A method and system for early warning and maintenance of sampling tubes for differential pressure level gauges in coal mills.

Abstract

This invention relates to the field of coal mill technology and discloses a method and system for early warning of maintenance of a sampling tube for a differential pressure level gauge in a coal mill. The method includes: acquiring an image of the sampling tube to be inspected to obtain a real-time grayscale image of the sampling tube; calibrating the pixels on the real-time grayscale image and determining the feature information of the real-time grayscale image based on the calibrated pixels, wherein the feature information includes image texture features and image color features; determining the standard feature information of a standard grayscale image and calculating the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information; determining whether to issue an early warning based on the relationship between the similarity and the preset similarity; if so, performing maintenance operations on the sampling tube to be inspected. This method can provide real-time maintenance early warning for the sampling tube, ensuring the stable operation of the coal mill, improving operational safety and reliability, and avoiding impact on the automatic level control of the coal mill and the rapid response of the unit load.

Description

Technical Field

[0001] This invention relates to the field of coal mill technology, and more specifically, to a method and system for inspection and early warning of the sampling tube of the differential pressure level gauge in a coal mill. Background Technology

[0002] As one of the most important auxiliary equipment in coal-fired power plants, the material level in the mill cylinder is the most important detection and control parameter. The BBD type double-inlet double-outlet ball mill is a type of coal mill commonly used in most thermal power plants. The mainstream method for measuring, controlling and adjusting the material level in the mill cylinder of this type of coal mill is differential pressure detection. This method utilizes the principle of deep cylinder level measurement. Two sampling tubes filled with low-pressure, constant-flow compressed air are inserted into the upper and lower halves of the hollow cylinder of the mill, respectively. The concentration of coal powder and the thickness of the powder layer cause different back pressures on the two sampling tubes. A differential pressure transmitter is used to measure the differential pressure between the two tubes to reflect the material level inside the mill.

[0003] However, the sampling tube installed at the bottom of the hollow cylinder has only 5 centimeters of space between it and the auger directly above it. If the auger is poorly installed, the auger blades break, or especially if the auger bearing is damaged and the auger falls, the sampling tube will quickly be worn through and leaked by the auger during operation, causing inaccurate material level measurements. This will affect the automatic control of the coal mill's material level, the unit's rapid load response, and the operators' judgment of the material level. In severe cases, it can cause the coal mill to be full of coal, directly endangering the unit's operational safety. Therefore, how to determine whether the sampling tube is damaged and issue a maintenance warning is a technical problem that urgently needs to be solved. Summary of the Invention

[0004] This invention provides a method and system for early warning and maintenance of the sampling tube of the differential pressure level gauge in a coal mill. This system addresses the technical problem in the prior art where timely damage monitoring and early warning of the sampling tube are not possible, which leads to the sampling tube damage affecting the automatic control of the coal mill level and the rapid response of the unit load.

[0005] To achieve the above objectives, the present invention provides a method for early warning and maintenance of the sampling tube of a differential pressure level gauge in a coal mill, comprising:

[0006] Based on the pre-deployed image acquisition device, the sample tube to be tested is image acquired to obtain the real-time grayscale image of the sample tube to be tested;

[0007] The pixels on the real-time grayscale image are calibrated, and the feature information of the real-time grayscale image is determined based on the calibrated pixels, wherein the feature information includes image texture features and image color features;

[0008] Determine the standard feature information of the standard grayscale image, and calculate the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information;

[0009] Whether to issue an early warning is determined based on the relationship between the aforementioned similarity and the preset similarity. If so, the sampling tube to be tested is inspected and repaired.

[0010] Furthermore, when acquiring images of the sampling tube to be detected based on a pre-deployed image acquisition device to obtain a real-time image of the sampling tube to be detected, the process includes:

[0011] Based on preset angle requirements, the sampling tube to be tested is photographed from multiple angles to obtain multiple images from different angles;

[0012] Analyze all images to determine the overlapping area between every two images;

[0013] The stitching position is determined based on all overlapping areas, and all images are stitched together to obtain a real-time image of the sampling tube to be detected.

