagram of the present invention;

[0064] Figure 9 This is a schematic diagram of the first boundary law curve of the present invention;

[0065] Figure 10 This is a schematic diagram of the first transition law of the present invention;

[0066] Figure 11 This is a flowchart of the steps of the method of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Example 1, please refer to Figure 1 As shown, this application provides a heating furnace operation monitoring system based on infrared thermal imaging, including an infrared monitoring module, an image preprocessing module, an area recognition module, a transition law analysis module, and an operation status judgment module; the infrared monitoring module, the image preprocessing module, the area recognition module, and the transition law analysis module are respectively connected to the operation status judgment module for data transmission.

[0069] The infrared monitoring module is used to set up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal imaging images;

[0070] The infrared monitoring module is configured with an infrared monitoring strategy, which includes:

[0071] The heating furnace includes a preheating section, a heating section, and a soaking section;

[0072] An infrared thermal imager was set up to monitor the preheating section, heating section, and heat soaking section in a unified manner, and thermal images were obtained.

[0073] In practical applications, the ranges of the preheating section, heating section, and soaking section in a heating furnace are not clearly defined, requiring precise identification through a region recognition module. This is only used to illustrate comprehensive monitoring of the heating furnace using an infrared thermal imager; the goal is simply to ensure that the preheating section, heating section, and soaking section can be identified in the thermal imaging image. Only the furnace body is captured in the thermal imaging image, with the background completely removed. Since the positions of the infrared thermal imager and the heating furnace are fixed, the heating furnace in the captured thermal imaging image is also at a fixed pixel point. Therefore, retaining only the fixed pixel point is sufficient to perfectly remove the background, preventing it from affecting region recognition. Typically, the heating furnace adjusts the temperature of the heating section, while the temperatures of the preheating and soaking sections change with the heating section temperature. There is a certain proportional relationship between them, and these are inherent settings of the heating furnace, requiring no adjustment in this embodiment. Therefore, each temperature value in the heating section has a corresponding temperature value in the preheating and soaking sections for balance.

[0074] The image preprocessing module is used to preprocess thermal imaging images and convert them into high-quality images. Image preprocessing of thermal imaging images includes denoising, enhancement, and grayscale processing to finally obtain high-quality images.

[0075] In practical applications, image preprocessing techniques are all performed using existing denoising and image enhancement techniques, and then the images are converted into grayscale images to obtain high-quality images.

[0076] The region recognition module is used to perform image analysis on high-quality images and identify different thermal energy regions and thermal energy transition regions in the heating furnace; the region recognition module includes a grayscale analysis unit and a region recognition unit.

[0077] The grayscale analysis unit is configured with a grayscale analysis strategy, which includes:

[0078] The pixel in the nth row and mth column of the high-quality image is labeled as P(n,m), and the gray value of P(n,m) is labeled as H(n,m), where n and m are both positive integers and (n,m) is the index of P and H.

[0079] When analyzing any H(n,m), it is named the center gray level, and the H(n,m) adjacent to the center gray level is named the adjacent gray level.

[0080] Calculate the absolute value of the difference between each adjacent gray level and the center gray level, and name the calculation result as gray level difference. Calculate all gray level differences with each H(n,m) as the center gray level.

[0081] Further analysis of the grayscale difference was conducted to identify different thermal energy zones and thermal energy transition zones in the heating furnace;

[0082] In practical applications, H(n,m) reflects the temperature of the heating furnace at point P(n,m). The number of adjacent gray values ​​for each H(n,m) is different, but they can be generally divided into eight adjacent gray values: top left, top right, top right, left, right, bottom left, bottom right, and bottom right. However, if H(n,m) is located at the edge of the heating furnace, its adjacent pixels are removed, and the corresponding adjacent gray values ​​are null values. For example, if a certain central gray value has three adjacent H(n,m), then three gray value differences are calculated. Assuming the central gray value is 156, and the adjacent gray values ​​are 177, 168, and 154, the gray value differences are 21, 12, and 2, respectively. Analyzing the gray value difference of each P(n,m), due to the large number of gray value differences, this embodiment is not convenient to show in detail, but only provides a simplified illustration through the subsequent gray value distribution diagram.

