Multi-sensor fusion self-diagnosis system for transmission line sealing operation

Abstract

The present invention discloses a self-diagnosis system for a transmission line network sealing operation integrated machine with multi-sensor fusion, which relates to the field of electric power construction technology. The system includes: a network sealing characteristic recognition module, which is used to collect network sealing images, and perform network sealing area recognition and network sealing uniformity recognition; a network sealing area fault self-diagnosis module, which is used to perform network sealing area fault self-diagnosis; a key network sealing area acquisition module, which is used to obtain multiple key network sealing areas whose uniformity information deviates from a preset uniformity interval; a network sealing damage recognition module, which is used to perform network sealing damage recognition; a movement anomaly recognition module; and a network sealing fault self-diagnosis module. The present invention solves the technical problem that the traditional method in the prior art can only diagnose a single fault, and the accuracy and comprehensiveness of the diagnosis are low, and achieves the technical effect of comprehensively combining multiple aspects of information such as network sealing area, network sealing damage, and guide wheel movement anomaly, and performing comprehensive fault self-diagnosis to improve the accuracy and comprehensiveness of the diagnosis.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power construction, and in particular to a multi-sensor fusion integrated machine self-diagnosis system for power transmission line sealing operations. Background Art

[0002] Network sealing operations are usually carried out in complex and dangerous environments, such as high altitude and high pressure environments, and the operation area is large and time is tight, requiring the equipment to be highly automated and stable. As the core equipment for network sealing operations, the all-in-one operation machine must be able to operate stably under different environmental conditions and promptly identify and handle faults during the operation.

[0003] There are many types of faults involved in net sealing operations, including mechanical faults (such as guide wheel jamming, net damage), environmental factors (such as temperature, humidity, friction changes), etc. These faults may occur independently or as a result of multiple factors. The identification and classification of faults require the support of highly complex diagnostic algorithms. However, traditional methods can often only diagnose a single fault, lack comprehensive analysis capabilities, and have low diagnostic accuracy and comprehensiveness. Summary of the invention

[0004] The present application provides a multi-sensor fusion power transmission line sealing operation integrated machine self-diagnosis system, which is used to solve the technical problem that the traditional method in the prior art can only diagnose a single fault and has low diagnostic accuracy and comprehensiveness.

[0005] The present application provides a self-diagnosis system for a transmission line network sealing operation integrated machine with multi-sensor fusion, the system comprising: a network sealing characteristic recognition module, the network sealing characteristic recognition module is used to collect network sealing images when the transmission line network sealing operation integrated machine performs network sealing operation in the maximum range based on multiple sensors, perform network sealing area recognition and network sealing uniformity recognition, obtain network sealing area deviation information and network sealing uniformity distribution, wherein the network sealing uniformity distribution includes uniformity information in multiple network sealing areas; a network sealing area fault self-diagnosis module, the network sealing area fault self-diagnosis module is used to perform network sealing area fault self-diagnosis according to the network sealing area deviation information, and obtain network sealing area fault diagnosis results; a key network sealing area acquisition module, the key network sealing area acquisition module is used to index and obtain multiple key sealing areas whose uniformity information deviates from a preset uniformity interval according to the network sealing uniformity distribution. a network area, and indexes multiple key guide wheels in the multiple key network sealing areas; a network sealing damage identification module, the network sealing damage identification module is used to collect multiple key area images of the multiple key network sealing areas, perform network sealing damage identification, and obtain multiple area network sealing damage identification results; a movement abnormality identification module, the movement abnormality identification module is used to collect the network sealing environment information of the current network sealing operation and the line feature information on the surface of the transmission line based on multiple sensors, perform movement abnormality identification of the multiple key guide wheels, and obtain multiple guide wheel movement abnormality identification results; a network sealing fault self-diagnosis module, the network sealing fault self-diagnosis module is used to perform network sealing fault self-diagnosis of the transmission line network sealing operation integrated machine according to the network sealing area fault diagnosis result, multiple area network sealing damage identification results and the multiple guide wheel movement abnormality identification results, and obtain the integrated machine self-diagnosis result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The multi-sensor fusion transmission line network sealing operation integrated machine self-diagnosis system provided in the present application relates to the field of electric power construction technology. By collecting the network sealing images when the transmission line network sealing operation integrated machine performs the maximum range network sealing operation, the network sealing area and network sealing uniformity are identified, and the network sealing area fault self-diagnosis, network sealing damage identification and key guide wheel movement abnormality identification are performed. Finally, according to the network sealing area fault diagnosis results, multiple area network sealing damage identification results and multiple guide wheel movement abnormality identification results, the network sealing fault self-diagnosis of the transmission line network sealing operation integrated machine is performed. It solves the technical problem that the traditional method in the prior art can only diagnose a single fault, and the accuracy and comprehensiveness of the diagnosis are low, and realizes comprehensive information such as network sealing area, network sealing damage and guide wheel movement abnormality, and performs comprehensive fault self-diagnosis, improving the accuracy and comprehensiveness of the diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic diagram of the structure of a self-diagnosis system for a multi-sensor fusion transmission line sealing operation all-in-one machine provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of the process of identifying the network sealing area and the network sealing uniformity in the self-diagnosis system of the multi-sensor fusion power transmission line network sealing operation integrated machine provided in the embodiment of the present application;