[0014] The real-time image is converted to grayscale to obtain the real-time grayscale image of the sampling tube to be detected.

[0015] Furthermore, when calibrating the pixels on the real-time grayscale image and determining the feature information of the real-time grayscale image based on the calibrated pixels, the process includes:

[0016] The overall detail of the real-time grayscale image is calculated based on the calibrated pixels;

[0017] Whether image enhancement processing is needed for the real-time grayscale image is determined based on the relationship between the overall detail level and the detail threshold.

[0018] When the overall detail is greater than or equal to the detail threshold, it is determined that image enhancement processing is not required for the real-time grayscale image.

[0019] When the overall detail is less than the detail threshold, it is determined that image enhancement processing needs to be performed on the real-time grayscale image to obtain the image enhancement information of the real-time grayscale image;

[0020] The image enhancement information is enhanced and adjusted according to the overall detail level and the detail threshold to obtain an enhanced real-time grayscale image, and the feature information is determined.

[0021] Furthermore, when calculating the overall detail of the real-time grayscale image based on the calibrated pixels, the following steps are included:

[0022] Extract all pixels and establish a coordinate system;

[0023] Determine the coordinates and grayscale values ​​of all pixels, and calculate the detail of the corresponding pixels based on the coordinates and grayscale values;

[0024] The detail level of a pixel is calculated using the following formula:

[0025]

[0026] Where A represents the detail level of a pixel, I(x, y) represents the grayscale value of the pixel at coordinates (x, y), and u(x, y) represents the expected value of the pixel at coordinates (x, y).

[0027] Construct a set of details based on all the levels of detail;

[0028] Obtain a pre-set detail threshold and classify details below the detail threshold into a low detail set;

[0029] Detail levels greater than or equal to the specified detail threshold are classified into the high detail set;

[0030] The overall detail of the real-time grayscale image is calculated based on the low detail set and the high detail set.

[0031] Furthermore, the overall detail of the real-time grayscale image is calculated according to the following formula:

[0032]

[0033] Where K is the overall detail of the real-time grayscale image, b1 is the calculation coefficient corresponding to the low detail set, n is the number of details in the low detail set, ΔS is the detail threshold, dc is the c-th detail in the low detail set, b2 is the calculation coefficient corresponding to the high detail set, m is the number of details in the high detail set, and fe is the e-th detail in the high detail set.

[0034] Further, when enhancing and adjusting the image enhancement information according to the overall detail and the detail threshold to obtain the enhanced real-time grayscale image, the process includes:

[0035] Based on the image enhancement information, construct an image enhancement information sequence H, where H = {h1, h2, h3, ..., hp}, and p is the number of image enhancement information.

[0036] Calculate the difference in detail between the overall detail level and the detail level threshold;

[0037] A set of detail difference values ​​is preset, wherein the set of detail difference values ​​includes a first preset detail difference value and a second preset detail difference value;

[0038] A set of enhancement coefficients is preset, wherein the set of enhancement coefficients includes a first preset enhancement coefficient r1, a second preset enhancement coefficient r2 and a third preset enhancement coefficient r3, and 1 < r1 < r2 < r3 < 1.25;

[0039] Based on the relationship between the detail difference and the set of detail differences, a corresponding preset enhancement coefficient is selected from the set of enhancement coefficients;

[0040] When the detail difference is less than the first preset detail difference, the first preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

[0041] When the detail difference is greater than or equal to the first preset detail difference and less than the second preset detail difference, the second preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

[0042] When the detail difference is greater than or equal to the second preset detail difference, the third preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

[0043] Further, when determining the standard feature information of the standard grayscale image and calculating the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information, the process includes:

[0044] Numerical extraction is performed on the image texture features and image color features of the real-time grayscale image to obtain the corresponding binary sequences of texture features and color features;

[0045] The texture feature binary sequence and the color feature binary sequence are combined respectively to obtain a high-dimensional texture feature vector and a high-dimensional color feature vector;

[0046] The high-dimensional texture feature vector and the high-dimensional color feature vector are subjected to dimensionality reduction processing to obtain texture feature values ​​and color feature values;

[0047] Obtain the standard texture feature value and standard color feature value corresponding to the standard grayscale image;

[0048] The similarity between the real-time grayscale image and the standard grayscale image is calculated based on the texture feature value, color feature value, standard texture feature value, and standard color feature value.

