Method and system for detecting mesh of drum screen of nuclear power plant and computer equipment

Abstract

The present application relates to a kind of nuclear power plant drum net mesh detection method, system and computer equipment, the nuclear power plant drum net mesh detection method includes: when the drum net rotates, the image of the mesh is fully covered by array camera Image acquisition is carried out, to obtain the multiple local images of the mesh in axial and circumferential direction respectively;Respectively to each local image is carried out image processing, to identify the mesh hole profile in corresponding local image, and according to the mesh hole profile in corresponding local image Mark the position of corresponding mesh hole;The multiple local images are registered and fused in axial and circumferential direction respectively, to obtain the splicing image of the mesh;The area of each mesh hole in the splicing image is calculated respectively, and whether the mesh hole of the mesh is abnormal is judged according to the calculated area.The technical scheme of the present application reduces the labor intensity and operation risk of personnel, and the accuracy is higher, and the safe operation of nuclear power plant cold source equipment is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of nuclear power, and in particular to a method, system, and computer equipment for detecting the mesh sheets of a nuclear power plant's bulging mesh. Background Technology

[0002] The drum filter (also known as a drum-shaped filter) is a key piece of equipment for filtering cooling water in nuclear power plants. Its surface is covered with several mesh panels to remove impurities, marine organisms, and other foreign matter from the water source. (Reference) Figure 1 , Figure 2 and Figure 3 The drum screen 100 consists of a rotating cylindrical skeleton structure with mesh panels 101 installed around its circumference. Water enters the drum screen 100 from both end faces, is filtered through the mesh openings 1011 on the mesh panels 101, and then flows out of the drum screen 100. When foreign objects in the water source clog the mesh panels 101, it can lead to a reduction in the cooling water flow rate of the nuclear power plant. When the cooling water flow rate drops to a safe limit, it can trigger an emergency shutdown of the nuclear power plant, thus affecting its safe operation. Furthermore, the mesh panels 101 typically operate in seawater with high salinity, and are susceptible to structural corrosion and damage due to water flow, which can affect the filtration capacity of the drum screen and allow oversized foreign objects in the water source to enter the cooling water pipes, threatening the safe operation of critical cooling equipment in the nuclear power plant. Therefore, regular inspections of the mesh panels of the nuclear power plant's drum screen are necessary to ensure that the mesh panels 101 are not clogged or structurally damaged.

[0003] To ensure that the drum filter function meets the safety requirements of nuclear power plants, corrosion status and cooling water flow monitoring devices are typically installed on the drum filters. The corrosion status of the filters is assessed using cathodic protection and potential measurement methods. However, this indirect detection method cannot directly detect the actual state of clogging and corrosion damage. Therefore, regular manual visual inspection of the mesh openings is necessary. Due to the large size of the drum filter equipment, the astonishing number of mesh openings, the humid and dimly lit environment, and the fact that it involves working at height, manual inspection carries high safety risks. It is not only time-consuming and labor-intensive, but the quality of the inspection is also directly related to the experience and physical strength of the personnel, making it highly susceptible to missed inspections and personal injury. Summary of the Invention

[0004] To address the problems of high workload and easy omissions in manual inspection of mesh panels in existing technologies, this invention provides a method, system, and computer equipment for inspecting mesh panels of nuclear power plant blasting mesh.

[0005] The technical solution adopted by this invention to solve its technical problem is: to provide a method for detecting the mesh of a nuclear power plant's bulging mesh, comprising:

[0006] Step 10: When the drum mesh rotates, the array camera performs full-coverage image acquisition of the mesh to obtain multiple local images of the mesh in the axial and circumferential directions respectively;

[0007] Step 20: Perform image processing on each local image to identify the mesh outline in the corresponding local image, and mark the position of the corresponding mesh in the corresponding local image according to the mesh outline;

[0008] Step 30: Register and fuse the multiple local images in the axial and circumferential directions respectively to obtain the stitched image of the mesh;

[0009] Step 40: Calculate the area of ​​each mesh in the stitched image, and determine whether each mesh of the mesh is abnormal based on the calculated area.