[0083] The region identification unit is configured with a region identification strategy, which includes:

[0084] Please see Figure 2 As shown, a one-dimensional coordinate system is constructed with the gray-level difference as the X-axis, named the gray-level distribution map, and the gray-level difference is entered into the gray-level distribution map;

[0085] Please see Figure 3As shown, cluster analysis is performed on the grayscale distribution image to obtain grayscale difference clusters. These grayscale difference clusters are then numbered and labeled with the symbol S. i This indicates that i is a positive integer and i is the index of S;

[0086] The gray-scale difference clustering population is divided into two clustering types: non-transitional clustering and transitional clustering. The statistical S... i The number of grayscale differences in the data is named the difference quantity. The two grayscale difference clusters with the largest difference quantity are marked as the initial parent clusters. The initial parent cluster on the left is marked as the non-transitional parent cluster, the initial parent cluster on the right is marked as the transitional parent cluster, and the remaining grayscale difference clusters are marked as the initial sub-clusters.

[0087] In practical applications, the constructed grayscale distribution map is as follows: Figure 2 As shown, cluster analysis yields gray-scale difference clusters as follows: Figure 3 As shown, there are 5 rectangles in total, each representing a gray-level difference cluster. In this embodiment, they are numbered S1 to S5 from left to right. Figure 2 It can be seen that S1 and S5 have the most differences. Since S1 is on the left, S1 is marked as a non-transitional mother cluster, S5 is marked as a transitional mother cluster, and S2 to S4 are all initial sub-clusters.

[0088] Calculate the distance between the initial sub-cluster and the initial parent cluster, named the cluster distance. The cluster distance is the difference between the nearest grayscale difference between the initial sub-cluster and the initial parent cluster. The cluster distance includes non-transition distance and transition distance.

[0089] If the non-transition distance is less than the transition distance, the initial sub-cluster is incorporated into the non-transition mother cluster; otherwise, the initial sub-cluster is incorporated into the transition mother cluster. After all the initial sub-clusters have been divided, the non-transition mother cluster is the non-transition cluster, and the transition mother cluster is the transition cluster.

[0090] In practical applications, taking S2 as an example, S2 is to the right of S1, meaning S2 is greater than S1. Therefore, the non-transition distance can be obtained by calculating the difference between the maximum value of X in S1 and the minimum value of X in S2. Similarly, S2 is to the left of S5, meaning S2 is less than S5. The transition distance can be obtained by calculating the difference between the maximum value of X in S2 and the minimum value of X in S5. Figure 3It is easy to see that the non-transition distance is less than the transition distance, so S2 is included in the non-transition parent cluster. Similarly, S3 and S4 are included in the transition parent cluster. Finally, the non-transition cluster includes S1 and S2, and the transition cluster includes S3 to S5. Since there is still a certain gap between S1 and S2, the final non-transition cluster is actually the range of minimum and maximum values ​​of S1 and S2, and the transition cluster is the range of minimum and maximum values ​​of S3 to S5. Finally, the non-transition cluster is [0,21] and the transition cluster is [27,57].

[0091] The grayscale difference calculated with any H(n,m) as the center grayscale is named the judgment difference. Non-transition parameter and transition parameter are set. The non-transition parameter and transition parameter are initially 0. If the judgment difference is in the non-transition cluster, the non-transition parameter is incremented by one. If the judgment difference is in the transition cluster, the transition parameter is incremented by one.

[0092] For each judgment difference, analyze it. After the analysis is completed, if the non-transition parameter is greater than the transition parameter, mark P(n,m) corresponding to the center gray level at this time as a non-transition pixel; otherwise, mark P(n,m) corresponding to the center gray level at this time as a transition pixel.

[0093] In practical applications, assuming H(164,263) is the center gray level, there are a total of 8 gray level differences. Among them, there are 6 gray level differences in the non-over-clustered gray level and 2 gray level differences in the transitional cluster. Thus, the non-over-clustered parameter is 6 and the transitional parameter is 2. The non-over-clustered parameter is greater than the transitional parameter, so P(164,263) is marked as a non-transitional pixel. Similarly, all P(n,m) are analyzed to obtain all non-transitional pixels and transitional pixels.

[0094] A non-transition region is formed by consecutive adjacent non-transition pixels, and a transition region is formed by consecutive adjacent transition pixels. The average value of H(n,m) in the non-transition region is calculated and named the non-transition average gray level. The average value of H(n,m) in the transition region is calculated and named the transition average gray level.

[0095] The non-transition region with the highest average gray value is marked as the preheating region, the non-transition region with the lowest average gray value is marked as the heating region, and the remaining non-transition regions are marked as the homogenization region. The transition region between the preheating region and the heating region is marked as the preheating-heating transition region, and the remaining transition regions are marked as the heating-homogenization transition region. The preheating region, the heating region, and the homogenization region are the thermal energy regions, and the preheating-heating transition region and the heating-homogenization transition region are the thermal energy transition regions.