[0049] W=|G1-G2|×t1+|V1-V2|×t2;

[0050] Where W is the similarity between the real-time grayscale image and the standard grayscale image, G1 is the texture feature value, G2 is the standard texture feature value, t1 is the weight corresponding to the texture feature, V1 is the color feature value, V2 is the standard color feature value, and t2 is the weight corresponding to the color feature.

[0051] Furthermore, when determining whether to issue a warning based on the relationship between the stated similarity and a preset similarity, the process includes:

[0052] If the similarity is less than the preset similarity, a warning is issued.

[0053] If the similarity is greater than or equal to the preset similarity, then it is determined that no warning will be issued.

[0054] To achieve the above objectives, the present invention also provides a maintenance and early warning system for the sampling tube of a coal mill differential pressure level gauge, comprising:

[0055] The first module is used to acquire images of the sampling tube to be tested based on a pre-deployed image acquisition device, and obtain a real-time grayscale image of the sampling tube to be tested;

[0056] The second module is used to calibrate the pixels on the real-time grayscale image and determine the feature information of the real-time grayscale image based on the calibrated pixels, wherein the feature information includes image texture features and image color features.

[0057] The third module is used to determine the standard feature information of the standard grayscale image, and calculate the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information;

[0058] The fourth module is used to determine whether to issue an early warning based on the relationship between the similarity and the preset similarity. If so, the sampling tube to be tested is inspected.

[0059] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0060] This invention discloses a maintenance early warning method and system for a sampling tube of a differential pressure level gauge in a coal mill. The method includes: acquiring an image of the sampling tube to be inspected to obtain a real-time grayscale image of the sampling tube; calibrating the pixels on the real-time grayscale image and determining the feature information of the real-time grayscale image based on the calibrated pixels, wherein the feature information includes image texture features and image color features; determining the standard feature information of a standard grayscale image; calculating the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information; determining whether to issue an early warning based on the relationship between the similarity and a preset similarity; if so, performing maintenance operations on the sampling tube to be inspected. This method can provide real-time maintenance early warning for the sampling tube, ensuring the stable operation of the coal mill, improving operational safety and reliability, and avoiding impacts on the automatic level control of the coal mill and the rapid response of the unit load. Attached Figure Description

[0061] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0062] Figure 1 A flowchart illustrating a maintenance and early warning method for a sampling tube of a coal mill differential pressure level gauge according to an embodiment of the present invention is shown.

[0063] Figure 2 A schematic diagram of the maintenance and early warning system for a coal mill differential pressure level gauge sampling tube is shown in an embodiment of the present invention. Detailed Implementation

[0064] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0065] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0066] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0067] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0068] The following is a description of preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0069] like Figure 1 As shown, an embodiment of the present invention discloses a maintenance and early warning method for a sampling tube of a differential pressure level gauge in a coal mill, comprising:

[0070] S110: Based on the pre-deployed image acquisition device, an image of the sampling tube to be detected is acquired to obtain a real-time grayscale image of the sampling tube to be detected;

[0071] In some embodiments of this application, when acquiring images of the sampling tube to be detected based on a pre-deployed image acquisition device to obtain a real-time image of the sampling tube to be detected, the process includes:

[0072] Based on preset angle requirements, the sampling tube to be tested is photographed from multiple angles to obtain multiple images from different angles;

[0073] Analyze all images to determine the overlapping area between every two images;

[0074] The stitching position is determined based on all overlapping areas, and all images are stitched together to obtain a real-time image of the sampling tube to be detected.

[0075] The real-time image is converted to grayscale to obtain the real-time grayscale image of the sampling tube to be detected.

[0076] In this embodiment, the image acquisition device can be a CCD camera.