[0010] Preferably, step S20 includes:

[0011] Step S21: Perform binarization processing on each local image;

[0012] Step S22: For each binarized local image, the gradient discrimination algorithm is used to determine the mesh boundary points;

[0013] Step S23: For each mesh boundary point, calculate its distance value with the adjacent mesh boundary points, and determine whether the mesh boundary point is a pixel point on the mesh outline set based on the distance value.

[0014] Step S24: Determine the mesh outline based on each pixel in the mesh outline set, and mark the position of the corresponding mesh with a preset marker graphic.

[0015] Preferably, step S30 includes:

[0016] By using feature matching, the multiple local images are registered and fused in the axial and circumferential directions respectively to obtain the stitched image of the mesh.

[0017] Preferably, step S30 includes:

[0018] Step S31: Obtain the template image of the mesh;

[0019] Step S32: Determine the spatial position of each local image based on the marked position of the mesh in each local image and the template image;

[0020] Step S33: Based on the spatial position of each local image and the rotation speed of the drum mesh, register the multiple local images in the axial and circumferential directions respectively;

[0021] Step S34: Remove overlapping regions from the registered local images and fuse the images after removing overlapping regions to obtain the stitched image of the mesh.

[0022] Preferably, after step S34, the method further includes:

[0023] Calculate the difference between the center coordinates of each mesh in the template image and the center point of the corresponding mesh mark in the stitched image, and determine the spatial position of each mesh in the stitched image based on the difference.

[0024] Preferably, between step S30 and step S40, the method further includes:

[0025] Based on the template image, the fused region in the stitched image is corrected.

[0026] Preferably, step S40 includes:

[0027] Step S41: Calculate the area of ​​each mesh in the stitched image, and subtract the calculated area from the standard area of ​​the mesh in the template image to obtain the area difference.

[0028] Step S42: Determine whether the area difference is greater than a first set value or less than a second set value, wherein the first set value is greater than zero and the second set value is less than zero;

[0029] Step S43: If the area difference is greater than the first set value, then it is determined that the corresponding mesh is corroded.

[0030] Step S44: If the area difference is less than the second set value, it is determined that the corresponding mesh is blocked;

[0031] Step S45: If the area difference is not greater than the first set value and not less than the second set value, then the corresponding mesh is determined to be normal.

[0032] Preferably, it further includes:

[0033] The anomaly detection results of each mesh hole of the mesh are sent to the remote monitoring module.

[0034] The present invention also provides a computer device, including a processor and a memory storing a computer program, wherein the processor, when executing the computer program, implements the steps of the above-described method for detecting the mesh of a nuclear power plant's blower screen.

[0035] This invention also constructs a mesh inspection system for bulging wire mesh in nuclear power plants, comprising:

[0036] A light source array is disposed on one side of the mesh;

[0037] A camera array is positioned on the other side of the mesh.

[0038] The computer equipment described above.

[0039] The technical solution of this invention utilizes the light-transmitting characteristics of the mesh openings in a mesh sheet. An array camera is used to capture images of the mesh sheet with full coverage. Then, the acquired images are processed to identify and mark the positions of the mesh openings. Multiple acquired partial images are then registered and fused. Finally, the area of ​​the pixels formed by the light transmission through the mesh openings is used to determine whether the mesh openings are abnormal, such as blocked or corroded defects. This detection method enables automated inspection of drum-type mesh sheets. Compared to manual inspection, it reduces labor intensity and operational risks, and offers higher accuracy, ensuring the safe operation of cold source equipment in nuclear power plants. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a schematic diagram of the structure of the heat exchanger in a nuclear power plant;

[0042] Figure 2 This is a schematic diagram of the AA-direction cross-section of the drum screen of a nuclear power plant;

[0043] Figure 3 This is a schematic diagram of the structure of the mesh on the drum screen of a nuclear power plant;

[0044] Figure 4 This is a flowchart of Embodiment 1 of the method for detecting the mesh of a nuclear power plant's bulging screen according to the present invention;

[0045] Figure 5 This is a partial structural schematic diagram of an embodiment of the nuclear power plant mesh inspection system of the present invention;

[0046] Figure 6 This is a logical structure diagram of Embodiment 2 of the nuclear power plant mesh detection system of the present invention. Detailed Implementation

[0047] 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.

