Boiler water-cooled wall inspection method, system and related components based on wall-climbing robot

Abstract

This application relates to a method, system, and related components for boiler water-cooled wall inspection based on a wall-climbing robot, belonging to the technical field of boiler water-cooled wall inspection. The method includes: acquiring image information of the area to be inspected on the boiler water-cooled wall captured by the wall-climbing robot; inputting the image information into a trained damage analysis neural network model to determine whether damage exists in the area to be inspected in the image information; if damage exists, acquiring damage information, searching for solutions to the damage in a database based on the damage information; and generating prompt information including the damage information and the solution. This application has the effect of improving work efficiency.

Description

Technical Field

[0001] This application relates to the technical field of boiler water-cooled wall inspection, and in particular to a boiler water-cooled wall inspection method, system and related components based on a wall-climbing robot. Background Technology

[0002] Boilers are important equipment in thermal power plants, and water-cooled walls are an important part of boilers. Depending on the unit capacity and boiler size, the height of water-cooled walls varies from 60 meters to 100 meters.

[0003] Currently, wall-climbing robots for inspecting boiler water-cooled walls are widely used. These robots use magnetic attraction to adhere to the water-cooled walls and employ a magnetically attached wheeled walking method. The intelligent climbing equipment platform is equipped with a wall thickness detector, visual equipment, storage device, and multi-functional load platform, enabling rapid and accurate detection of the boiler furnace environment and key areas. The relevant measurement results can be summarized in a visual and systematic way for maintenance and management personnel. This intelligent equipment has functions such as inspection, rust removal, and marking of boiler water-cooled walls, providing assistance for hazard elimination and scientific decision-making.

[0004] Currently, after the wall-climbing robot identifies an anomaly in the boiler's water-cooled wall, it provides a notification about the anomaly, which is then handled by staff. This process involves a large workload and low efficiency. Summary of the Invention

[0005] To improve work efficiency, this application provides a boiler water-cooled wall inspection method, system, and related components based on a wall-climbing robot.

[0006] In a first aspect, this application provides a boiler water-cooled wall inspection method based on a wall-climbing robot, employing the following technical solution:

[0007] Acquire image information of the test area of ​​the boiler water-cooled wall captured by the wall-climbing robot;

[0008] The image information is input into a trained damage analysis neural network model to determine whether there is damage in the area to be tested in the image information.

[0009] If damage exists, damage information is obtained, and a solution to resolve the damage is searched in the database based on the damage information;

[0010] Generate a prompt message that includes the damage information and the solution.

[0011] By adopting the above technical solution, the electronic device acquires image information of the test area captured by the wall-climbing robot, inputs the image information into the trained damage analysis neural network model, and then determines whether the test area has damage. If damage exists, the damage information is determined, and then the corresponding solution is searched in the database, and prompt information is generated to prompt the inspection personnel to check the relevant information. Therefore, after detecting damage, the solution is automatically queried, providing reference for the inspection personnel and improving work efficiency.

[0012] Furthermore, before inputting the image information into the trained damage analysis neural network model, the method further includes:

[0013] The image information is segmented to determine at least one sub-image information of a preset size;

[0014] Each of the sub-image information is compared with the undamaged image, and the sub-image information with inconsistent comparison is determined as the first image information;

[0015] The first image information corresponding to the image information is labeled and input into the trained damage analysis neural network model.

[0016] By adopting the above technical solution, the electronic device divides the image information into multiple sub-image information, compares the sub-image information with the undamaged image, filters out the undamaged sub-image information, and determines the first sub-image information that may have damage as the first image information. Therefore, it can improve the analysis speed of the damage analysis neural network model and improve work efficiency.

[0017] Furthermore, determining whether there is damage to the boiler water-cooled wall in the image information includes:

[0018] The analysis results are obtained after the first image information is input into the trained damage analysis neural network model.

[0019] Add a damage label to the label of the first image information that shows damage in the analysis results.

[0020] By adopting the above technical solution, the electronic device obtains the analysis results of the first image information. When the analysis results indicate that there is damage, a damage label is added to the first image information to facilitate unified management.