[0077] In this embodiment, the angle requirement can be up, down, front, back, left, or right.

[0078] In this embodiment, the grayscale conversion method for images is complex and mature, and will not be described in detail here.

[0079] The beneficial effects of the above technical solution are: by determining the real-time grayscale image of the sampling tube to be detected, the present invention can lay the foundation for early warning of the sampling tube.

[0080] S120: The pixels on the real-time grayscale image are calibrated, and the feature information of the real-time grayscale image is determined based on the calibrated pixels, wherein the feature information includes image texture features and image color features;

[0081] In some embodiments of this application, when calibrating the pixels on the real-time grayscale image and determining the feature information of the real-time grayscale image based on the calibrated pixels, the process includes:

[0082] The overall detail of the real-time grayscale image is calculated based on the calibrated pixels;

[0083] Whether image enhancement processing is needed for the real-time grayscale image is determined based on the relationship between the overall detail level and the detail threshold.

[0084] When the overall detail is greater than or equal to the detail threshold, it is determined that image enhancement processing is not required for the real-time grayscale image.

[0085] When the overall detail is less than the detail threshold, it is determined that image enhancement processing needs to be performed on the real-time grayscale image to obtain the image enhancement information of the real-time grayscale image;

[0086] The image enhancement information is enhanced and adjusted according to the overall detail level and the detail threshold to obtain an enhanced real-time grayscale image, and the feature information is determined.

[0087] In this embodiment, overall detail is a numerical value used to measure whether an image is clear.

[0088] In this embodiment, the detail threshold can be set according to the actual situation.

[0089] The beneficial effects of the above technical solution are: the present invention determines whether image enhancement processing of real-time grayscale images is needed based on the relationship between overall detail and detail threshold, which can ensure the accuracy of feature information acquisition and avoid image blurring and errors.

[0090] In some embodiments of this application, calculating the overall detail of the real-time grayscale image based on calibrated pixels includes:

[0091] Extract all pixels and establish a coordinate system;

[0092] Determine the coordinates and grayscale values ​​of all pixels, and calculate the detail of the corresponding pixels based on the coordinates and grayscale values;

[0093] The detail level of a pixel is calculated using the following formula:

[0094]

[0095] Where A represents the detail level of a pixel, I(x, y) represents the grayscale value of the pixel at coordinates (x, y), and u(x, y) represents the expected value of the pixel at coordinates (x, y).

[0096]

[0097] Construct a set of details based on all the levels of detail;

[0098] Obtain a pre-set detail threshold and classify details below the detail threshold into a low detail set;

[0099] Detail levels greater than or equal to the specified detail threshold are classified into the high detail set;

[0100] The overall detail of the real-time grayscale image is calculated based on the low detail set and the high detail set.

[0101] In this embodiment, each pixel has corresponding coordinates and grayscale value.

[0102] In this embodiment, the detail set includes a low detail set and a high detail set.

[0103] The beneficial effects of the above technical solution are: the present invention calculates the overall detail of a real-time grayscale image based on the low detail set and the high detail set, ensuring the calculation efficiency and accuracy of the overall detail, avoiding errors caused by manual judgment, and providing a basis for judging whether the image needs to be enhanced.

[0104] In some embodiments of this application, the overall detail of the real-time grayscale image is calculated according to the following formula:

[0105]

[0106] Where K is the overall detail of the real-time grayscale image, b1 is the calculation coefficient corresponding to the low detail set, n is the number of details in the low detail set, ΔS is the detail threshold, dc is the c-th detail in the low detail set, b2 is the calculation coefficient corresponding to the high detail set, m is the number of details in the high detail set, and fe is the e-th detail in the high detail set.

[0107] In some embodiments of this application, when enhancing and adjusting the image enhancement information according to the overall detail and the detail threshold to obtain an enhanced real-time grayscale image, the following steps are included:

[0108] Based on the image enhancement information, construct an image enhancement information sequence H, where H = {h1, h2, h3, ..., hp}, and p is the number of image enhancement information.