[0096] In practical applications, under normal circumstances, only 3 non-transition areas and 2 transition areas will be generated. If there are not 3 non-transition areas or not 2 transition areas, it indicates that there is an abnormality in the heating furnace, and an abnormality alarm signal will be output directly. Under normal circumstances, the temperature of the preheating section is lower than the temperature of the homogenizing section, which is lower than the temperature of the heating section. In high-quality images, the larger the gray value, the lower the temperature, and the smaller the gray value, the higher the temperature. Therefore, the non-transition area with the largest average gray value belongs to the preheating area, the non-transition area with the smallest average gray value belongs to the heating area, and the other non-transition area belongs to the homogenizing area. The transition area between the preheating area and the heating area is the preheating-heating transition area.

[0097] The transition law analysis module is used to analyze the relationship between thermal energy transition regions and thermal energy regions to obtain thermal energy transition laws; the transition law analysis module includes boundary midpoint analysis units and transition law analysis units.

[0098] The boundary midpoint analysis unit is used to acquire historical infrared data of a normally operating heating furnace and analyze the boundary midpoint.

[0099] The boundary midpoint analysis unit is configured with a boundary midpoint analysis strategy, which includes:

[0100] Historical infrared data includes historical data of thermal energy areas, thermal energy transition areas, and high-quality images. When the heating temperature in the furnace changes, the grayscale value, thermal energy areas, and thermal energy transition areas of the furnace in the high-quality image will also change accordingly.

[0101] Please see Figures 4 to 5 As shown, the horizontal and vertical axes are drawn with the side length of a pixel in the high-quality image as a unit length, and a two-dimensional coordinate system is constructed, named the boundary line confirmation map. The thermal energy region and the thermal energy transition region are delineated in the high-quality image to obtain the region division map. The lower left vertex of the region division map coincides with the origin, and the entire region division map is placed in the first quadrant.

[0102] Please see Figure 6 As shown, the rectangular squares corresponding to the pixels in the region division map are named the first squares, the first squares adjacent to the thermal energy region and the thermal energy transition region are named the second squares, and the first squares are removed.

[0103] The intersection line between two adjacent second squares in the same row is named the boundary intersection line, and the midpoint of the boundary intersection line is marked as the boundary intersection point. The boundary intersection points include the first boundary intersection point, the second boundary intersection point, the third boundary intersection point, and the fourth boundary intersection point, which correspond to the intersection points between the preheating area and the preheating heating transition area, the preheating heating transition area and the heating area, the heating area and the heating homogenization transition area, and the heating homogenization transition area and the homogenization area, respectively.

[0104] In practical applications, with Figure 4 For example, Figure 4 This is a cross-sectional view of the heating furnace. Under normal circumstances, the distribution range of the preheating section, heating section, and soaking section is as follows: Figure 4 As shown in the diagram, due to heat transfer effects, there exists a transition zone between the preheating section and the heating section, as well as between the heating section and the soaking section. This transition zone is further divided into a preheating-heating transition zone and a heating-soaking transition zone. Furthermore, due to the external structure of the furnace and factors such as heat conduction, the distribution of these zones is not regular. Figure 5 As shown, Figure 5 The dashed lines in the diagram are used to indicate the intersections between the second squares. Figure 6 The diagram illustrates two first squares. The first square on the left belongs to the preheating area, and the first square on the right belongs to the preheating transition area. Therefore... Figure 6 The two first squares belong to the second square, and they are in the same row and adjacent to each other, therefore Figure 6 The line of intersection between two pixels is the boundary line, and the midpoint of the boundary line is the boundary point.

[0105] Please see Figure 7 As shown, linear regression is performed on the first, second, third, and fourth boundary intersection points, respectively. The resulting line segments are named the first boundary line segment, the second boundary line segment, the third boundary line segment, and the fourth boundary line segment, respectively. The midpoints of the first, second, third, and fourth boundary line segments are obtained and named the first boundary midpoint, the second boundary midpoint, the third boundary midpoint, and the fourth boundary midpoint, respectively.

[0106] The midpoints of the first, second, third, and fourth boundaries are the boundary midpoints. By analyzing each set of historical infrared data, different boundary midpoints can be obtained.

[0107] In practical applications, since the analysis process for the first, second, third, and fourth boundary intersections is exactly the same, this embodiment only takes the first boundary intersection as an example to illustrate the specific process of obtaining the first transition law through subsequent analysis. Figure 7 The image shows the first boundary intersection point, the first boundary line segment, and the first boundary midpoint; a different boundary midpoint can be calculated for each set of historical infrared data.