[0021] Furthermore, the damage information includes the damage location, and determining the damage information includes:

[0022] Obtain the actual location information of the area to be tested;

[0023] Obtain the damaged area from the first image information;

[0024] The first image information with the same label is combined to determine the combined damage area;

[0025] A first coordinate system is established based on the image information;

[0026] A second coordinate system is established based on the actual location information, and the first coordinate system is mapped to the second coordinate system.

[0027] Determine the first coordinate group of the damaged area in the first coordinate system;

[0028] The damage location of the damaged area is determined in the second coordinate system based on the first coordinate set.

[0029] By adopting the above technical solution, the electronic device acquires the actual location information of the area to be tested, establishes a second coordinate system based on the actual location information, establishes a first coordinate system on the first image information, and then determines the damaged area in the first image information. After combining the first image information, the combined damaged area is determined, and then the first coordinate group of the damaged area is determined in the first coordinate system. Since the first coordinate system and the second coordinate system can correspond, the damaged area can be determined in the second coordinate system, and thus the damage information can be determined accurately.

[0030] Furthermore, if the damaged area is located at the edge of the image information, the method further includes:

[0031] Determine the location of the damaged area at the edge in the image information;

[0032] The wall-climbing robot is controlled to move a preset distance in the specified direction to acquire new image information of the new test area, which is adjacent to the previous test area.

[0033] The new image information is segmented to determine at least one sub-image information of a preset size;

[0034] The sub-image information adjacent to the previous image information is determined as the first image information. The remaining sub-image information is compared with the undamaged image, and the sub-image information with inconsistent comparison is determined as the first image information.

[0035] The first image information is input into the trained damage analysis neural network model.

[0036] By adopting the above technical solution, the electronic device can determine the damaged area at the edge of the image information, and then determine the location of the damaged area, so that the wall-climbing robot can move to the corresponding location, monitor whether there is damage in the adjacent area, and after acquiring new image information, it can perform segmentation processing, retain the sub-image information that may have a higher probability of damage, and then perform damage recognition, thereby improving processing speed and work efficiency.

[0037] Furthermore, the step of searching for a solution to the damage in the database based on the damage information includes:

[0038] Obtain the type and area of ​​the damage;

[0039] Based on the type of damage, search the database for alternative solutions corresponding to that type;

[0040] The solution is determined from the alternative solutions based on the area of ​​the damage.

[0041] By adopting the above technical solution, electronic devices can quickly find a suitable solution by selecting a solution from the alternatives based on the type and area of ​​damage.

[0042] Furthermore, when at least two solutions exist, the generation of prompt information including the solutions includes:

[0043] At least two solutions were identified as solutions to be processed;

[0044] Obtain the success rate, time, and cost of each of the proposed solutions for handling the damage;

[0045] A first descending sequence of solutions to be processed is generated based on the success rate; a second ascending sequence of solutions to be processed is generated based on the time; and a third ascending sequence of solutions to be processed is generated based on the cost.

[0046] The fourth sequence is determined by multiplying the index of the solution to be processed in each sequence with the weight value corresponding to each sequence;

[0047] The solution is displayed in the prompt message in the order of the fourth sequence.

[0048] By adopting the above technical solution, the electronic device generates a sequence of solutions to be processed based on factors such as the success rate, time, and cost of the solutions to be processed. Then, it processes each sequence as needed to obtain an updated fourth sequence. The fourth sequence is closer to the manufacturer's requirements, so it can recommend the most suitable solution to the user.

[0049] Secondly, this application provides a boiler water-cooled wall inspection system based on a wall-climbing robot, which adopts the following technical solution:

[0050] The image information acquisition module is used to acquire image information captured by the wall-climbing robot on the test area of ​​the boiler water-cooled wall;

[0051] The damage analysis module is used to input the image information into a trained damage analysis neural network model to determine whether there is damage in the area to be tested in the image information;

[0052] The solution search module is used to determine damage information when the damage analysis module determines that damage exists, and search for solutions to the damage in the database based on the damage information;

[0053] The prompt information generation module is used to generate prompt information including the damage information and the solution.