[0011] Figure 3 A schematic diagram of the process of performing self-diagnosis of a transmission line network sealing operation integrated machine for network sealing failure in the self-diagnosis system of the transmission line network sealing operation integrated machine with multi-sensor fusion provided in an embodiment of the present application.

[0012] Explanation of the reference numerals: network blocking characteristic identification module 11, network blocking area fault self-diagnosis module 12, key network blocking area acquisition module 13, network blocking damage identification module 14, movement anomaly identification module 15, network blocking fault self-diagnosis module 16. DETAILED DESCRIPTION

[0013] The present application provides a multi-sensor fusion power transmission line sealing operation integrated machine self-diagnosis system, which is used to solve the technical problem that the traditional method in the prior art can only diagnose a single fault and has low diagnostic accuracy and comprehensiveness.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Example

[0016] like Figure 1 As shown, the present application provides a multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system, the system comprising:

[0017] The network sealing characteristic identification module 11 is used to collect network sealing images when the power transmission line network sealing operation integrated machine performs the maximum range network sealing operation based on multiple sensors, perform network sealing area identification and network sealing uniformity identification, and obtain network sealing area deviation information and network sealing uniformity distribution, wherein the network sealing uniformity distribution includes uniformity information in multiple network sealing areas.

[0018] Further, such as Figure 2 As shown, the network blocking characteristic identification module 11 is further used to perform the following steps:

[0019] P11: Based on multiple sensors, collect network sealing images when the integrated transmission line network sealing operation machine performs the maximum range network sealing operation; P12: Train a network sealing identifier for identifying the network sealing area and the uniformity of the network sealing, wherein the network sealing identifier includes a network sealing area identification branch and a network sealing uniformity identification branch; P13: Use the network sealing identifier to identify the network sealing image to obtain the network sealing area information and the network sealing uniformity distribution; P14: Based on the network sealing area information and the standard network sealing area information of the maximum range network sealing, calculate and obtain the network sealing area deviation information.

[0020] It should be understood that the network sealing characteristic identification module 11 of the present application is responsible for accurately capturing the real-time image of the network sealing operation by integrating a multi-sensor system when the all-in-one machine performs the maximum range network sealing operation, and performing network sealing area identification and network sealing uniformity identification through image processing and data analysis technology, and obtaining network sealing area deviation information and network sealing uniformity distribution. Among them, the network sealing area deviation information reflects the difference between the actual network sealing area and the preset standard area, and the preset standard area represents the network sealing coverage range under ideal conditions. The actual network sealing area may change due to reasons such as the obstruction of the guide wheel movement and the damage of the network sealing. The module identifies the deviation information of the network sealing area by comparing the standard area and the actual network sealing area. The network sealing uniformity distribution reveals the density distribution of the network sealing in each area. The network sealing area can be divided into multiple sub-areas, and the network sealing density in each area can be identified to generate uniformity distribution information. If the density of a certain area changes (possibly due to the obstruction of the guide wheel movement or the damage of the network sealing), the uniformity distribution will show the area deviating from the normal.

[0021] Specifically, firstly, based on the collaborative work of multiple sensors, network blocking images are collected when the all-in-one machine performs the maximum range of network blocking operations. These sensors can include visual sensors (such as cameras), laser radars, etc., which are used to capture the network blocking status at different angles and areas during the operation. These images will become the basic data for subsequent identification and analysis.

[0022] Furthermore, a large amount of network blocking image data is trained through machine learning or deep learning technology to generate a network blocking identifier. The network blocking identifier can automatically process and analyze the network blocking images collected from the sensor. In addition, the network blocking identifier contains two main branches: a network blocking area identification branch and a network blocking uniformity identification branch. The network blocking area identification branch focuses on identifying the area actually covered in the network blocking operation, while the network blocking uniformity identification branch is used to analyze the density and uniformity of the network blocking area.