[0109] Calculate the difference in detail between the overall detail level and the detail level threshold;

[0110] A set of detail difference values ​​is preset, wherein the set of detail difference values ​​includes a first preset detail difference value and a second preset detail difference value;

[0111] A set of enhancement coefficients is preset, wherein the set of enhancement coefficients includes a first preset enhancement coefficient r1, a second preset enhancement coefficient r2 and a third preset enhancement coefficient r3, and 1 < r1 < r2 < r3 < 1.25;

[0112] Based on the relationship between the detail difference and the set of detail differences, a corresponding preset enhancement coefficient is selected from the set of enhancement coefficients;

[0113] When the detail difference is less than the first preset detail difference, the first preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

[0114] When the detail difference is greater than or equal to the first preset detail difference and less than the second preset detail difference, the second preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

[0115] When the detail difference is greater than or equal to the second preset detail difference, the third preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

[0116] In this embodiment, the image enhancement information includes hue value, color saturation value, brightness value, etc.

[0117] In this embodiment, the first preset detail difference is less than the second preset detail difference.

[0118] In this embodiment, the product of the preset enhancement coefficient and each image enhancement information in the image enhancement information sequence H is calculated, and the enhanced real-time grayscale image is obtained based on the obtained product value.

[0119] The beneficial effects of the above technical solution are: the present invention selects the corresponding preset enhancement coefficient from the enhancement coefficient set according to the relationship between the detail difference and the detail difference set, which can ensure the accuracy of enhancement processing and correction processing, avoid the occurrence of color cast, and avoid over-enhancement or under-enhancement.

[0120] S130: Determine the standard feature information of the standard grayscale image, and calculate the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information;

[0121] In some embodiments of this application, determining standard feature information of a standard grayscale image and calculating the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information includes:

[0122] Numerical extraction is performed on the image texture features and image color features of the real-time grayscale image to obtain the corresponding binary sequences of texture features and color features;

[0123] The texture feature binary sequence and the color feature binary sequence are combined respectively to obtain a high-dimensional texture feature vector and a high-dimensional color feature vector;

[0124] The high-dimensional texture feature vector and the high-dimensional color feature vector are subjected to dimensionality reduction processing to obtain texture feature values ​​and color feature values;

[0125] Obtain the standard texture feature value and standard color feature value corresponding to the standard grayscale image;

[0126] The similarity between the real-time grayscale image and the standard grayscale image is calculated based on the texture feature value, color feature value, standard texture feature value, and standard color feature value.

[0127] W=|G1-G2|×t1+|V1-V2|×t2;

[0128] Where W is the similarity between the real-time grayscale image and the standard grayscale image, G1 is the texture feature value, G2 is the standard texture feature value, t1 is the weight corresponding to the texture feature, V1 is the color feature value, V2 is the standard color feature value, and t2 is the weight corresponding to the color feature.

[0129] In this embodiment, color features are global features that describe the surface properties of objects corresponding to an image or image region. These features exhibit good stability to changes in image scale, orientation, and viewpoint, making them excellent for tasks such as image recognition and classification.

[0130] Texture features: Texture features are also a type of global feature, describing the surface properties of objects corresponding to an image or image region. Unlike color features, texture features focus more on describing the structural information of an object's surface, such as roughness, directionality, and repetition patterns.

[0131] The beneficial effects of the above technical solution are: the present invention calculates the similarity between a real-time grayscale image and a standard grayscale image based on texture feature values, color feature values, standard texture feature values ​​and standard color feature values, providing reliable data support for whether the sampling tube needs an early warning.

[0132] S140: Determine whether to issue an early warning based on the relationship between the similarity and the preset similarity. If so, perform maintenance operations on the sampling tube to be tested.

[0133] In some embodiments of this application, determining whether to issue a warning based on the relationship between the similarity and a preset similarity includes:

[0134] If the similarity is less than the preset similarity, a warning is issued.

[0135] If the similarity is greater than or equal to the preset similarity, then it is determined that no warning will be issued.

[0136] The beneficial effects of the above technical solution are: the present invention can accurately determine whether the sampling tube is damaged and whether it needs to be repaired or replaced, thus ensuring the normal operation of the coal mill and avoiding affecting the stable operation of the unit.