[0054] By adopting the above technical solution, the image information acquisition module acquires image information of the test area captured by the wall-climbing robot, the damage analysis module inputs the image information into the trained damage analysis neural network model, and then determines whether the test area has damage. If damage exists, the damage information is determined, and then the solution search module searches for the corresponding solution in the database. The prompt information generation module generates prompt information to prompt the inspection personnel to check the relevant information. Therefore, after detecting damage, the solution is automatically searched, providing reference for the inspection personnel and improving work efficiency.

[0055] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0056] An electronic device, comprising:

[0057] At least one processor;

[0058] Memory;

[0059] At least one computer program, wherein the at least one computer program is stored in the memory and configured to be executed by the at least one processor, the at least one computer program being configured to: perform the method as described in any one of the first aspects.

[0060] By adopting the above technical solution, the processor executes the computer program in the memory to obtain image information of the test area taken by the wall-climbing robot, inputs the image information into the trained damage analysis neural network model, and then determines whether the test area has damage. If damage exists, the damage information is determined, and then the corresponding solution is searched in the database, and prompt information is generated to prompt the inspection personnel to check the relevant information. Therefore, after detecting damage, the solution is automatically queried, providing reference for the inspection personnel and improving work efficiency.

[0061] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0062] A computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method as described in any one of the first aspects.

[0063] By adopting the above technical solution, the processor executes a computer program in a computer-readable storage medium to obtain image information of the area to be tested taken by the wall-climbing robot. The image information is then input into a trained damage analysis neural network model to determine whether the area to be tested has damage. If damage exists, the damage information is determined, and a corresponding solution is searched in the database. A prompt message is generated to remind the inspection personnel to check the relevant information. Therefore, after detecting damage, the solution is automatically queried, providing a reference for the inspection personnel and improving work efficiency.

[0064] In summary, this application includes at least one of the following beneficial technical effects:

[0065] 1. Acquire image information of the test area taken by the wall-climbing robot, input the image information into the trained damage analysis neural network model, and then determine whether the test area has damage. If damage exists, the damage information is determined, and then the corresponding solution is searched in the database and a prompt message is generated to prompt the inspection personnel to check the relevant information. Therefore, after detecting damage, the solution is automatically searched to provide reference for the inspection personnel and improve work efficiency.

[0066] 2. By segmenting image information into multiple sub-image information, filtering out undamaged sub-image information, and identifying the first sub-image information that may contain damage as the first image information, the analysis speed of the damage analysis neural network model can be improved, thus increasing work efficiency.

[0067] 3. It can detect whether there is damage in adjacent areas in a timely manner. Attached Figure Description

[0068] Figure 1 This is a flowchart illustrating the boiler water-cooled wall detection method based on a wall-climbing robot in an embodiment of this application.

[0069] Figure 2 This is a schematic diagram of the structure after image information segmentation processing in the embodiments of this application.

[0070] Figure 3 This is a structural block diagram of the boiler water-cooled wall detection system based on a wall-climbing robot in an embodiment of this application.

[0071] Figure 4 This is a structural block diagram of the electronic device in the embodiments of this application. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0073] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0074] This application discloses a boiler water-cooled wall inspection method based on a wall-climbing robot. (Refer to...) Figure 1 This process is executed by an electronic device, which can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet, desktop computer, etc., but is not limited to these. The process includes the following steps (S101 to S104):

[0075] Step S101: Obtain image information of the area to be tested on the boiler water-cooled wall captured by the wall-climbing robot.

[0076] Specifically, electronic devices communicate with the wall-climbing robot to acquire image information captured by the robot during its operation. The wall-climbing robot moves along the boiler's water-cooled wall, which can be controlled by personnel or move automatically within a certain range. The area traversed by the wall-climbing robot is the test area, and the camera on the robot captures image information of this area.

[0077] Step S102: Input the image information into the trained damage analysis neural network model to determine whether there is damage in the area to be tested in the image information; if so, proceed to step S103.