[0108] The transition law analysis unit is used to analyze the thermal energy transition law based on the midpoint of the boundary;

[0109] The transition law analysis unit is configured with transition law analysis strategies, which include:

[0110] Please see Figure 8 As shown, the heating temperature corresponding to the historical infrared data is obtained. With the heating temperature as the X-axis, four two-dimensional coordinate systems are established with the midpoints of the first boundary, the second boundary, the third boundary, and the fourth boundary as the Y-axis. These systems are named the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram, respectively, and are collectively named the boundary pattern diagram.

[0111] Please see Figure 9 As shown, discrete regression analysis was performed on the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram respectively to obtain the first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve. The first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve are collectively named the boundary pattern curve.

[0112] Please see Figure 10 As shown, by increasing the line width of the boundary law curve so that the boundary law curve exactly covers all the coordinate points in the boundary law diagram, the final boundary law curve is the thermal energy transition law, which includes the first transition law, the second transition law, the third transition law, and the fourth transition law.

[0113] In practical applications, since the length of the boundary line segment on the Y-axis is fixed, meaning the midpoint of the boundary is always at the same height and only changes on the horizontal axis, the Y-axis of the boundary pattern diagram only needs to record the value of the midpoint of the boundary on the horizontal axis. The resulting first boundary pattern diagram is shown below. Figure 8 As shown, discrete regression analysis was performed on the first boundary law diagram, and the first boundary law curve was obtained as follows. Figure 9 As shown, increasing the line width ultimately yields the first transition law, as follows: Figure 10 As shown, the process of analyzing the second, third, and fourth transition laws is exactly the same as the process of analyzing the first process law, and will not be described in detail in this embodiment.

[0114] The operation status judgment module is used to determine whether the heating furnace is operating normally based on the thermal energy transition law;

[0115] The running status determination module is configured with running status determination strategies, which include:

[0116] Obtain the current heating temperature of the furnace, name it the real-time temperature, and label it T;

[0117] The midpoints of the first, second, third, and fourth boundaries of the heating furnace are analyzed and calculated, and labeled as Q1, Q2, Q3, and Q4, respectively.

[0118] Enter the coordinates (T,Q1), (T,Q2), (T,Q3), and (T,Q4) into the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram, respectively. Determine whether they are in the first transition pattern, the second transition pattern, the third transition pattern, and the fourth transition pattern, respectively. If they are, output a normal operation signal; otherwise, output an abnormal operation signal.

[0119] In practical applications, if T is obtained as 1050℃, and the midpoint Q1 of the current first boundary of the heating furnace is calculated to be 246, then the coordinates (1050, 246) are substituted into the first boundary law diagram. It is found that (1050, 246) is within the first transition law. If (T, Q2) is within the second transition law, (T, Q3) is within the third transition law, and (T, Q4) is within the fourth transition law, then a normal operation signal is output. If any of the above conditions are not met, it means that there is an abnormality in the internal temperature balance of the heating furnace, and an abnormal operation signal is output.

[0120] Example 2, please refer to Figure 11 As shown, this application provides a method for monitoring the operation of a heating furnace based on infrared thermal imaging, including the following steps:

[0121] Step S1: Set up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal imaging images; Step S1 includes the following sub-steps:

[0122] Step S101: The heating furnace includes a preheating section, a heating section, and a soaking section;

[0123] Step S102: Set up an infrared thermal imager to uniformly monitor the preheating section, heating section and heat soaking section to obtain thermal imaging images;

[0124] Step S2: Perform image preprocessing on the thermal imaging image to convert it into a high-quality image; image preprocessing on the thermal imaging image includes denoising, enhancement and grayscale processing to finally obtain a high-quality image.

[0125] Step S3 involves image analysis of the high-quality images to identify different thermal energy regions and thermal energy transition regions within the heating furnace. Step S3 includes the following sub-steps:

[0126] Step S301: Mark the pixel in the nth row and mth column of the high-quality image as P(n,m), and mark the gray value of P(n,m) as H(n,m), where n and m are both positive integers and (n,m) is the index of P and H.

[0127] Step S302: When analyzing any H(n,m), name it the center gray level and name the H(n,m) adjacent to the center gray level as the adjacent gray level.

[0128] Step S303: Calculate the absolute value of the difference between each adjacent gray level and the center gray level, name the calculation result as gray level difference, and calculate all gray level differences with each H(n,m) as the center gray level.

