Device detection method and system based on narrowband communication, storage medium and device

Abstract

The embodiment of the invention discloses an equipment detection method based on narrowband communication. The method comprises the following steps: determining to-be-detected power grid equipment without public network coverage in a target area; acquiring an operation state image or video stream of the power grid equipment; performing semantic compression coding on the running state image or video stream of the power grid equipment to generate low-bandwidth compressed data; transmitting the low-bandwidth compressed data to a remote monitoring center by a satellite link through narrow-band satellite communication equipment; the remote monitoring center is used for decoding the received low-bandwidth compressed data and reconstructing a monitoring image of the power grid equipment; and detecting the operation state of the power grid equipment based on the reconstructed monitoring image. According to the method, massive images and video data are converted into semantic information with an extremely low code rate through semantic compression coding, and the problem that a narrow-band satellite channel cannot transmit traditional videos is solved.

Description

Technical Field

[0001] This invention relates to the field of power systems, and more particularly to a device detection method, system, storage medium, and device based on narrowband communication. Background Technology

[0002] With the continuous expansion of power grids, a large number of power grid equipment are distributed in remote mountainous areas, forests, and other geographically complex regions. These areas often lack reliable public network communication coverage, making the monitoring of the operational status of power grid equipment a long-standing management challenge. Currently, the inspection of equipment in these areas without public network coverage mainly relies on manual on-site operations. This method is not only inefficient and costly, but also poses personal safety risks to inspection personnel, and makes it difficult to detect sudden equipment failures in a timely manner.

[0003] Satellite communication technology, especially narrowband satellite communication, offers a potential solution to communication problems in areas without public networks due to its wide signal coverage and strong penetration capabilities. However, narrowband satellite channels inherently have extremely limited bandwidth resources, while the amount of video and image data generated by equipment status monitoring is enormous. Traditional image compression techniques, at extremely low bit rates, can lead to severe image distortion, failing to meet the accuracy requirements for precise identification and analysis of equipment status. Summary of the Invention

[0004] Therefore, it is necessary to propose a device detection method based on narrowband communication to address the above problems.

[0005] A device detection method based on narrowband communication, the method comprising the following steps: Identify the power grid equipment to be tested within the target area that is not covered by public network; Collect images or video streams showing the operating status of the power grid equipment; The operating status images or video streams of the power grid equipment are semantically compressed and encoded to generate low-bandwidth compressed data. The low-bandwidth compressed data is transmitted to a remote monitoring center via a satellite link using narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Based on the reconstructed monitoring images, the operating status of power grid equipment is detected.

[0006] In the above scheme, determining the power grid equipment to be tested within the target area that is not covered by a public network specifically includes: Obtain the geographic location information of all power grid equipment from the power grid asset management system; The geographical location information is compared with the public network communication coverage map to filter out power grid equipment located in the public network coverage blind spot and mark them as equipment to be tested.

[0007] In the above scheme, the step of semantically compressing and encoding the operating status image or video stream of the power grid equipment to generate low-bandwidth compressed data specifically includes: Identify key device components and their status features in the image or video stream, and generate relevant semantic information; The relevant semantic information is quantized and entropy encoded to form low-bandwidth compressed data suitable for narrowband satellite transmission.

[0008] In the above scheme, the key components of the equipment and their condition characteristics include: the thickness of ice covering the conductor, the condition of the insulator, the degree of corrosion of the fittings, and the intrusion of foreign objects.

[0009] In the above scheme, the step of identifying key device components and their state features in the image or video stream and generating relevant semantic information specifically includes: Perform Gaussian filtering to denoise each video frame in the image or video stream; Feature extraction is performed on the denoised image to obtain a high-dimensional feature vector. ; The high-dimensional feature vector Mapped to low-dimensional semantic vectors The mapping transformation relationship is as follows:

[0010] in, For the preset weight matrix, For bias vectors, It is a low-dimensional semantic vector. It is a high-dimensional feature vector; The low-dimensional semantic vector S contains the identifiers of key components of the device and their status feature information, which constitute the relevant semantic information.