[0078] Specifically, the electronic device presets a training sample set and a test sample set. Both the training sample set and the test sample set include multiple sets of image information and corresponding damage analysis results. The electronic device uses the training sample set to train the neural network model, and after the training is completed, it uses the test sample set to verify it. Once the analysis accuracy reaches the preset accuracy, the damage analysis neural network model is trained.

[0079] After acquiring image information, the electronic device inputs the image information into a trained damage analysis neural network model to obtain the output damage analysis results. Based on the damage analysis results, it is determined whether there is damage in the area to be tested.

[0080] In another possible implementation, damage may only exist in local locations in the region to be tested. In order to improve the analysis speed of the damage analysis neural network model, the size of the sample images in the training sample set used to train the damage analysis neural network model is smaller than the image information.

[0081] Therefore, before step S102, the method also includes (steps S11 to S13):

[0082] Step S11: Segment the image information to determine at least one sub-image information of a preset size.

[0083] For example, an electronic device can divide image information into 4×4 segments to obtain 16 sub-image segments, so the area of ​​each sub-image segment is smaller than the image information.

[0084] Step S12: Compare each sub-image information with the undamaged image, and determine the sub-image information with inconsistent comparison as the first image information.

[0085] Specifically, the electronic device pre-stores a large number of non-destructive images, which can be represented by the uniform color tone of the water-cooled walls in the image. The electronic device compares the sub-image information with the non-destructive images and determines the image information that is inconsistent in the comparison as the first image information, which can be represented by the presence of inconsistent color blocks in the first image information.

[0086] Step S13: Label the first image information corresponding to the image information and input it into the trained damage analysis neural network model.

[0087] For example, refer to Figure 2 Image information A is divided into 4 blocks of first image information (shaded areas). The 4 blocks of first image information are labeled with the label of image information A. Therefore, it can be determined that the 4 blocks of first image information belong to image information A.

[0088] Furthermore, once the electronic device inputs the first image information into the trained damage analysis neural network model, images that are not helpful in identifying damage have been removed, thereby improving the calculation speed.

[0089] Therefore, after the image information is divided into first image information, step S102 includes: obtaining the analysis result after the first image information is input into the trained damage analysis neural network model; and adding damage labels to the labels of the first image information that the analysis result indicates damage.

[0090] Specifically, when the analysis result of any first image information is that there is damage, a damage label is added to the label of the first image information, so that the damaged first image information can be recorded.

[0091] Step S103: Determine the damage information and search for solutions to the damage in the database based on the damage information.

[0092] Specifically, when image information is input into the damage analysis neural network model, the damage analysis neural network model can determine the damage information of the image information, including the damage location, damage area, and damage type.

[0093] The electronic device can determine the actual location of the damage based on the actual location of the wall-climbing robot.

[0094] Electronic devices have a pre-installed database that stores solutions for various types of damage. Based on different damage types and varying damage area sizes within each type, multiple solutions are available. Specific steps include:

[0095] Obtain the type and area of ​​damage; search for alternative solutions in the database based on the type of damage; determine the solution from the alternative solutions based on the area of ​​damage.

[0096] Specifically, electronic devices first determine alternative solutions based on type, then further filter the alternative solutions based on area, and through layers of filtering, quickly determine the most suitable solution.

[0097] When the image information is divided into multiple first image information, and the damage information includes the damage location, step S103, in determining the damage information, includes (steps S21 to S27):

[0098] Step S21: Obtain the actual location information of the area to be tested.

[0099] Specifically, during the inspection process, the electronic equipment can determine the position of the wall-climbing robot, and thus determine the actual position information of the area to be tested.

[0100] Step S22: Obtain the damaged area from the first image information.

[0101] Specifically, after the first image information is input into the trained damage analysis neural network model, the damage analysis neural network model can determine the damage area in the first image information.

[0102] Step S23: Combine the first image information with the same label to determine the combined damage area.