[0023] Furthermore, the collected network sealing images are processed using the trained network sealing identifier. The network sealing area recognition branch extracts the actual network sealing area information in the image, while the network sealing uniformity recognition branch analyzes the density distribution of each network sealing area to generate network sealing uniformity distribution information. Then, based on the obtained network sealing area information, it is compared with the preset maximum range network sealing standard area information. Through this comparison, the difference between the actual network sealing area and the standard area is calculated, that is, the network sealing area deviation information. This indicator directly reflects the gap between the actual effect of the network sealing operation and the expected target, and provides a key basis for subsequent fault self-diagnosis. At the same time, the network sealing uniformity distribution map also provides an intuitive visualization method for identifying potential network sealing quality problems, which helps operators quickly locate and deal with uneven phenomena in the network sealing process.

[0024] Furthermore, step P12 of the embodiment of the present application also includes:

[0025] P12-1: Based on the recorded data of the largest-scale network sealing operation of the power transmission line network sealing operation integrated machine in the historical time, a set of sample network sealing images is collected; P12-2: According to the actual network sealing area in each sample network sealing image, a set of sample network sealing area information is obtained, and according to the network sealing density of multiple network sealing areas in each sample network sealing image, a set of sample network sealing uniformity distribution is obtained; P12-3: Using the sample network sealing image set as input data, and using the sample network sealing area information set and the sample network sealing uniformity distribution set as supervision data respectively, a network sealing area identification branch and a network sealing uniformity identification branch are trained; P12-4: Combining the network sealing area identification branch and the network sealing uniformity identification branch to obtain the network sealing identifier.

[0026] Specifically, the detailed process of training the network blocking identifier can be, first, based on the recorded data of the transmission line network blocking operation integrated machine when performing the largest range of network blocking operations in historical time, collect relevant sample network blocking image sets. These sample images are image data actually collected in historical operations, which not only cover various operating conditions and environmental changes, but also include network blocking examples of different shapes and sizes, providing a rich data foundation for subsequent model training.

[0027] Afterwards, professional labelers can label each sample network blocking image. For the network blocking area, accurate measurement tools and methods are used to obtain the actual network blocking area information in each sample image, and a sample network blocking area information set is established accordingly. At the same time, in order to evaluate the uniformity of the network blocking, each sample image is subdivided into multiple network blocking areas, and labeled according to the network blocking density of each area, and finally a sample network blocking uniformity distribution set is generated.

[0028] Furthermore, the sample network blocking image set is used as input data, and the sample network blocking area information set and the sample network blocking uniformity distribution set are used as supervision data to train the network blocking area recognition branch and the network blocking uniformity recognition branch in the network blocking identifier. After the training of the above branches is completed, the network blocking area recognition branch and the network blocking uniformity recognition branch are combined to form a complete network blocking identifier. This identifier can process network blocking images at the same time and output information about network blocking area and network blocking uniformity.

[0029] The network sealing area fault self-diagnosis module 12 is used to perform network sealing area fault self-diagnosis according to the network sealing area deviation information to obtain the network sealing area fault diagnosis result.

[0030] Furthermore, the network blocking area fault self-diagnosis module 12 is also used to perform the following steps:

[0031] P21: Collect a set of sample network cover area deviation information, and mark and obtain a set of sample network cover area fault diagnosis results according to the size of the sample cover area deviation information; P22: Use the sample network cover area deviation information set and the sample network cover area fault diagnosis result set to construct an area fault diagnosis classifier, diagnose and classify the network cover area deviation information, and obtain the network cover area fault diagnosis result.

[0032] Optionally, in order to realize automatic diagnosis of network blocking area fault, the network blocking area fault self-diagnosis module 12 performs self-diagnosis of the network blocking area fault by analyzing the network blocking area deviation information, and generates a network blocking area fault diagnosis result.

[0033] First, a wide range of sample network closure area deviation information is collected from historical data or experimental environments. These samples not only cover area deviations within the normal range, but also include abnormal deviations caused by various types of faults. In order to train the area fault diagnosis classifier, the system carefully annotates the network closure area deviation information of each sample, that is, according to the size and characteristics of the deviation information, it is classified into different fault types or normal states, thus forming a sample network closure area fault diagnosis result set. This step is the basis for building a classifier, and its accuracy directly affects the reliability of subsequent diagnosis.