[0011] In the above scheme, the quantization and entropy encoding of the relevant semantic information to form low-bandwidth compressed data suitable for narrowband satellite transmission specifically includes: To obtain the real-time bandwidth and packet loss rate of a narrowband satellite communication link, the quantization bit depth is determined using the following formula:

[0012] in, For the number of quantization bits, The preset bitrate adaptation coefficient, For the real-time bandwidth of narrowband satellite communication links, The original code rate of the semantic information. This refers to the packet loss rate of narrowband satellite communication links. Based on the quantization bit depth, for each semantic element in the relevant semantic information Uniform quantization is performed, and the quantization formula is:

[0013] in, The quantized value after quantizing a single semantic element. For the number of quantization bits, For the i-th semantic element, The minimum value of the semantic element. The maximum value of the semantic elements. Preset quantization offset coefficient; Based on the quantized values ​​of all semantic elements Generate a quantized index sequence; Statistical analysis of each quantization index in the quantization index sequence The frequency of occurrence is used to obtain the probability of occurrence of each quantization index. ; Based on the occurrence probability of each quantization index The codeword length is assigned according to the following formula to form low-bandwidth compressed data suitable for narrowband satellite transmission:

[0014] in, This is a coding efficiency adjustment coefficient. Let be the probability of the i-th index appearing.

[0015] In the above scheme, detecting the operating status of power grid equipment based on the reconstructed monitoring images specifically includes: The reconstructed surveillance images are compared with a preset library of normal device status templates. Calculate the feature distance between the reconstructed surveillance image and the most similar normal state template; If the feature distance is greater than a preset abnormal threshold, the power grid equipment is determined to be in an abnormal operating state, and an alarm message is generated.

[0016] This application also proposes a device detection system based on narrowband communication, the system comprising: a device positioning unit, a data acquisition unit, a semantic compression coding unit, a communication transmission unit, and a detection unit; The device positioning unit is used to determine the power grid equipment to be tested that is not covered by the public network within the target area; The data acquisition unit is used to acquire images or video streams of the operating status of the power grid equipment; The semantic compression coding unit is used to perform semantic compression coding on the operating status image or video stream of the power grid equipment to generate low-bandwidth compressed data. The communication transmission unit transmits the low-bandwidth compressed data to the remote monitoring center via a satellite link through narrowband satellite communication equipment. The remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. The detection unit is used to detect the operating status of power grid equipment based on the reconstructed monitoring images.

[0017] This application also proposes a readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps: Identify the power grid equipment to be tested within the target area that is not covered by public network; Collect images or video streams showing the operating status of the power grid equipment; The operating status images or video streams of the power grid equipment are semantically compressed and encoded to generate low-bandwidth compressed data. The low-bandwidth compressed data is transmitted to a remote monitoring center via a satellite link using narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Based on the reconstructed monitoring images, the operating status of power grid equipment is detected.

[0018] This application also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the following steps: Identify the power grid equipment to be tested within the target area that is not covered by public network; Collect images or video streams showing the operating status of the power grid equipment; The operating status images or video streams of the power grid equipment are semantically compressed and encoded to generate low-bandwidth compressed data. The low-bandwidth compressed data is transmitted to a remote monitoring center via a satellite link using narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Based on the reconstructed monitoring images, the operating status of power grid equipment is detected.

[0019] The present invention has the following beneficial effects: By integrating narrowband satellite communication and semantic compression technology, the present invention solves the contradiction between the high bandwidth requirements of video data and the low transmission capacity of narrowband satellites. It can apply the wide coverage advantage of narrowband satellites to video monitoring of power grid equipment, making remote and real-time visual monitoring of equipment status possible and eliminating the dependence on manual on-site inspections. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] in: Figure 1 This is a schematic diagram of a device detection method based on narrowband communication in one embodiment. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the invention; however, it will be apparent to those skilled in the art that the invention may be practiced without one or more of these details; in other instances, certain technical features well-known in the art have not been described in order to avoid confusion with the invention. It should be understood that the invention can be practiced in different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided to make the disclosure thorough and complete and to fully convey the scope of the invention to those skilled in the art. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms, unless the context clearly indicates otherwise. The terms “comprising” and / or “including,” when used in this specification, identify the presence of said features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.

[0024] To fully understand the present invention, a detailed structure will be presented in the following description in order to illustrate the technical solution proposed by the present invention; optional embodiments of the present invention are described in detail below, however, in addition to these detailed descriptions, the present invention may have other embodiments.