[0129] Step S304: Further analyze the grayscale difference and identify different thermal energy regions and thermal energy transition regions in the heating furnace;

[0130] Step S304 includes the following sub-steps:

[0131] Step S3041: Construct a one-dimensional coordinate system with the gray-level difference as the X-axis, name it gray-level distribution map, and enter the gray-level difference into the gray-level distribution map;

[0132] Step S3042: Perform cluster analysis on the grayscale distribution image to obtain grayscale difference clusters. Number the grayscale difference clusters and assign them the symbol S. i This indicates that i is a positive integer and i is the index of S;

[0133] Step S3043: Divide the gray-scale difference clustering population into two clustering types: non-transitional clustering and transitional clustering, and calculate S. i The number of grayscale differences in the data is named the difference quantity. The two grayscale difference clusters with the largest difference quantity are marked as the initial parent clusters. The initial parent cluster on the left is marked as the non-transitional parent cluster, the initial parent cluster on the right is marked as the transitional parent cluster, and the remaining grayscale difference clusters are marked as the initial sub-clusters.

[0134] Step S3044: Calculate the distance between the initial sub-cluster and the initial parent cluster, named the cluster distance. The cluster distance is the difference between the closest grayscale difference between the initial sub-cluster and the initial parent cluster. The cluster distance includes non-transition distance and transition distance.

[0135] Step S3045: If the non-transition distance is less than the transition distance, the initial sub-cluster is incorporated into the non-transition mother cluster; otherwise, the initial sub-cluster is incorporated into the transition mother cluster. When all initial sub-clusters are divided, the non-transition mother cluster is the non-transition cluster, and the transition mother cluster is the transition cluster.

[0136] Step S3046: Name the gray difference calculated with any H(n,m) as the center gray value as the judgment difference. Set non-transition parameter and transition parameter. The non-transition parameter and transition parameter are initially 0. If the judgment difference is in the non-transition cluster, the non-transition parameter is incremented by one. If the judgment difference is in the transition cluster, the transition parameter is incremented by one.

[0137] Step S3047: Analyze each judgment difference. After the analysis is completed, if the non-transition parameter is greater than the transition parameter, mark P(n,m) corresponding to the center gray level at this time as a non-transition pixel; otherwise, mark P(n,m) corresponding to the center gray level at this time as a transition pixel.

[0138] Step S3048: A non-transition region is formed by consecutive adjacent non-transition pixels, and a transition region is formed by consecutive adjacent transition pixels. Calculate the average value of H(n,m) in the non-transition region and name it as the non-transition average gray level. Calculate the average value of H(n,m) in the transition region and name it as the transition average gray level.

[0139] Step S3049: Mark the non-transition region with the largest average gray value as the preheating region, mark the non-transition region with the smallest average gray value as the heating region, mark the remaining non-transition regions as the homogenization region, mark the transition region between the preheating region and the heating region as the preheating-heating transition region, mark the remaining transition regions as the heating-homogenization transition region, and the preheating region, heating region, and homogenization region together constitute the thermal energy region, and the preheating-heating transition region and the heating-homogenization transition region together constitute the thermal energy transition region.

[0140] Step S4 involves analyzing the relationship between the thermal energy transition region and the thermal energy region to obtain the thermal energy transition law. Step S4 includes the following sub-steps:

[0141] Step S401: Obtain historical infrared data of a normally operating heating furnace and analyze the midpoint of the boundary;

[0142] Step S401 includes the following sub-steps:

[0143] Step S4011: Historical infrared data refers to the historical data of thermal energy area, thermal energy transition area and high-quality image. When the heating temperature in the heating furnace changes, the gray value, thermal energy area and thermal energy transition area division of the heating furnace in the high-quality image will also change accordingly.

[0144] Step S4012: Draw the horizontal and vertical axes with the side length of a pixel in the high-quality image as a unit length to construct a two-dimensional coordinate system, named the boundary line confirmation map. Outline the thermal energy region and the thermal energy transition region in the high-quality image to obtain the region division map. Make the lower left vertex of the region division map coincide with the origin, and make the entire region division map within the first quadrant.

[0145] Step S4013: Name the rectangular square corresponding to the pixel in the region division map as the first square, name the first square adjacent to the thermal energy region and the thermal energy transition region as the second square, and remove the first square.

[0146] Step S4014: Name the intersection line between two adjacent second squares in the same row as the boundary intersection line, and mark the midpoint of the boundary intersection line as the boundary intersection point. The boundary intersection point includes the first boundary intersection point, the second boundary intersection point, the third boundary intersection point, and the fourth boundary intersection point, which correspond to the intersection points between the preheating area and the preheating heating transition area, the preheating heating transition area and the heating area, the heating area and the heating homogenization transition area, and the heating homogenization transition area and the homogenization area, respectively.