[0034] Subsequently, machine learning or deep learning technology is used to construct an area fault diagnosis classifier with the sample network closure area deviation information set as input features and the sample network closure area fault diagnosis result set as supervision labels. This classifier can learn and understand the mapping relationship between different deviation information and fault types, so as to have the ability to automatically diagnose and classify the newly input network closure area deviation information. During the construction process, a variety of algorithms and technical means can be used to optimize the performance of the classifier, such as feature selection, parameter tuning, model fusion, etc., to ensure that it can maintain high accuracy and robustness in complex and changing operating environments.

[0035] Finally, when the network sealing machine performs real-time operations, the network sealing area recognition module will continuously generate new network sealing area deviation information. This information is sent to the trained area fault diagnosis classifier, which will quickly diagnose and classify it and output the corresponding network sealing area fault diagnosis results. These results can not only help operators quickly understand the status and potential problems of network sealing operations, but also provide strong support for subsequent fault handling and maintenance, and realize intelligent monitoring and evaluation of the quality of network sealing operations.

[0036] The key network blocking area acquisition module 13 is used to index and acquire multiple key network blocking areas whose uniformity information deviates from a preset uniformity interval according to the network blocking uniformity distribution, and index multiple key guide wheels of the multiple key network blocking areas.

[0037] Specifically, the main task of the key network blocking area acquisition module 13 of the present application is to accurately identify the key network blocking areas whose uniformity information significantly deviates from the preset uniformity interval based on the analysis of the network blocking uniformity distribution, and further index the key guide wheels in these areas that may affect the network blocking quality.

[0038] Specifically, the key network blocking area acquisition module 13 first analyzes the entire network blocking operation area using the network blocking uniformity distribution information generated by the network blocking characteristic identification module 11. The network blocking uniformity distribution information is obtained by detecting and analyzing the density of the network blocking area, which reflects the density of network blocking in each area.

[0039] The uniformity distribution information of the network closure is then compared with the preset uniformity interval. If the uniformity information of certain areas deviates from the preset uniformity interval, these areas will be identified as critical network closure areas. The preset uniformity interval is set according to the expected value of the network closure density under ideal conditions, and any deviation will indicate possible problems in the operation.

[0040] In order to locate the problem more accurately, after identifying the multiple key net sealing areas, the module will further index the key guide wheels in these key areas. These guide wheels are key components used to guide and control the net sealing materials when the net sealing machine performs the net sealing task. Their position, state or operation accuracy may have a direct impact on the uniformity of the net sealing. Therefore, by indexing the key guide wheels, the module can provide specific and targeted guidance for subsequent troubleshooting and maintenance.

[0041] In order to achieve this function, the key network sealing area acquisition module 13 can adopt a variety of technical means such as image recognition, data analysis, and machine learning. For example, image recognition technology is used to accurately divide the network sealing area and extract uniformity information, data analysis technology is used to compare the difference between the uniformity information and the preset interval, and machine learning algorithms are used to continuously optimize and improve the recognition accuracy and efficiency. Through the comprehensive application of these technologies, the module can ensure that the key network sealing area and key guide wheels are quickly and accurately identified in a complex and changeable operating environment. This not only helps the system quickly locate potential problem areas in the network sealing operation, but also provides important data support for further fault diagnosis and maintenance work.

[0042] The network blocking damage identification module 14 is used to collect multiple key area images of the multiple key network blocking areas, perform network blocking damage identification, and obtain multiple area network blocking damage identification results.

[0043] Furthermore, the network blocking damage identification module 14 is further configured to perform the following steps:

[0044] P41: Collect multiple key area images of the multiple key network sealing areas; P42: Collect a set of sample area network sealing images based on historical network sealing operation data, and annotate them according to the damage and deformation amplitude of the network sealing in each sample area network sealing image to obtain a set of sample area network sealing damage identification results; P43: Use the sample area network sealing image set and the sample area network sealing damage identification result set to train a network sealing damage identifier; P44: Use the network sealing damage identifier to perform network sealing damage identification on the multiple key area images to obtain multiple area network sealing damage identification results.

[0045] It should be understood that the main function of the network sealing damage identification module 14 of the present application is to collect and analyze images of multiple key network sealing areas, identify possible damage in the network sealing operation, and generate corresponding network sealing damage identification results to provide key information for timely repair and maintenance. First, focus on the multiple key network sealing areas identified by the key network sealing area acquisition module 13. For these areas, the module collects multiple key area images through an integrated high-definition camera or other image acquisition device. These images not only clearly show the surface condition of the network sealing, but also contain sufficient detailed information, providing a reliable data basis for subsequent damage identification.