[0025] like Figure 1As shown, in one embodiment, a device detection method based on narrowband communication is provided. This device detection method based on narrowband communication includes steps S101 to S105, which are detailed below: S101. Identify the power grid equipment to be tested within the target area that is not covered by a public network; In some embodiments, determining the power grid equipment to be tested that is not covered by a public network within a target area specifically includes: Obtain the geographic location information of all power grid equipment from the power grid asset management system; By comparing geographical location information with the public network communication coverage map, power grid equipment located in the public network coverage blind spot is screened out and marked as equipment to be tested.

[0026] By systematically identifying devices without public network coverage, redundant construction in areas with existing public network coverage is avoided, thus precisely deploying satellite communication resources and monitoring equipment to the areas most in need. This not only improves the targeting and efficiency of the solution implementation but also significantly reduces unnecessary investment costs, achieving optimal resource allocation.

[0027] Specifically, the system automatically exports a list of all power grid equipment requiring monitoring, along with their precise geographical location information, from the power grid company's SAP asset management system via API. Simultaneously, it accesses public network coverage raster map data services provided by telecommunications operators. This data is presented in layers, identifying the 4G / 5G signal strength of different areas, such as urban areas, suburbs, rural areas, and uninhabited areas. In the GIS geographic information system platform, the equipment location layer and the public network coverage layer are overlaid and analyzed. A signal strength threshold is set, automatically filtering out all equipment located in areas with signal strength below this threshold. These devices are marked in red on the system interface, and a list of devices without public network access to be monitored is generated, including equipment ID, equipment name, latitude and longitude, and the line to which they belong.

[0028] Preferably, for the initially screened power grid equipment, a test data packet is sent to it via the public network. If no response is received after three consecutive attempts with a 5-minute interval between each attempt, or if the data packet loss rate exceeds 90%, the equipment is confirmed to be in a state without effective public network coverage, and its status is updated from pending confirmation to pending detection. This effectively avoids misjudgments caused by untimely updates to the coverage map or terrain undulations, improving the accuracy of screening power grid equipment to be tested.

[0029] S102. Acquire images or video streams of the operating status of power grid equipment; Compared to traditional single-sensor data such as temperature and humidity, image and video streams contain massive amounts of multi-dimensional information about the equipment's appearance, structure, and surrounding environment. For example, it can visually show whether the conductor is covered in ice, whether the insulator is damaged, or whether the tower base is tilted. This step transforms the equipment status from abstract numerical data into visual evidence, providing a rich data foundation for subsequent accurate diagnosis and intelligent analysis.

[0030] Preferably, an integrated monitoring PTZ camera is installed on selected power grid equipment in areas without public grid access, such as a transmission tower. This camera is equipped with a 2-megapixel CMOS sensor, a wide-angle lens, and is powered by a 20W solar panel and a battery pack. To conserve energy, the monitoring equipment does not operate continuously. Its operating strategy is set to automatically capture panoramic images of the tower and close-up images of key components at a frequency of one frame per hour during peak daily inspection periods, such as 8-10 AM and 2-4 PM. In case of severe weather, such as when a sudden drop in temperature below 0°C is detected, it automatically switches to icing monitoring mode, increasing the capture frequency to one frame every 15 minutes.

[0031] S103. Perform semantic compression encoding on the operating status images or video streams of power grid equipment to generate low-bandwidth compressed data; Traditional video compression techniques lead to severe image distortion at extremely low bitrates, rendering them unusable for device detection. Semantic compression, however, doesn't simply compress pixels; instead, it uses AI models to understand image content, abstracting high-dimensional pixel information into low-dimensional semantic information. It intelligently retains key information related to device status while removing redundant information irrelevant to detection, such as weather or birds. This reduces video data volume by tens or even hundreds of times, making it suitable for the demanding bandwidth requirements of narrowband satellites and enabling video surveillance transmission over them.

[0032] In some embodiments, semantic compression encoding is performed on images or video streams of the operating status of power grid equipment to generate low-bandwidth compressed data, specifically including: Identify key device components and their status features in images or video streams, and generate relevant semantic information; The relevant semantic information is quantized and entropy encoded to form low-bandwidth compressed data suitable for narrowband satellite transmission.