[0048] Figure 4This is a flowchart of a first embodiment of the method for inspecting the mesh of a nuclear power plant mesh according to the present invention. The mesh inspection method of this embodiment includes the following steps:

[0049] Step 10: When the drum mesh rotates, the array camera performs full-coverage image acquisition of the mesh to obtain multiple local images of the mesh in the axial and circumferential directions respectively;

[0050] In this step, an array of light sources and array cameras arranged on opposite sides can be used to acquire images of the drum screen. Specifically, combined with... Figure 5 Based on the dimensions of the drum mesh and the mesh sheet, an array of multiple cameras and light sources arranged linearly is used to acquire images of the mesh sheet 101 covering the drum mesh along the axial direction. Utilizing the rotational movement of the drum mesh, images of the mesh sheet 101 covering the drum mesh are acquired periodically and synchronously, thus achieving full coverage acquisition of the images of the mesh sheet 101. Furthermore, as... Figure 5 As shown, to avoid the influence of the rotating drum mesh components, camera, and light source bracket installation, the light source and camera array are evenly divided into two parts, and the array camera 11 and array light source 12 are arranged on opposite sides. That is, the array camera 11 and array light source 12 are respectively set on both sides of the mesh 101. For example, the light source array 12 is suspended and installed on the wall of the factory roof, and the camera array 11 is symmetrically installed on the walls on both sides of the factory using cantilever brackets. Of course, in other embodiments, the arrangement positions of the array camera 11 and array light source 12 can be interchanged, as long as the camera and light source are not on the same side.

[0051] Step 20: Perform image processing on each local image to identify the mesh outline in the corresponding local image, and mark the position of the corresponding mesh in the corresponding local image according to the mesh outline;

[0052] Step 30: Register and fuse the multiple local images in the axial and circumferential directions respectively to obtain the stitched image of the mesh;

[0053] Step 40: Calculate the area of ​​each mesh in the stitched image, and determine whether each mesh of the mesh is abnormal based on the calculated area.

[0054] This embodiment utilizes the light-transmitting characteristics of the mesh openings in the drum mesh to acquire full-coverage images of the mesh using an array camera. The acquired images are then processed to identify and mark the positions of the mesh openings. Multiple acquired local images are then registered and fused. Finally, the area of ​​the pixels formed by the light transmission through the mesh openings is used to determine whether the mesh openings are abnormal, such as blockages or corrosion. This detection method enables automated inspection of drum mesh, reducing labor intensity and operational risks compared to manual inspection. Furthermore, it offers higher accuracy, ensuring the safe operation of the cold source equipment in nuclear power plants.

[0055] Further, in an optional embodiment, step S20 includes:

[0056] Step S21: Perform binarization processing on each local image;

[0057] In this step, image binarization is the process of setting the grayscale value of each pixel in the image to 0 or 255, which gives the entire image a distinct black and white effect.

[0058] Step S22: For each binarized local image, the gradient discrimination algorithm is used to determine the mesh boundary points;

[0059] In this step, after binarizing the local image, the gradient value of the pixel can be calculated, and then the two-dimensional gradient value can be converted into a one-dimensional absolute gradient value. Then, all pixels are traversed to search for pixels whose absolute gradient value is greater than the threshold (a positive value that is zero or close to zero), and these pixels are used as mesh boundary points.

[0060] Step S23: For each mesh boundary point, calculate its distance value with the adjacent mesh boundary points, and determine whether the mesh boundary point is a pixel point on the mesh outline set based on the distance value.