[0046] Next, in order to train an efficient and accurate network closure damage identifier, the module needs to build a set of high-quality sample data. To this end, the module collected a set of sample area network closure images based on historical network closure operation data. These sample images cover a variety of operating conditions and network closure states, including normal network closure, slight damage, severe damage, and other situations. Subsequently, professional labelers will label the network closure damage deformation amplitude in each sample image and generate a sample area network closure damage identification result set. This set includes various degrees of damage, providing supervision data for subsequent identifier training.

[0047] Furthermore, deep learning or other machine learning techniques are used, with the sample area network blocking image set as input and the sample area network blocking damage identification result set as supervision labels, to train and obtain a network blocking damage identifier. The network blocking damage identifier can automatically extract feature information in the image and accurately determine whether the network blocking is damaged and the degree of damage through continuous learning and optimization.

[0048] Once the network damage identifier is trained, it is applied to multiple key area images for network damage identification. Through complex image processing and feature matching processes, the identifier can quickly and accurately identify the network damage in each key area and generate multiple regional network damage identification results. These results include not only basic information such as the location and size of the damage, but may also include a quantitative assessment of the degree of damage, providing strong support for subsequent repair and maintenance work.

[0049] The movement abnormality identification module 15 is used to collect the network sealing environment information of the current network sealing operation and the line characteristic information on the surface of the transmission line based on multiple sensors, perform movement abnormality identification of the multiple key guide wheels, and obtain multiple guide wheel movement abnormality identification results.

[0050] Furthermore, the movement anomaly identification module 15 is also used to perform the following steps:

[0051] P51: Based on multiple sensors, the humidity information and temperature information of the current network sealing operation are collected as the network sealing environment information; P52: The friction information on the surface of the transmission line is collected; P53: According to the network sealing environment information and friction information, the movement influence anomaly of the multiple key guide wheels is identified, and multiple guide wheel movement anomaly identification results are obtained, wherein a sample network sealing environment information set and a sample friction information set are collected, and according to the magnitude of the influence of each sample network sealing environment information and sample friction information on the guide wheel movement, a sample guide wheel movement anomaly identification result set is obtained by marking, and a guide wheel movement anomaly identifier is trained for identification.

[0052] Optionally, the mobile anomaly identification module 15 of the present application is a key component of the transmission line network sealing operation integrated machine, which is responsible for monitoring and evaluating whether there are any abnormalities in the movement behavior of the key guide wheel in a complex environment during the network sealing operation. In order to ensure the accuracy and real-time performance of the identification, the module makes full use of multi-sensor technology, combines the network sealing environment information and the line feature information on the surface of the transmission line, and conducts an in-depth analysis of the movement status of the key guide wheel.

[0053] Specifically, first, the network sealing environment information of the current network sealing operation is collected through multiple sensors, including humidity information and temperature information. These environmental factors may directly affect the smooth progress of the network sealing operation, especially under high humidity or extreme temperatures, the movement of the guide wheel may be significantly affected. At the same time, the friction information on the surface of the transmission line is collected. Friction is an important factor affecting the movement of the guide wheel. Changes in friction may cause the guide wheel to slip or block, thereby affecting the uniformity and area of ​​the network sealing. The friction information refers to the friction coefficient when the surface of the transmission line contacts the guide wheel, which directly affects the smoothness of the movement of the guide wheel.

[0054] Furthermore, based on the collected network sealing environment information (humidity and temperature) and friction information, the guide wheel movement abnormality identification is performed. By analyzing these data, the system can determine whether the guide wheel has moved abnormally during the operation, such as slipping or unstable movement.

[0055] In order to realize anomaly recognition, it is necessary to train a guide wheel movement anomaly identifier. Specifically, a sample network closure environment information set and a sample friction information set are collected from the historical operation data, and the amplitude of the guide wheel movement influence is marked according to these sample data to form a sample guide wheel movement anomaly recognition result set. This set records the specific manifestations of guide wheel movement anomalies under different environmental conditions and friction conditions, and provides supervision data for the training of subsequent models. Using the guide wheel movement anomaly recognition result set, combined with machine learning or deep learning technology, a guide wheel movement anomaly identifier is trained. Through continuous learning and optimization, this identifier can automatically extract key features from the network closure environment information and friction information, and accurately determine whether there is an abnormality in the guide wheel movement.

[0056] In actual application, when a new network sealing operation is carried out, the movement abnormality recognition module 15 will collect relevant data in real time and input it into the trained recognizer to quickly obtain multiple guide wheel movement abnormality recognition results. These recognition results can not only help operators to find and solve potential problems in time, but also provide valuable reference information for subsequent maintenance and optimization.