[0147] Step S4015: Perform linear regression on the first intersection point, the second intersection point, the third intersection point, and the fourth intersection point of the boundary respectively, and name the line segments obtained from the regression as the first boundary line segment, the second boundary line segment, the third boundary line segment, and the fourth boundary line segment respectively. Obtain the midpoints of the first boundary line segment, the second boundary line segment, the third boundary line segment, and the fourth boundary line segment respectively, and name them as the first boundary midpoint, the second boundary midpoint, the third boundary midpoint, and the fourth boundary midpoint.

[0148] Step S4016: The midpoints of the first boundary, the second boundary, the third boundary, and the fourth boundary are the boundary midpoints. Analyze each set of historical infrared data to obtain different boundary midpoints.

[0149] Step S402: Analyze the thermal energy transition law based on the midpoint of the boundary;

[0150] Step S402 includes the following sub-steps:

[0151] Step S4021: Obtain the heating temperature corresponding to the historical infrared data. With the heating temperature as the X-axis, establish four two-dimensional coordinate systems with the midpoints of the first, second, third, and fourth boundaries as the Y-axis. Name them as the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram, respectively. They are collectively named as boundary pattern diagrams.

[0152] Step S4022: Perform discrete regression analysis on the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram respectively to obtain the first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve. Name the first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve together as boundary pattern curves.

[0153] Step S4023: Increase the line width of the boundary law curve so that the boundary law curve exactly covers all the coordinate points in the boundary law diagram. The final boundary law curve is the thermal energy transition law, which includes the first transition law, the second transition law, the third transition law, and the fourth transition law.

[0154] Step S5: Determine whether the heating furnace is operating normally based on the thermal energy transition law; Step S5 includes the following sub-steps:

[0155] Step S501: Obtain the current heating temperature of the furnace, name it the real-time temperature, and label it as T;

[0156] Step S502: Analyze and calculate the current midpoints of the first boundary, second boundary, third boundary, and fourth boundary of the heating furnace, and label them as Q1, Q2, Q3, and Q4, respectively.

[0157] Step S503: Input the coordinates (T,Q1), (T,Q2), (T,Q3) and (T,Q4) into the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram and the fourth boundary pattern diagram respectively, and determine whether they are in the first transition pattern, the second transition pattern, the third transition pattern and the fourth transition pattern respectively. If they are, output a normal operation signal; otherwise, output an abnormal operation signal.

[0158] Example 3: This application provides an electronic device, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes a computer-readable instruction, it performs steps such as those in the infrared thermal imaging-based furnace operation monitoring method to achieve the following functions: setting up an infrared thermal imager to comprehensively monitor the furnace and acquire thermal imaging images; preprocessing the thermal imaging images to convert them into high-quality images; performing image analysis on the high-quality images to identify different thermal energy regions and thermal energy transition regions in the furnace; analyzing the relationship between thermal energy transition regions and thermal energy regions to obtain thermal energy transition patterns; and determining whether the furnace is operating normally based on the thermal energy transition patterns.

[0159] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0160] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-described method for monitoring the operation of a heating furnace based on infrared thermal imaging, to achieve the following functions: setting up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal imaging images; performing image preprocessing on the thermal imaging images to convert them into high-quality images; performing image analysis on the high-quality images to identify different thermal energy regions and thermal energy transition regions in the heating furnace; analyzing the relationship between the thermal energy transition regions and the thermal energy regions to obtain the thermal energy transition law; and determining whether the heating furnace is operating normally based on the thermal energy transition law.

[0161] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the technical solutions described above, or the parts that contribute to the prior art, can be embodied in the form of software products. These computer software products can be stored in computer-readable storage media, such as ROM / RAM, magnetic disks, optical disks, etc., and include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0162] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0163] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring the operation of a heating furnace based on infrared thermal imaging, characterized in that, Includes the following steps: An infrared thermal imager was installed to comprehensively monitor the heating furnace and acquire thermal imaging images; Perform image preprocessing on thermal imaging images to convert them into high-quality images; Image analysis is performed on high-quality images to identify different thermal energy regions and thermal energy transition regions in the heating furnace; By analyzing the relationship between the thermal energy transition region and the thermal energy region, the thermal energy transition law is obtained. Determine whether the heating furnace is operating normally based on the thermal energy transition law.

2. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 1, characterized in that, Setting up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal images includes the following sub-steps: The heating furnace includes a preheating section, a heating section, and a soaking section; An infrared thermal imager was set up to monitor the preheating section, heating section, and heat soaking section in a unified manner, and thermal images were obtained.

3. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 2, characterized in that, The image preprocessing of the thermal imaging image includes denoising, enhancement, and grayscale processing to obtain a high-quality image.

4. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 3, characterized in that, Image analysis of high-quality images to identify different thermal energy regions and thermal transition regions in the heating furnace includes the following sub-steps: The pixel in the nth row and mth column of the high-quality image is labeled as P(n,m), and the gray value of P(n,m) is labeled as H(n,m), where n and m are both positive integers and (n,m) is the index of P and H. When analyzing any H(n,m), we name it the center gray level and the H(n,m) adjacent to the center gray level are named the adjacent gray levels. Calculate the absolute value of the difference between each adjacent gray level and the center gray level, and name the calculation result as gray level difference. Calculate all gray level differences with each H(n,m) as the center gray level. Further analysis of the grayscale difference was conducted to identify different thermal energy zones and thermal energy transition zones in the heating furnace.

5. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 4, characterized in that, Further analysis of the grayscale difference and identification of different thermal energy zones and thermal energy transition zones in the heating furnace includes the following sub-steps: Construct a one-dimensional coordinate system with the gray-level difference as the X-axis, name it gray-level distribution map, and enter the gray-level difference values ​​into the gray-level distribution map; Cluster analysis was performed on the grayscale distribution image to obtain grayscale difference clusters. These grayscale difference clusters were then numbered and labeled with the symbol S. i This indicates that i is a positive integer and i is the index of S; The gray-scale difference clustering population is divided into two clustering types: non-transitional clustering and transitional clustering. The statistical S... i The number of grayscale differences in the data is named the difference quantity. The two grayscale difference clusters with the largest difference quantity are marked as the initial parent clusters. The initial parent cluster on the left is marked as the non-transitional parent cluster, the initial parent cluster on the right is marked as the transitional parent cluster, and the remaining grayscale difference clusters are marked as the initial sub-clusters. Calculate the distance between the initial sub-cluster and the initial parent cluster, named the cluster distance. The cluster distance is the difference between the nearest grayscale difference between the initial sub-cluster and the initial parent cluster. The cluster distance includes non-transition distance and transition distance. If the non-transition distance is less than the transition distance, the initial sub-cluster is incorporated into the non-transition mother cluster; otherwise, the initial sub-cluster is incorporated into the transition mother cluster. After all the initial sub-clusters have been divided, the non-transition mother cluster is the non-transition cluster, and the transition mother cluster is the transition cluster. The grayscale difference calculated with any H(n,m) as the center grayscale is named the judgment difference. Non-transition parameter and transition parameter are set. The non-transition parameter and transition parameter are initially 0. If the judgment difference is in the non-transition cluster, the non-transition parameter is incremented by one. If the judgment difference is in the transition cluster, the transition parameter is incremented by one. For each judgment difference, analyze it. After the analysis is completed, if the non-transition parameter is greater than the transition parameter, mark P(n,m) corresponding to the center gray level at this time as a non-transition pixel; otherwise, mark P(n,m) corresponding to the center gray level at this time as a transition pixel. A non-transition region is formed by consecutive adjacent non-transition pixels, and a transition region is formed by consecutive adjacent transition pixels. The average value of H(n,m) in the non-transition region is calculated and named the non-transition average gray level. The average value of H(n,m) in the transition region is calculated and named the transition average gray level. The non-transition region with the highest average gray value is marked as the preheating region, the non-transition region with the lowest average gray value is marked as the heating region, and the remaining non-transition regions are marked as the homogenization region. The transition region between the preheating region and the heating region is marked as the preheating-heating transition region, and the remaining transition regions are marked as the heating-homogenization transition region. The preheating region, the heating region, and the homogenization region constitute the thermal energy region, and the preheating-heating transition region and the heating-homogenization transition region constitute the thermal energy transition region.

6. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 5, characterized in that, Analyzing the relationship between the thermal energy transition regions and the thermal energy transition law yields the following sub-steps: Acquire historical infrared data of a normally operating heating furnace and analyze the midpoint of the boundary; The thermal energy transition law is analyzed based on the midpoint of the boundary.

7. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 6, characterized in that, Acquiring historical infrared data of a normally operating heating furnace and analyzing the boundary midpoints includes the following sub-steps: The historical infrared data refers to the historical data of thermal energy region, thermal energy transition region and high-quality image. When the heating temperature in the heating furnace changes, the gray value, thermal energy region and thermal energy transition region division of the heating furnace in the high-quality image will also change accordingly. Using the side length of a pixel in the high-quality image as a unit length, draw the horizontal and vertical axes to construct a two-dimensional coordinate system, named the boundary line confirmation map. Delineate the thermal energy region and the thermal energy transition region within the high-quality image to obtain the region division map. Align the lower left vertex of the region division map with the origin and ensure that the entire region division map is within the first quadrant. Name the rectangular square corresponding to the pixel in the region division map as the first square, name the first square adjacent to the thermal energy region and the thermal energy transition region as the second square, and remove the first square. The intersection line between two adjacent second squares in the same row is named the boundary intersection line, and the midpoint of the boundary intersection line is marked as the boundary intersection point. The boundary intersection point includes the first boundary intersection point, the second boundary intersection point, the third boundary intersection point, and the fourth boundary intersection point, which correspond to the intersection point between the preheating area and the preheating heating transition area, the intersection point between the preheating heating transition area and the heating area, the intersection point between the heating area and the heating homogenization transition area, and the intersection point between the heating homogenization transition area and the homogenization area, respectively. Linear regression was performed on the first, second, third, and fourth boundary intersection points, respectively. The resulting line segments were named the first boundary line segment, the second boundary line segment, the third boundary line segment, and the fourth boundary line segment, respectively. The midpoints of the first, second, third, and fourth boundary line segments were obtained and named the first boundary midpoint, the second boundary midpoint, the third boundary midpoint, and the fourth boundary midpoint, respectively. The midpoints of the first, second, third, and fourth boundaries are the boundary midpoints. Different boundary midpoints are obtained by analyzing each set of historical infrared data.

8. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 7, characterized in that, The analysis of thermal energy transition based on the midpoint of the boundary includes the following sub-steps: Obtain the heating temperature corresponding to the historical infrared data. With the heating temperature as the X-axis, establish four two-dimensional coordinate systems with the midpoints of the first, second, third, and fourth boundaries as the Y-axis. Name them the First Boundary Pattern Map, Second Boundary Pattern Map, Third Boundary Pattern Map, and Fourth Boundary Pattern Map respectively, and collectively name them Boundary Pattern Map. Discrete regression analysis was performed on the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram to obtain the first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve. The first boundary pattern curve, the second boundary pattern curve, the third boundary pattern curve, and the fourth boundary pattern curve were collectively named the boundary pattern curve. Increase the line width of the boundary law curve so that the boundary law curve exactly covers all the coordinate points in the boundary law diagram. The final boundary law curve is the thermal energy transition law, which includes the first transition law, the second transition law, the third transition law, and the fourth transition law.

9. The method for monitoring the operation of a heating furnace based on infrared thermal imaging according to claim 8, characterized in that, Determining whether a heating furnace is operating normally based on the thermal energy transition law includes the following sub-steps: Obtain the current heating temperature of the furnace, name it the real-time temperature, and label it T; The midpoints of the first, second, third, and fourth boundaries of the heating furnace are analyzed and calculated, and labeled as Q1, Q2, Q3, and Q4, respectively. Enter the coordinates (T,Q1), (T,Q2), (T,Q3), and (T,Q4) into the first boundary pattern diagram, the second boundary pattern diagram, the third boundary pattern diagram, and the fourth boundary pattern diagram, respectively. Determine whether they are in the first transition pattern, the second transition pattern, the third transition pattern, and the fourth transition pattern, respectively. If they are, output a normal operation signal; otherwise, output an abnormal operation signal.

10. A furnace operation monitoring system based on infrared thermal imaging, used to implement the furnace operation monitoring method based on infrared thermal imaging as described in any one of claims 1-9, characterized in that, It includes an infrared monitoring module, an image preprocessing module, a region recognition module, a transition pattern analysis module, and a running status judgment module; the infrared monitoring module, the image preprocessing module, the region recognition module, and the transition pattern analysis module are respectively connected to the running status judgment module for data transmission. The infrared monitoring module is used to set up an infrared thermal imager to comprehensively monitor the heating furnace and acquire thermal imaging images. The image preprocessing module is used to preprocess the thermal imaging image and convert it into a high-quality image. The region recognition module is used to perform image analysis on high-quality images and identify different thermal energy regions and thermal energy transition regions in the heating furnace. The transition law analysis module is used to analyze the relationship between the thermal energy transition region and the thermal energy region to obtain the thermal energy transition law; The operating status judgment module is used to determine whether the heating furnace is operating normally based on the thermal energy transition law.

Images