[0137] To further illustrate the technical concept of this invention, the technical solution of this invention will now be described in conjunction with specific application scenarios.

[0138] Correspondingly, such as Figure 2 As shown, this application also provides a maintenance early warning system for the sampling tube of the differential pressure level gauge in a coal mill, comprising:

[0139] The first module is used to acquire images of the sampling tube to be tested based on a pre-deployed image acquisition device, and obtain a real-time grayscale image of the sampling tube to be tested;

[0140] The second module is used to calibrate the pixels on the real-time grayscale image and determine the feature information of the real-time grayscale image based on the calibrated pixels, wherein the feature information includes image texture features and image color features.

[0141] The third module is used to determine the standard feature information of the standard grayscale image, and calculate the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information;

[0142] The fourth module is used to determine whether to issue an early warning based on the relationship between the similarity and the preset similarity. If so, the sampling tube to be tested is inspected.

[0143] In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0144] Although the invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in this invention can be combined with each other in any way. The fact that not all of these combinations are described in this specification is merely for the sake of brevity and resource conservation.

[0145] It will be understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for inspection and early warning of the sampling tube of a differential pressure level gauge in a coal mill, characterized in that, include: Based on the pre-deployed image acquisition device, the sample tube to be tested is image acquired to obtain the real-time grayscale image of the sample tube to be tested; The pixels on the real-time grayscale image are calibrated, and the feature information of the real-time grayscale image is determined based on the calibrated pixels, wherein the feature information includes image texture features and image color features; Determine the standard feature information of the standard grayscale image, and calculate the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information; Whether to issue an early warning is determined based on the relationship between the similarity and the preset similarity. If so, the sampling tube to be tested is inspected and repaired. The process of calibrating the pixels in the real-time grayscale image and determining the feature information of the real-time grayscale image based on the calibrated pixels includes: The overall detail of the real-time grayscale image is calculated based on the calibrated pixels; Whether image enhancement processing is needed for the real-time grayscale image is determined based on the relationship between the overall detail level and the detail threshold. When the overall detail is greater than or equal to the detail threshold, it is determined that image enhancement processing is not required for the real-time grayscale image. When the overall detail is less than the detail threshold, it is determined that image enhancement processing needs to be performed on the real-time grayscale image to obtain the image enhancement information of the real-time grayscale image; The image enhancement information is enhanced and adjusted according to the overall detail level and the detail level threshold to obtain an enhanced real-time grayscale image, and the feature information is determined. Calculating the overall detail of the real-time grayscale image based on calibrated pixels includes: Extract all pixels and establish a coordinate system; Determine the coordinates and grayscale values ​​of all pixels, and calculate the detail of the corresponding pixels based on the coordinates and grayscale values; The detail level of a pixel is calculated using the following formula: ； Where A represents the detail level of a pixel, I(x, y) represents the grayscale value of the pixel at coordinates (x, y), and u(x, y) represents the expected value of the pixel at coordinates (x, y). ; Construct a set of details based on all the levels of detail; Obtain a pre-set detail threshold and classify details below the detail threshold into a low detail set; Detail levels greater than or equal to the specified detail threshold are classified into the high detail set; The overall detail level of the real-time grayscale image is calculated based on the low detail set and the high detail set; The overall detail level of the real-time grayscale image is calculated according to the following formula: ； Where K is the overall detail of the real-time grayscale image, b1 is the calculation coefficient corresponding to the low detail set, n is the number of details in the low detail set, ΔS is the detail threshold, dc is the c-th detail in the low detail set, b2 is the calculation coefficient corresponding to the high detail set, m is the number of details in the high detail set, and fe is the e-th detail in the high detail set.