[0057] The network sealing fault self-diagnosis module 16 is used to perform network sealing fault self-diagnosis of the transmission line network sealing operation integrated machine according to the network sealing area fault diagnosis result, the multiple area network sealing damage identification results and the multiple guide wheel movement abnormality identification results, and obtain the integrated machine self-diagnosis result.

[0058] Further, such as Figure 3 As shown, the network blocking fault self-diagnosis module 16 is also used to perform the following steps:

[0059] P61: Based on the network sealing damage identification results of the multiple areas, multiple network sealing damage uniformity deviation information is obtained by mapping; P62: Based on the multiple guide wheel movement abnormality identification results, multiple movement abnormality uniformity deviation information is obtained by mapping; P63: Based on the network sealing uniformity distribution, multiple key uniformity deviation information of the multiple key network sealing areas is obtained, and the multiple network sealing damage uniformity deviation information and the multiple movement abnormality uniformity deviation information are respectively subtracted to obtain multiple unknown uniformity deviation information; P64: Based on the network sealing area fault diagnosis results, multiple network sealing damage uniformity deviation information, multiple movement abnormality uniformity deviation information and multiple unknown uniformity deviation information, fault self-diagnosis analysis is performed to obtain fault scale information and fault cause distribution as the all-in-one machine self-diagnosis result.

[0060] Specifically, the main function of the network sealing fault self-diagnosis module 16 of the present application is to comprehensively analyze the network sealing area fault diagnosis results, the network sealing damage identification results of multiple areas, and the abnormal movement identification results of multiple guide wheels, perform comprehensive fault self-diagnosis on the transmission line network sealing operation integrated machine, and finally obtain the self-diagnosis results of the integrated machine.

[0061] First, based on the results of the identification of the damage to the sealing net in multiple regions, the corresponding deviation information of the uniformity of the sealing net damage is mapped out. These deviation information reflects the changes in the uniformity of the sealing net caused by the damage to the sealing net, such as uneven density of the sealing net due to damage or deformation in some areas. Next, based on the results of the identification of multiple guide wheel movement anomalies, the corresponding movement anomaly uniformity deviation information is mapped out, that is, the degree of influence of the guide wheel movement anomaly on the uniform distribution of the sealing net. The movement anomaly of the guide wheel (such as slipping or blocking) will directly affect the uniform distribution of the sealing net, and these deviation information is obtained through the analysis of the guide wheel movement anomaly results.

[0062] Furthermore, according to the uniformity distribution of the network sealing, uniformity deviation information of multiple key network sealing areas is obtained. Then, the uniformity deviation information of network sealing damage and uniformity deviation information of movement anomaly obtained above are respectively subtracted to calculate multiple unknown uniformity deviation information. These unknown deviation information represent the part of the network sealing uniformity change that cannot be directly explained by damage or movement anomaly, which may be caused by other unidentified factors.

[0063] Finally, a comprehensive fault self-diagnosis analysis is performed based on the fault diagnosis results of the network blocking area, the uniformity deviation information of the network blocking damage, the uniformity deviation information of the movement abnormality, and the unknown uniformity deviation information. This analysis process may involve complex algorithms and models, such as rule-based reasoning, Bayesian networks, neural networks, etc., to achieve accurate judgment of the fault scale, fault cause, and its distribution. By integrating this information, the scale of the fault and its cause distribution are determined, and detailed fault scale information and cause distribution are output as the self-diagnosis result of the all-in-one machine. This information is of great significance for guiding operators to quickly locate faults and formulate maintenance plans.

[0064] Furthermore, step P64 of the embodiment of the present application also includes:

[0065] P64-1: Collect a set of sample network blocking area fault diagnosis results, and mark and obtain a set of sample fault scale information to construct a fault scale classifier; P64-2: Input the network blocking area fault diagnosis results into the fault scale classifier and classify to obtain the fault scale information; P64-3: Based on the multiple network blocking damage uniformity deviation information, multiple movement abnormality uniformity deviation information and multiple unknown uniformity deviation information, calculate the fault cause distribution, and combine the fault scale information to obtain the all-in-one self-diagnosis result.

[0066] Optionally, the specific process of performing fault self-diagnosis analysis can be that, in order to accurately classify the fault scale, a large number of sample network closure area fault diagnosis result sets are first collected. These samples cover fault conditions of different scales and types, providing a rich data basis for building a fault scale classifier. Subsequently, the fault scale of each sample is carefully labeled to form a sample fault scale information set. Based on these labeled data, a machine learning algorithm (such as decision tree, random forest, neural network, etc.) is used to train a fault scale classifier. This classifier can learn and identify the characteristic patterns of different fault scales, providing strong support for subsequent classification tasks.