[0103] For example, an electronic device combines the first image information with tag A and damage tag. During the combination, the first image information is combined in the same position as when it was segmented, thereby determining the damage area after combination. In this process, the first image information without damage is screened out after analysis, which can also accurately determine the damage analysis results.

[0104] Step S24: Establish the first coordinate system based on the image information.

[0105] Step S25: Establish a second coordinate system based on the actual location information, and make the first coordinate system correspond to the second coordinate system.

[0106] Specifically, the electronic device acquires the inner plane unfolded view of the boiler water-cooled wall, determines the actual position information of the area to be measured on the inner plane unfolded view, the actual position information is an area, and then the electronic device establishes a second coordinate system on the actual position information.

[0107] Electronic devices can obtain the ratio of unit length between a first coordinate system and a second coordinate system, and then correspond the first coordinate system to the second coordinate system. For example, if the ratio of unit length between the second coordinate system and the first coordinate system is 10:1, then the point (10,10) in the second coordinate system corresponds to the point (1,1) in the first coordinate system.

[0108] Step S26: Determine the first coordinate group of the damaged area in the first coordinate system.

[0109] Step S27: Determine the location of the damage area in the second coordinate system based on the first coordinate group.

[0110] Specifically, the electronic device determines the coordinates of the edge of the damaged area in a first coordinate system, thereby obtaining a first coordinate set. Then, the electronic device can determine the coordinates of each coordinate in the first coordinate set in a second coordinate system, and thus determine the location of the damage based on the coordinates in the second coordinate system.

[0111] Step S104: Generate a prompt message including damage information and solutions.

[0112] Specifically, after the electronic device identifies the damage information and finds the corresponding solution, it generates a prompt message. Based on the prompt message, staff can determine the location of the damage and refer to the solution to handle the damage.

[0113] In another possible implementation, the damaged regions may be connected. If an edge region is determined to be damaged in an image, then the region adjacent to the current region to be tested is also likely to be damaged. Therefore, the above method also includes (steps S31 to S35):

[0114] Step S31: Determine the location of the damaged area at the edge in the image information.

[0115] Specifically, the orientation includes four directions: up, down, left, and right. If the damaged area is located at the upper edge of the image information, the damaged area is above the image information.

[0116] Step S32: Control the wall-climbing robot to move a preset distance in the orientation direction to obtain new image information of the new test area. The new test area is adjacent to the previous test area.

[0117] Specifically, the electronic device determines the orientation of the image information to correspond to the movement orientation of the wall-climbing robot inside the boiler water-cooled wall. If the upper part of the image information corresponds to the right side of the boiler water-cooled wall, the electronic device controls the wall-climbing robot to move a preset distance to the right. After the wall-climbing robot moves the preset distance, the corresponding new test area is adjacent to the previous test area, so the acquired new image information is adjacent to the previous image information.

[0118] Step S33: Segment the new image information to determine at least one sub-image information of a preset size.

[0119] Step S34: Determine the sub-image information adjacent to the previous image information as the first image information, compare the remaining sub-image information with the undamaged image, and determine the sub-image information with inconsistent comparison as the first image information.

[0120] Furthermore, in order to speed up the processing, the electronic device segments the image information of the new test area, for example, into 4×4 sub-image information of a preset size.

[0121] After obtaining the sub-image information, the electronic device first identifies a column or row of sub-image information adjacent to the previous image as the first image information. The remaining sub-image information is then subjected to non-destructive screening, and sub-image information that may be damaged is identified as the first image information.

[0122] Step S35: Input the first image information into the trained damage analysis neural network model.

[0123] Then, the electronic device uses the determined first image information to focus on investigating the adjacent areas of the damaged area.

[0124] In another possible implementation, when there are at least two solutions, step S104 generates a prompt message including the solutions, including (steps S41 to S45):

[0125] Step S41: Identify at least two solutions as options to be processed.

[0126] Step S42: Obtain the success rate, time, and cost of each proposed solution for handling damage.

[0127] Specifically, the electronic device stores information about the success rate, time, and cost of the damage treatment solution. This information can be obtained from big data or from historical data statistics.