[0033] Preferably, a lightweight YOLO-fastest model is deployed on an embedded AI device on the tower side for object detection. This model performs real-time inference on the acquired images, identifying key components such as insulators and conductors, and accurately outlining their locations. Subsequently, a lightweight classification network analyzes each outlined component region to determine its state, for example, outputting "insulator: intact" and "conductor: slightly iced." These structured recognition results—component type, location, state description, and confidence level—are then combined into a JSON-formatted text data segment. This JSON data contains the relevant semantic information, and its data volume is compressed hundreds of times compared to the original image.

[0034] In some embodiments, key components of the equipment and their condition characteristics include: conductor icing thickness, insulator damage, hardware corrosion degree, and foreign object intrusion.

[0035] By specifically defining key components and their status characteristics as conductor icing thickness, insulator damage, hardware corrosion, and foreign object intrusion, the AI ​​semantic compression model can prioritize and retain the visual information most relevant to these four core power grid safety hazards during compression. This ensures that the most business-value data is transmitted within limited narrowband bandwidth. Ultimately, the remotely reconstructed images are not only visually similar but also highly faithful in key diagnostic dimensions, enabling monitoring centers to directly and reliably identify typical faults such as icing, damage, corrosion, and foreign objects. This significantly improves the targeting and effectiveness of power grid equipment status monitoring in areas without public grid access.

[0036] In some embodiments, identifying key device components and their state characteristics in an image or video stream, and generating relevant semantic information, specifically includes: Perform Gaussian filtering to denoise each video frame in an image or video stream; Feature extraction is performed on the denoised image to obtain a high-dimensional feature vector. ; High-dimensional feature vectors Mapped to low-dimensional semantic vectors The mapping transformation relationship is as follows:

[0037] in, For the preset weight matrix, For bias vectors, It is a low-dimensional semantic vector. It is a high-dimensional feature vector; The low-dimensional semantic vector S contains the identifiers of key components of the device and their status characteristics, which constitute relevant semantic information.

[0038] Specifically, by employing a Gaussian filtering denoising and then feature extraction to semantic vector mapping technique, high-dimensional pixel data is compressed into low-dimensional semantic vectors containing the identifiers of key equipment components and their status features, achieving a fundamental shift from pixel-level transmission to semantic-level transmission. This process uses learnable weight matrices and bias vectors to perform nonlinear transformations on high-dimensional features, effectively eliminating noise redundancy in the image while concisely retaining key diagnostic information such as conductor icing and insulator damage. This drastically reduces the amount of data transmitted subsequently, while accurately conveying the core semantics of the equipment status, thus achieving both high compression ratios and high diagnostic value information transmission in narrowband channels.

[0039] In some embodiments, relevant semantic information is quantized and entropy encoded to form low-bandwidth compressed data suitable for narrowband satellite transmission, specifically including: To obtain the real-time bandwidth and packet loss rate of a narrowband satellite communication link, the quantization bit depth is determined using the following formula:

[0040] in, For the number of quantization bits, The preset bitrate adaptation coefficient, For the real-time bandwidth of narrowband satellite communication links, The original code rate of the semantic information. This refers to the packet loss rate of narrowband satellite communication links. Based on the quantization bit depth, each semantic element in the relevant semantic information is... Uniform quantization is performed, and the quantization formula is:

[0041] in, The quantized value after quantizing a single semantic element. For the number of quantization bits, For the i-th semantic element, The minimum value of the semantic element. The maximum value of the semantic elements. Preset quantization offset coefficient; Based on the quantized values ​​of all semantic elements Generate a quantized index sequence; Each quantization index in the statistical quantization index sequence The frequency of occurrence is used to obtain the probability of occurrence of each quantization index. ; Based on the occurrence probability of each quantization index The codeword length is assigned according to the following formula to form low-bandwidth compressed data suitable for narrowband satellite transmission:

[0042] in, This is a coding efficiency adjustment coefficient. Let be the probability of the i-th index appearing.

[0043] By introducing a channel-adaptive dynamic quantization and probabilistically optimized entropy coding mechanism, extreme compression of semantic information is achieved in extremely narrowband channels. This scheme first dynamically adjusts the quantization bit depth based on the real-time bandwidth and packet loss rate of the satellite link to ensure the coding rate always matches the channel tolerance. Then, it converts semantic elements into discrete indices through uniform quantization and employs information entropy-optimized variable-length coding based on the probability of index occurrence, ensuring that high-frequency indices correspond to short codewords and low-frequency indices correspond to long codewords. This dual compression strategy achieves data simplification through quantization and eliminates coding redundancy through statistical properties. Ultimately, while maintaining semantic integrity, it compresses the data to less than one-tenth of the original bitrate, completely solving the core bottleneck of transmitting video semantic information via narrowband satellite.