[0067] Furthermore, the fault diagnosis result of the current network closure operation is input into the trained fault scale classifier. The classifier will automatically match the learned feature pattern according to the input diagnosis result and output the corresponding fault scale information to achieve fast and accurate classification of the fault scale.

[0068] Next, based on multiple network blocking damage uniformity deviation information, multiple mobile abnormality uniformity deviation information, and multiple unknown uniformity deviation information, the distribution of fault causes is further calculated. Specifically, these deviation information are first added up to obtain the total network blocking damage uniformity deviation information, the total mobile abnormality uniformity deviation information, and the total uniformity deviation information. Then, the proportion of each of the three in the total uniformity deviation information is calculated. These proportions represent the contribution of different causes to the network blocking uniformity deviation, thereby constituting the distribution of fault causes. Finally, this distribution information is combined with the previously obtained fault scale information to form a complete all-in-one self-diagnosis result. This result not only points out the specific scale and scope of the fault, but also reveals the main causes of the fault and its distribution, providing comprehensive and in-depth guidance for subsequent fault handling and maintenance.

[0069] In summary, the embodiments of the present application have at least the following technical effects:

[0070] The present application collects network sealing images when the power transmission line network sealing operation integrated machine performs the maximum range network sealing operation, identifies the network sealing area and the uniformity of the network sealing, and performs self-diagnosis of network sealing area faults, network sealing damage identification, and abnormal movement of key guide wheels. Finally, based on the network sealing area fault diagnosis results, the network sealing damage identification results of multiple areas, and the abnormal movement identification results of multiple guide wheels, the power transmission line network sealing operation integrated machine performs self-diagnosis of network sealing faults.

[0071] It achieves the goal of integrating multiple information such as the network sealing area, network sealing damage, and abnormal guide wheel movement, and conducts comprehensive fault self-diagnosis to improve the accuracy and comprehensiveness of the diagnosis.

[0072] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0073] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0074] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. Multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system, It is characterized in that The system comprises: A network sealing characteristic identification module, which is used to collect network sealing images when the transmission line network sealing operation integrated machine performs network sealing operation in the maximum range based on multiple sensors, perform network sealing area identification and network sealing uniformity identification, and obtain network sealing area deviation information and network sealing uniformity distribution, wherein the network sealing uniformity distribution includes uniformity information in multiple network sealing areas; A network sealing area fault self-diagnosis module, wherein the network sealing area fault self-diagnosis module is used to perform a network sealing area fault self-diagnosis according to the network sealing area deviation information to obtain a network sealing area fault diagnosis result; A key network blocking area acquisition module, the key network blocking area acquisition module is used to index and acquire multiple key network blocking areas whose uniformity information deviates from a preset uniformity interval according to the network blocking uniformity distribution, and index multiple key guide wheels of the multiple key network blocking areas; A network blocking damage identification module, the network blocking damage identification module is used to collect multiple key area images of the multiple key network blocking areas, perform network blocking damage identification, and obtain multiple area network blocking damage identification results; A movement anomaly recognition module, the movement anomaly recognition module is used to collect the network sealing environment information of the current network sealing operation and the line characteristic information on the surface of the transmission line based on multiple sensors, perform movement anomaly recognition of the multiple key guide wheels, and obtain multiple guide wheel movement anomaly recognition results; A network sealing fault self-diagnosis module is used to perform network sealing fault self-diagnosis of the transmission line network sealing operation integrated machine according to the network sealing area fault diagnosis result, multiple area network sealing damage identification results and multiple guide wheel movement abnormality identification results, and obtain the integrated machine self-diagnosis result.

2. According to the multi-sensor fusion power transmission line sealing operation integrated machine self-diagnosis system of claim 1, It is characterized in that Based on multiple sensors, the network sealing images of the transmission line network sealing machine are collected when the network sealing operation is carried out in the maximum range, and the network sealing area and network sealing uniformity are identified, including: Based on multiple sensors, the network blocking images of the transmission line network blocking machine are collected when the network blocking operation is carried out in the largest range; Training a network blocking identifier for identifying a network blocking area and a network blocking uniformity, wherein the network blocking identifier includes a network blocking area identification branch and a network blocking uniformity identification branch; Using the network blocking identifier, the network blocking image is identified to obtain network blocking area information and network blocking uniformity distribution; The network blocking area deviation information is calculated based on the network blocking area information and the standard network blocking area information of the maximum range of network blocking.