[0128] Step S43: Generate a first descending sequence of pending solutions based on success rate, a second ascending sequence of pending solutions based on time, and a third ascending sequence of pending solutions based on cost.

[0129] Specifically, the higher the success rate, the shorter the time, and the lower the cost, the higher the value of the processing solution. Therefore, multiple sequences are generated based on the characteristics of each piece of relevant information.

[0130] Step S44: Multiply the index of the solution to be processed in each sequence by the weight value corresponding to each sequence to determine the fourth sequence.

[0131] Specifically, the weight value corresponding to each sequence is determined according to the manufacturer's own needs. For example, if the manufacturer has no requirements on cost and only seeks to quickly resolve the damage, then the weight value corresponding to the third ascending sequence is smaller, and the weight value corresponding to the second ascending sequence is larger. Therefore, the resulting fourth sequence is more in line with the manufacturer's needs.

[0132] Step S45: Display the solution in the prompt message according to the order of the fourth sequence.

[0133] To better implement the above method, this application also provides a boiler water-cooled wall inspection system based on a wall-climbing robot, referring to... Figure 3 The boiler water-cooled wall inspection system 200 based on a wall-climbing robot includes:

[0134] The image information acquisition module 201 is used to acquire image information captured by the wall-climbing robot on the test area of ​​the boiler water-cooled wall;

[0135] Damage analysis module 202 is used to input image information into a trained damage analysis neural network model to determine whether there is damage in the area to be tested in the image information;

[0136] The solution search module 203 is used to determine the damage information when the damage analysis module 202 determines that damage exists, and to search for solutions to the damage in the database based on the damage information.

[0137] The prompt message generation module 204 is used to generate prompt messages that include damage information and solutions.

[0138] Furthermore, the boiler water-cooled wall inspection system 200 based on a wall-climbing robot also includes:

[0139] The first segmentation module is used to segment the image information and determine at least one sub-image information of a preset size;

[0140] The first image information first determination module is used to compare each sub-image information with the undamaged image and determine the sub-image information with inconsistent comparison as the first image information.

[0141] The labeling module is used to label the first image information corresponding to the image information and input it into the trained damage analysis neural network model.

[0142] Furthermore, when the damage analysis module 202 determines whether there is damage to the boiler water-cooled wall in the image information, it is specifically used for:

[0143] The analysis results are obtained after the first image information is input into the trained damage analysis neural network model.

[0144] Add a damage label to the label of the first image information that shows damage in the analysis results.

[0145] Furthermore, when determining damage information, the solution lookup module 203 is specifically used for:

[0146] Obtain the actual location information of the area to be tested;

[0147] Obtain the damaged area from the first image information;

[0148] The first image information with the same label is combined to determine the combined damage area;

[0149] Establish a first coordinate system based on image information;

[0150] A second coordinate system is established based on the actual location information, and the first coordinate system is mapped to the second coordinate system.

[0151] Determine the first coordinate group of the damaged area in the first coordinate system;

[0152] The location of the damage area is determined in the second coordinate system based on the first coordinate set.

[0153] Furthermore, if the damaged area is located at the edge of the image information, the boiler water-cooled wall inspection system 200 based on the wall-climbing robot also includes:

[0154] The orientation determination module is used to determine the orientation of the damaged area located at the edge in the image information;

[0155] The new image information acquisition module is used to control the wall-climbing robot to move a preset distance in the orientation direction and acquire new image information of the new test area, which is adjacent to the previous test area.

[0156] The second segmentation module is used to segment the new image information and determine at least one sub-image information of a preset size;

[0157] The first image information second determination module is used to determine the sub-image information adjacent to the previous image information as the first image information, compare the remaining sub-image information with the undamaged image, and determine the sub-image information with inconsistent comparison as the first image information;

[0158] The second analysis module is used to input the first image information into the trained damage analysis neural network model.

[0159] Furthermore, the solution search module 203 searches the database for solutions to address the damage based on the damage information, specifically for:

[0160] Obtain the type and area of ​​the damage;

[0161] Search the database for alternative solutions corresponding to the type of damage;

[0162] The solution is determined from the alternative solutions based on the area of ​​the damage.