[0044] S104. Low-bandwidth compressed data is transmitted to the remote monitoring center via a satellite link through narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Narrowband satellites, with their wide coverage and strong signal penetration, can reliably transmit data from communication dead zones such as deep mountains and forests back to monitoring centers thousands of miles away, regardless of terrain limitations. Simultaneously, decoding and reconstruction at the monitoring center involves using a supporting AI model to translate the transmitted low-dimensional semantic information back into a visually recognizable monitoring image that is highly similar to the original. This transmission and reception completes a full closed loop of information from the field to the backend, serving as a crucial bridge connecting front-end perception and back-end analysis.

[0045] Preferably, after receiving compressed data from the satellite, the server at the remote monitoring center first decompresses and decodes it to restore the structured semantic information. Then, based on the component type and status description in the semantic information, it retrieves corresponding standard images of normal component status and typical fault status from a pre-set high-definition material library. Using image generation technology, these standard images are fused with the semantic information to generate a reconstructed image that is visually highly similar to the original image, providing a visual reference for maintenance personnel.

[0046] Furthermore, the monitoring center platform has constructed digital twin models of key power grid equipment. When it receives semantic data transmitted from the field, such as a 10% increase in conductor sag, the monitoring center platform synchronously injects this data into the digital twin of the corresponding equipment. The digital twin, combined with real-time environmental data such as wind speed and temperature, performs simulation calculations to predict the development trend of the situation, such as whether the sag will continue to increase and fall below the safe distance under strong wind conditions. The monitoring center platform not only reports current anomalies but also provides a predictive risk assessment report, such as a high risk of discharge tripping within the next 3 hours, thus upgrading from post-event alarms to pre-event warnings.

[0047] S105. Based on the reconstructed monitoring images, detect the operating status of power grid equipment.

[0048] The reconstructed monitoring images are automatically analyzed to determine whether the equipment is in normal condition. Once an anomaly is detected, such as excessive icing or the presence of foreign objects, an alarm is immediately generated. This transforms power grid management from a passive response mode relying on periodic manual inspections to a proactive early warning and intelligent diagnostic mode operating 24 / 7 without human intervention, significantly improving the reliability and safety of power grid operation.

[0049] In some embodiments, detecting the operating status of power grid equipment based on the reconstructed monitoring image specifically includes: The reconstructed surveillance images are compared with a preset library of normal device status templates. Calculate the feature distance between the reconstructed surveillance image and the most similar normal state template; If the feature distance is greater than the preset abnormal threshold, the power grid equipment is determined to be in an abnormal operating state, and an alarm message is generated.

[0050] This application also proposes a device detection system based on narrowband communication, the system comprising: a device positioning unit, a data acquisition unit, a semantic compression coding unit, a communication transmission unit, and a detection unit; The equipment positioning unit is used to locate the power grid equipment to be tested within the target area that is not covered by the public network. The data acquisition unit is used to acquire images or video streams of the operating status of power grid equipment; The semantic compression coding unit is used to perform semantic compression coding on images or video streams of the operating status of power grid equipment to generate low-bandwidth compressed data. The communication transmission unit transmits low-bandwidth compressed data to the remote monitoring center via a satellite link through narrowband satellite communication equipment. The remote monitoring center decodes the received low-bandwidth compressed data and reconstructs the monitoring image of the power grid equipment. The detection unit is used to detect the operating status of power grid equipment based on the reconstructed monitoring images.

[0051] This application also proposes a readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps: Identify the power grid equipment to be tested within the target area that is not covered by public network; Acquire images or video streams showing the operational status of power grid equipment; Semantic compression encoding is performed on images or video streams of the operating status of power grid equipment to generate low-bandwidth compressed data; Low-bandwidth compressed data is transmitted via satellite link to a remote monitoring center through narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Based on the reconstructed monitoring images, the operating status of power grid equipment is detected.

[0052] This application also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: Identify the power grid equipment to be tested within the target area that is not covered by public network; Acquire images or video streams showing the operational status of power grid equipment; Semantic compression encoding is performed on images or video streams of the operating status of power grid equipment to generate low-bandwidth compressed data; Low-bandwidth compressed data is transmitted via satellite link to a remote monitoring center through narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Based on the reconstructed monitoring images, the operating status of power grid equipment is detected.