3. The multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system according to claim 2, It is characterized in that Training of network blocking detectors for identifying network blocking area and network blocking uniformity, including: Collecting a sample network sealing image set according to the recorded data of the largest range network sealing operation of the transmission line network sealing operation integrated machine in the historical period; Marking is performed according to the actual network blocking area in each sample network blocking image to obtain a sample network blocking area information set, and marking is performed according to the network blocking density of multiple network blocking areas in each sample network blocking image to obtain a sample network blocking uniformity distribution set; The sample network blocking image set is used as input data, and the sample network blocking area information set and the sample network blocking uniformity distribution set are used as supervision data to train the network blocking area recognition branch and the network blocking uniformity recognition branch; The network blocking area identification branch and the network blocking uniformity identification branch are combined to obtain the network blocking identifier.

4. The multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system according to claim 1, It is characterized in that According to the network sealing area deviation information, a network sealing area fault self-diagnosis is performed to obtain a network sealing area fault diagnosis result, including: Collect the sample network cover area deviation information set, and mark and obtain the sample network cover area fault diagnosis result set according to the size of the sample cover area deviation information; The sample network sealing area deviation information set and the sample network sealing area fault diagnosis result set are used to construct an area fault diagnosis classifier, diagnose and classify the network sealing area deviation information, and obtain the network sealing area fault diagnosis result.

5. The multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system according to claim 1, It is characterized in that Collecting multiple key area images of the multiple key network blocking areas, performing network blocking damage identification, and obtaining multiple area network blocking damage identification results, including: Collecting multiple key area images of the multiple key network blocking areas; According to the historical network sealing operation data, a set of network sealing images of the sample area is collected, and according to the damage and deformation extent of the network sealing in each sample area network sealing image, annotated to obtain a set of network sealing damage identification results of the sample area; Using the sample area network blocking image set and the sample area network blocking damage recognition result set to train a network blocking damage identifier; The network blocking damage identifier is used to perform network blocking damage identification on the multiple key area images to obtain multiple area network blocking damage identification results.

6. The multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system according to claim 1, It is characterized in that Based on multiple sensors, the network sealing environment information of the current network sealing operation and the line feature information on the surface of the transmission line are collected to identify the movement anomalies of the multiple key guide wheels, including: Based on multiple sensors, the humidity and temperature information of the current network blocking operation is collected as network blocking environment information; Collecting friction information on the surface of the transmission line; According to the network sealing environment information and friction information, the movement influence anomaly identification of the multiple key guide wheels is performed to obtain multiple guide wheel movement anomaly identification results, wherein a sample network sealing environment information set and a sample friction information set are collected, and according to the magnitude of the influence of each sample network sealing environment information and sample friction information on the guide wheel movement, a sample guide wheel movement anomaly identification result set is obtained by marking, and a guide wheel movement anomaly identifier is trained for identification.

7. The multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system according to claim 1, It is characterized in that According to the network sealing area fault diagnosis result, the multiple area network sealing damage identification results and the multiple guide wheel movement abnormality identification results, the network sealing fault self-diagnosis of the transmission line network sealing operation integrated machine is performed, including: Mapping and obtaining a plurality of network blocking damage uniformity deviation information according to the plurality of regional network blocking damage identification results; Mapping and obtaining a plurality of movement abnormality uniformity deviation information according to the plurality of guide wheel movement abnormality identification results; According to the network sealing uniformity distribution, a plurality of key uniformity deviation information of the plurality of key network sealing areas are obtained, and the plurality of network sealing damage uniformity deviation information and the plurality of movement abnormality uniformity deviation information are respectively subtracted to obtain a plurality of unknown uniformity deviation information; According to the network blocking area fault diagnosis results, multiple network blocking damage uniformity deviation information, multiple movement abnormality uniformity deviation information and multiple unknown uniformity deviation information, fault self-diagnosis analysis is performed to obtain fault scale information and fault cause distribution as the all-in-one machine self-diagnosis result.

8. The multi-sensor fusion transmission line sealing operation integrated machine self-diagnosis system according to claim 7, It is characterized in that According to the network blocking area fault diagnosis result, multiple network blocking damage uniformity deviation information, multiple movement abnormality uniformity deviation information and multiple unknown uniformity deviation information, a fault self-diagnosis analysis is performed, including: Collect the sample network closure area fault diagnosis result set, annotate and obtain the sample fault scale information set, and build a fault scale classifier; Inputting the network blocking area fault diagnosis result into the fault scale classifier to obtain fault scale information through classification; According to the multiple network blocking damage uniformity deviation information, multiple movement abnormality uniformity deviation information and multiple unknown uniformity deviation information, the fault cause distribution is calculated and combined with the fault scale information to obtain the all-in-one machine self-diagnosis result.

Images