2. The maintenance and early warning method for the sampling tube of the differential pressure level gauge in a coal mill according to claim 1, characterized in that, When acquiring images of the sampling tube to be inspected based on a pre-deployed image acquisition device to obtain a real-time image of the sampling tube to be inspected, the process includes: Based on preset angle requirements, the sampling tube to be tested is photographed from multiple angles to obtain multiple images from different angles; Analyze all images to determine the overlapping area between every two images; The stitching position is determined based on all overlapping areas, and all images are stitched together to obtain a real-time image of the sampling tube to be detected. The real-time image is converted to grayscale to obtain the real-time grayscale image of the sampling tube to be detected.

3. The maintenance and early warning method for the sampling tube of the differential pressure level gauge in a coal mill according to claim 1, characterized in that, When enhancing and adjusting the image enhancement information according to the overall detail and the detail threshold to obtain an enhanced real-time grayscale image, the process includes: Based on the image enhancement information, construct an image enhancement information sequence H, where H = {h1, h2, h3, ..., hp}, and p is the number of image enhancement information. Calculate the difference in detail between the overall detail level and the detail level threshold; A set of detail difference values ​​is preset, wherein the set of detail difference values ​​includes a first preset detail difference value and a second preset detail difference value; A set of enhancement coefficients is preset, wherein the set of enhancement coefficients includes a first preset enhancement coefficient r1, a second preset enhancement coefficient r2 and a third preset enhancement coefficient r3, and 1 < r1 < r2 < r3 < 1.25; Based on the relationship between the detail difference and the set of detail differences, a corresponding preset enhancement coefficient is selected from the set of enhancement coefficients; When the detail difference is less than the first preset detail difference, the first preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H. When the detail difference is greater than or equal to the first preset detail difference and less than the second preset detail difference, the second preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H. When the detail difference is greater than or equal to the second preset detail difference, the third preset enhancement coefficient is selected to enhance and adjust the image enhancement information sequence H.

4. The maintenance and early warning method for the sampling tube of the differential pressure level gauge in a coal mill according to claim 1, characterized in that, When determining the standard feature information of a standard grayscale image, and calculating the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information, the process includes: Numerical extraction is performed on the image texture features and image color features of the real-time grayscale image to obtain the corresponding binary sequences of texture features and color features; The texture feature binary sequence and the color feature binary sequence are combined respectively to obtain a high-dimensional texture feature vector and a high-dimensional color feature vector; The high-dimensional texture feature vector and the high-dimensional color feature vector are subjected to dimensionality reduction processing to obtain texture feature values ​​and color feature values; Obtain the standard texture feature value and standard color feature value corresponding to the standard grayscale image; The similarity between the real-time grayscale image and the standard grayscale image is calculated based on the texture feature value, color feature value, standard texture feature value, and standard color feature value. ； Where W is the similarity between the real-time grayscale image and the standard grayscale image, G1 is the texture feature value, G2 is the standard texture feature value, t1 is the weight corresponding to the texture feature, V1 is the color feature value, V2 is the standard color feature value, and t2 is the weight corresponding to the color feature.

5. The maintenance and early warning method for the sampling tube of the differential pressure level gauge in a coal mill according to claim 1, characterized in that, When determining whether to issue a warning based on the relationship between the aforementioned similarity and a preset similarity, the following steps are included: If the similarity is less than the preset similarity, a warning is issued. If the similarity is greater than or equal to the preset similarity, then it is determined that no warning will be issued.

6. A maintenance early warning system for a coal mill differential pressure level gauge sampling tube, applied to the maintenance early warning method for the coal mill differential pressure level gauge sampling tube as described in any one of claims 1-5, characterized in that, include: The first module is used to acquire images of the sampling tube to be tested based on a pre-deployed image acquisition device, and obtain a real-time grayscale image of the sampling tube to be tested; The second module is used to calibrate the pixels on the real-time grayscale image and determine the feature information of the real-time grayscale image based on the calibrated pixels, wherein the feature information includes image texture features and image color features. The third module is used to determine the standard feature information of the standard grayscale image, and calculate the similarity between the real-time grayscale image and the standard grayscale image based on the feature information and the standard feature information; The fourth module is used to determine whether to issue an early warning based on the relationship between the similarity and the preset similarity. If so, the sampling tube to be tested is inspected.

Images