[0061] In this step, the distance between two adjacent mesh boundary points is calculated by searching, and the value of the distance (close to 1) is used to determine whether it is a pixel on the mesh outline set.

[0062] Step S24: Determine the mesh outline based on each pixel in the mesh outline set, and mark the position of the corresponding mesh with a preset marker graphic.

[0063] In this step, the pixels on the mesh outline set form a closed graphic set, and the position of the mesh outline set is marked with a marker image (e.g., rectangle, circle, etc.).

[0064] Further, in an optional embodiment, step S30 includes: using feature matching to register and fuse the multiple local images in the axial and circumferential directions respectively to obtain a stitched image of the mesh. In this embodiment, some specially shaped features can be set on the mesh, or the mesh holes on the mesh can be used as features. After obtaining multiple local images of the mesh, the features in the images can be identified first, and then the image registration and stitching can be performed using feature matching.

[0065] Further, in an optional embodiment, step S30 includes:

[0066] Step S31: Obtain the template image of the mesh;

[0067] In this step, it should be noted that the template image is formed by analyzing and stitching the images captured by the camera array after the drum screen, screen, camera array, etc. are installed and before the formal screen inspection is carried out.

[0068] Step S32: Determine the spatial position of each local image based on the marked position of the mesh in each local image and the template image;

[0069] In this step, the spatial location of each local image (or the image after binarization) is determined by marking the positional feature information of the mesh holes.

[0070] Step S33: Based on the spatial position of each local image and the rotation speed of the drum mesh, register the multiple local images in the axial and circumferential directions respectively;

[0071] Step S34: Remove overlapping regions from the registered local images and fuse the images after removing overlapping regions to obtain the stitched image of the mesh.

[0072] In this step, after image registration, redundant overlapping areas in each local image can be removed before fusion and stitching.

[0073] Furthermore, after step S34, the following steps are also included:

[0074] Calculate the difference between the center coordinates of each mesh in the template image and the center point of the corresponding mesh mark in the stitched image, and determine the spatial position of each mesh in the stitched image based on the difference.

[0075] In this step, the actual spatial position of each mesh is determined by calculating the difference between the coordinates of the center of each mesh in the template image and the center point of the corresponding mark in the stitched image.

[0076] Furthermore, in an optional embodiment, between steps S30 and S40, the method further includes: correcting the fusion region in the stitched image based on the template image. In this embodiment, it should be noted that due to the influence of light and other factors, some local images may be distorted; therefore, the template image can be used to enhance and correct the image in the fusion region.

[0077] Further, step S40 includes:

[0078] Step S41: Calculate the area of ​​each mesh in the stitched image, and subtract the calculated area from the standard area of ​​the mesh in the template image to obtain the area difference.

[0079] Step S42: Determine whether the area difference is greater than a first set value or less than a second set value, wherein the first set value is greater than zero and the second set value is less than zero;

[0080] Step S43: If the area difference is greater than the first set value, then it is determined that the corresponding mesh is corroded.

[0081] Step S44: If the area difference is less than the second set value, it is determined that the corresponding mesh is blocked;

[0082] Step S45: If the area difference is not greater than the first set value and not less than the second set value, then the corresponding mesh is determined to be normal.

[0083] In this embodiment, a closed contour graphic area threshold discrimination method is used to automatically identify and locate mesh blockage and corrosion defects. Specifically, the spatial calibration position information of the array camera is used to calculate the area of ​​each mesh in the stitched image, that is, the area s enclosed by each closed contour set. Then, the area difference is calculated using the following formula: Δs = s - k2, where Δs is the area difference, s is the calculated mesh area, and k2 is the standard mesh area in the template image, that is, the pixel area corresponding to a normal mesh. When the absolute value of Δs is less than or equal to k3 (k3 = 0 or a positive value close to 0), the mesh is judged to be normal; when Δs > 0 and the absolute value of Δs is greater than k3, the mesh is judged to be corroded, and the spatial position information of the mesh is output and the mesh is marked; when Δs < 0 and the absolute value of Δs is less than k3, the mesh is judged to be blocked. It should be noted that in this embodiment, the absolute values ​​of the first set value and the second set value are equal. It should be understood that in other embodiments, the absolute values ​​of the two may not be equal. Finally, the spatial location information corresponding to abnormal meshes can be output, and the mesh area can be marked for easy manual confirmation.