[0163] Furthermore, when at least two solutions exist, the prompt message generation module 204 generates a prompt message including the solutions, specifically for:

[0164] Identify at least two solutions as pending solutions;

[0165] Obtain the success rate, time, and cost of each proposed solution for handling damage;

[0166] Generate a first descending sequence of pending solutions based on success rate, a second ascending sequence based on time, and a third ascending sequence based on cost.

[0167] The fourth sequence is determined by multiplying the index of the solution to be processed in each sequence with the weight value corresponding to each sequence;

[0168] The solution is displayed in the prompt message in the order of the fourth sequence.

[0169] The various variations and specific examples of the methods in the foregoing embodiments are also applicable to the boiler water-cooled wall detection device based on the wall-climbing robot in this embodiment. Through the foregoing detailed description of the boiler water-cooled wall detection method based on the wall-climbing robot, those skilled in the art can clearly understand the implementation method of the boiler water-cooled wall detection device based on the wall-climbing robot in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0170] To better implement the above methods, embodiments of this application provide an electronic device, referring to... Figure 4 The electronic device 300 includes a processor 301, a memory 303, and a display screen 305. The memory 303 and the display screen 305 are both connected to the processor 301, such as via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in practical applications, the transceiver 304 is not limited to one type, and the structure of this electronic device 300 does not constitute a limitation on the embodiments of this application.

[0171] Processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 301 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0172] Bus 302 may include a pathway for transmitting information between the aforementioned components. Bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 302 may be divided into address bus, data bus, control bus, etc.

[0173] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0174] The memory 303 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0175] Figure 4 The electronic device 300 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0176] This application embodiment also provides a computer-readable storage medium storing a computer program. When executed by a processor, this program implements the boiler water-cooled wall inspection method based on a wall-climbing robot provided in the above embodiment. The processor executes the computer program in the computer-readable storage medium to acquire image information of the area to be tested captured by the wall-climbing robot, inputs the image information into a trained damage analysis neural network model, and then determines whether the area to be tested has damage. If damage exists, the damage information is determined, and then the corresponding solution is searched in the database, and a prompt message is generated to prompt the inspection personnel to view the relevant information. Therefore, after detecting damage, the solution is automatically queried, providing a reference for the inspection personnel and improving work efficiency.

[0177] In this embodiment, the computer-readable storage medium can be a tangible device that holds and stores instructions used by an instruction execution device. The computer-readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination thereof. Specifically, the computer-readable storage medium can be a portable computer disk, a hard disk, a USB flash drive, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), staging random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory stick, floppy disk, optical disk, magnetic disk, mechanical encoding device, or any combination thereof.

[0178] The computer program in this embodiment includes program code for performing all the aforementioned methods. The program code may include instructions corresponding to the method steps provided in the above embodiments. The computer program can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded to an external computer or external storage device via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network). The computer program can be executed entirely on the user's computer as a standalone software package.

[0179] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

[0180] Additionally, it should be understood that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

Claims

1. A method for detecting boiler water-cooled walls based on a wall-climbing robot, characterized in that, include: Acquire image information of the test area of ​​the boiler water-cooled wall captured by the wall-climbing robot; The image information is input into a trained damage analysis neural network model to determine whether there is damage in the area to be tested in the image information. If damage exists, damage information is obtained, and a solution to resolve the damage is searched in the database based on the damage information; Generate a prompt message that includes the damage information and the solution; If the damaged area is located at the edge of the image information, the method further includes: Determine the location of the damaged area at the edge in the image information; The wall-climbing robot is controlled to move a preset distance in the specified direction to acquire new image information of the new test area, which is adjacent to the previous test area. The new image information is segmented to determine at least one sub-image information of a preset size; The sub-image information adjacent to the previous image information is determined as the first image information. The remaining sub-image information is compared with the undamaged image, and the sub-image information with inconsistent comparison is determined as the first image information. The first image information is input into the trained damage analysis neural network model; The step of searching for a solution to the damage in the database based on the damage information includes: Obtain the type and area of ​​the damage; Based on the type of damage, search the database for alternative solutions corresponding to that type; The solution is determined from the alternative solutions based on the area of ​​the damage. When at least two solutions exist, the generation of prompt information including the solutions includes: At least two solutions were identified as solutions to be processed; Obtain the success rate, time, and cost of each of the proposed solutions for handling damage; A first descending sequence of solutions to be processed is generated based on the success rate; a second ascending sequence of solutions to be processed is generated based on the time; and a third ascending sequence of solutions to be processed is generated based on the cost. The fourth sequence is determined by multiplying the index of the solution to be processed in each sequence with the weight value corresponding to each sequence; The solution is displayed in the prompt message in the order of the fourth sequence.

2. The method according to claim 1, characterized in that, Before inputting the image information into the trained damage analysis neural network model, the method further includes: The image information is segmented to determine at least one sub-image information of a preset size; Each of the sub-image information is compared with the undamaged image, and the sub-image information with inconsistent comparison is determined as the first image information; The first image information corresponding to the image information is labeled and input into the trained damage analysis neural network model.

3. The method according to claim 2, characterized in that, The step of determining whether there is damage to the boiler water-cooled wall in the image information includes: The analysis results are obtained after the first image information is input into the trained damage analysis neural network model. Add a damage label to the label of the first image information that shows damage in the analysis results.

4. The method according to claim 3, characterized in that, The damage information includes the damage location, and determining the damage information includes: Obtain the actual location information of the area to be tested; Obtain the damaged area from the first image information; The first image information with the same label is combined to determine the combined damage area; A first coordinate system is established based on the image information; A second coordinate system is established based on the actual location information, and the first coordinate system is mapped to the second coordinate system. Determine the first coordinate group of the damaged area in the first coordinate system; The damage location of the damaged area is determined in the second coordinate system based on the first coordinate set.

5. A boiler water-cooled wall inspection system based on a wall-climbing robot, characterized in that, include: The image information acquisition module is used to acquire image information captured by the wall-climbing robot on the test area of ​​the boiler water-cooled wall; The damage analysis module is used to input the image information into a trained damage analysis neural network model to determine whether there is damage in the area to be tested in the image information; The solution search module is used to determine damage information when the damage analysis module determines that damage exists, and search for solutions to the damage in the database based on the damage information; The prompt information generation module is used to generate prompt information including the damage information and the solution; If the damaged area is located at the edge of the image information, the system further includes: The orientation determination module is used to determine the orientation of the damaged area located at the edge in the image information; The new image information acquisition module is used to control the wall-climbing robot to move a preset distance in the orientation direction and acquire new image information of the new test area, which is adjacent to the previous test area. The second segmentation module is used to segment the new image information and determine at least one sub-image information of a preset size; The first image information second determination module is used to determine the sub-image information adjacent to the previous image information as the first image information, compare the remaining sub-image information with the undamaged image, and determine the sub-image information with inconsistent comparison as the first image information; The second analysis module is used to input the first image information into the trained damage analysis neural network model; The solution search module searches the database for solutions to the damage based on the damage information, specifically for: Obtain the type and area of ​​the damage; Search the database for alternative solutions corresponding to the type of damage; The solution is determined from the alternative solutions based on the area of ​​the damage; When at least two solutions exist, the prompt information generation module generates prompt information including the solutions, specifically for: Identify at least two solutions as pending solutions; Obtain the success rate, time, and cost of each proposed solution for handling damage; Generate a first descending sequence of pending solutions based on success rate, a second ascending sequence based on time, and a third ascending sequence based on cost. The fourth sequence is determined by multiplying the index of the solution to be processed in each sequence with the weight value corresponding to each sequence; The solution is displayed in the prompt message in the order of the fourth sequence.

6. An electronic device, characterized in that, include: At least one processor; Memory; At least one computer program, wherein the at least one computer program is stored in the memory and configured to be executed by the at least one processor, the at least one computer program being configured to perform the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed as described in any one of claims 1 to 4.

Images