[0084] Furthermore, in an optional embodiment, the method for detecting the mesh of a nuclear power plant blasting grid of the present invention further includes: sending the abnormality judgment result of each mesh opening of the mesh to a remote monitoring module.

[0085] The present invention also constructs a computer device, including a processor and a memory storing a computer program, wherein the processor, when executing the computer program, implements the steps of the above-described method for detecting the mesh of a nuclear power plant's blower screen.

[0086] The processor of this invention provides computing and control capabilities to support the operation of the entire mesh inspection system. It should be understood that, in the embodiments of this application, the processor may be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0087] This invention also constructs a mesh detection system for bulging wire mesh in nuclear power plants, combined with Figure 5 The mesh inspection system of this embodiment includes: a computer device, a light source array 12 disposed on one side of the mesh 101, and a camera array 11 disposed on the other side of the mesh 101. The logical structure diagram of the computer device can be referred to the above description and will not be repeated here.

[0088] Figure 6This is a logical structure diagram of a second embodiment of the nuclear power plant mesh inspection system of the present invention. The mesh inspection system of this embodiment includes: a camera array 11, a light source array 12, a computer device 13, a camera drive control module 14, a light source control module 15, an image acquisition and preprocessing module 16, a local communication and power supply module 17, a remote communication and power supply module 18, and a remote monitoring and interaction module 19. The camera array 11, light source array 12, camera drive control module 14, light source control module 15, image acquisition and preprocessing module 16, and local communication and power supply module 17 are located on the local side, i.e., at the mesh installation site. Furthermore, the camera array 11 is located on one side of the mesh 101, and the light source array 12 is located on the other side of the mesh 101. The camera drive control module 14 is used to drive and control the camera array 11. The light source control module 15 is used to control the light source array 12. The image acquisition and preprocessing module 16 is used to preprocess the images acquired by the camera array. The local communication power supply module 17 is used to supply power to the camera array 11, the light source array 12, the camera drive control module 14, the light source control module 15, and the image acquisition preprocessing module 16, and to send the preprocessed image to a remote computer device. In addition, the computer equipment 13, the remote communication power supply module 18, and the remote monitoring and interaction module 19 are located remotely. The remote communication power supply module 18 supplies power to the computer equipment 13 and the remote monitoring and interaction module 19, and receives image information transmitted from the local side. The computer equipment 13 includes an image binarization and mesh stitching positioning module, a mesh blockage and corrosion identification and analysis module, and a data storage and recording module. The image binarization and mesh stitching positioning module binarizes each pre-processed local image before stitching and positioning it to obtain a stitched image of the mesh. The mesh blockage and corrosion identification and analysis module identifies whether the mesh has experienced blockage or corrosion based on the area of ​​the closed contour graphic in the stitched image. The data storage and recording module stores and records the received image data, intermediate processing data, and processing result data, facilitating the formation of standardized data that is easily analyzed and processed by the computer over a long period, thus benefiting the long-term storage and intelligent analysis of drum mesh inspection data. The remote monitoring and interaction module 19 allows inspectors to view the inspection results. Finally, it should be noted that this embodiment only shows the main functional modules of the mesh inspection system. These functional modules can be recombined according to the specific hardware implementation scheme. For example, the camera drive control 14 and the light source control module 15 can be combined, the communication power supply module can be separated, and the local and remote power supply modules can also use different power supplies, etc.

[0089] Finally, it should be noted that the steps in the method of this embodiment can be adjusted, combined, or deleted according to actual needs. Similarly, the units or sub-units in the device of this embodiment can be combined, divided, or deleted according to actual needs.

[0090] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that all or part of the processes for implementing the above embodiments, and equivalent variations made in accordance with the claims of the present invention, still fall within the scope of the invention.

Claims

1. A method of inspecting a mesh of a drum screen of a nuclear power plant, characterized by, The method comprises the following steps: Step 10: acquiring multiple partial images of the mesh in axial and circumferential directions respectively by full-coverage image acquisition of the mesh by an array camera while the drum mesh rotates; Step 20: performing image processing on each partial image respectively to identify the mesh hole contour in the corresponding partial image and mark the position of the corresponding mesh hole in the corresponding partial image according to the mesh hole contour; Step 30: registering and fusing the multiple partial images in axial and circumferential directions respectively by feature matching to acquire a spliced image of the mesh; Step 40: calculating the area of each mesh hole in the spliced image respectively and judging whether each mesh hole of the mesh is abnormal according to the calculated area; The step 30 comprises: Step 31: acquiring a template image of the mesh; Step 32: determining the spatial position of each partial image according to the marked position of the mesh hole in each partial image and the template image; Step 33: registering the multiple partial images in axial and circumferential directions respectively according to the spatial position of each partial image and the rotation speed of the drum mesh; Step 34: removing the overlapping area in the registered partial images and fusing the images after removing the overlapping area to acquire the spliced image of the mesh; calculating the difference between the center coordinate position of each mesh hole in the template image and the center point position of the marked pattern of the corresponding mesh hole in the spliced image and determining the spatial position corresponding to each mesh hole in the spliced image according to the difference.

2. The method of claim 1, wherein the mesh detection method of a drum screen of a nuclear power plant is characterized by, The step 20 comprises: Step 21: performing binaryzation processing on each partial image respectively; Step 22: determining the mesh hole boundary point by gradient discrimination algorithm for each binaryzation-processed partial image; Step 23: calculating the distance value between each mesh hole boundary point and the adjacent mesh hole boundary point and judging whether the mesh hole boundary point is a pixel point on the mesh hole contour set according to the distance value; Step 24: determining the mesh hole contour according to each pixel point on the mesh hole contour set and marking the position of the corresponding mesh hole by a preset marked pattern.

3. The method of claim 1, wherein the mesh detection method of a drum screen of a nuclear power plant is characterized by, Between the step 30 and the step 40, it further comprises: correcting the fusion area in the spliced image according to the template image.

4. The method of claim 1, wherein the mesh detection method of a drum screen of a nuclear power plant is characterized by, The step 40 comprises: Step 41: calculating the area of each mesh hole in the spliced image respectively and subtracting the calculated area from the standard area of the mesh hole in the template image to acquire an area difference value; Step 42: judging whether the area difference value is greater than a first set value or less than a second set value, wherein the first set value is greater than zero and the second set value is less than zero; Step 43: if the area difference value is greater than the first set value, it is determined that the corresponding mesh hole is corroded; Step 44: if the area difference value is less than the second set value, it is determined that the corresponding mesh hole is blocked; Step 45: if the area difference value is neither greater than the first set value nor less than the second set value, it is determined that the corresponding mesh hole is normal.

5. The method of claim 1, wherein the mesh detection method of a drum screen of a nuclear power plant is characterized by, It further comprises: sending the abnormal judgment result of each mesh hole of the mesh to a remote monitoring module.

6. A computer device comprising a processor and a memory having stored therein a computer program, characterized in that, The processor implements the steps of the mesh detection method of the nuclear power plant drum mesh of any one of claims 1-5 when executing the computer program.

7. A system for detecting a mesh of a drum screen of a nuclear power plant, characterized by Comprising: an array of light sources disposed on one side of the mesh; an array of cameras disposed on the other side of the mesh; The computer device of claim 6.

Images