[0053] Those skilled in the art will understand that implementing all or part of the processes in the above embodiments can be accomplished by instructing related hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0054] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] The embodiments described above are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application's patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. The embodiments disclosed above are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made according to the claims of this invention are still within the scope of this invention.

Claims

1. A device detection method based on narrowband communication, characterized in that, The method includes: Identify the power grid equipment to be tested within the target area that is not covered by public network; Collect images or video streams of the operating status of the power grid equipment; The operating status images or video streams of the power grid equipment are semantically compressed and encoded to generate low-bandwidth compressed data. The low-bandwidth compressed data is transmitted to a remote monitoring center via a satellite link using narrowband satellite communication equipment; the remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. Based on the reconstructed monitoring images, the operating status of power grid equipment is detected.

2. The device detection method based on narrowband communication according to claim 1, characterized in that, The determination of the power grid equipment to be tested within the target area that is not covered by public network specifically includes: Obtain the geographic location information of all power grid equipment from the power grid asset management system; The geographical location information is compared with the public network communication coverage map to filter out power grid equipment located in the public network coverage blind spot and mark them as equipment to be tested.

3. The device detection method based on narrowband communication according to claim 2, characterized in that, The step of semantically compressing and encoding the operating status images or video streams of the power grid equipment to generate low-bandwidth compressed data specifically includes: Identify key device components and their status features in the image or video stream, and generate relevant semantic information; The relevant semantic information is quantized and entropy encoded to form low-bandwidth compressed data suitable for narrowband satellite transmission.

4. The device detection method based on narrowband communication according to claim 3, characterized in that, The key components of the equipment and their condition characteristics include: the thickness of ice covering the conductor, the condition of the insulator, the degree of corrosion of the fittings, and the intrusion of foreign objects.

5. The device detection method based on narrowband communication according to claim 3, characterized in that, The process of identifying key device components and their status features in the image or video stream, and generating relevant semantic information, specifically includes: Perform Gaussian filtering to denoise each video frame in the image or video stream; Feature extraction is performed on the denoised image to obtain a high-dimensional feature vector. ; The high-dimensional feature vector Mapped to low-dimensional semantic vectors The mapping transformation relationship is as follows: in, For the preset weight matrix, For bias vectors, It is a low-dimensional semantic vector. It is a high-dimensional feature vector; The low-dimensional semantic vector S contains the identifiers of key components of the device and their status feature information, which constitute the relevant semantic information.

6. The device detection method based on narrowband communication according to claim 5, characterized in that, The process of quantizing and entropy encoding the relevant semantic information to form low-bandwidth compressed data suitable for narrowband satellite transmission specifically includes: To obtain the real-time bandwidth and packet loss rate of a narrowband satellite communication link, the quantization bit depth is determined using the following formula: in, For the number of quantization bits, The preset bitrate adaptation coefficient, This refers to the real-time bandwidth of narrowband satellite communication links. The original bitrate of the semantic information. This refers to the packet loss rate of narrowband satellite communication links. Based on the quantization bit depth, for each semantic element in the relevant semantic information Uniform quantization is performed, and the quantization formula is: in, The quantized value after quantizing a single semantic element. For the number of quantization bits, For the i-th semantic element, The minimum value of the semantic element. The maximum value of the semantic elements. Preset quantization offset coefficient; Based on the quantized values ​​of all semantic elements Generate a quantized index sequence; Statistical analysis of each quantization index in the quantization index sequence The frequency of occurrence is used to obtain the probability of occurrence of each quantization index. ; Based on the occurrence probability of each quantization index The codeword length is assigned according to the following formula to form low-bandwidth compressed data suitable for narrowband satellite transmission: in, This is a coding efficiency adjustment coefficient. Let be the probability of the i-th index appearing.

7. The device detection method based on narrowband communication according to claim 1, characterized in that, The detection of the operating status of power grid equipment based on the reconstructed monitoring images specifically includes: The reconstructed surveillance images are compared with a preset library of normal device status templates. Calculate the feature distance between the reconstructed surveillance image and the most similar normal state template; If the feature distance is greater than a preset abnormal threshold, the power grid equipment is determined to be in an abnormal operating state, and an alarm message is generated.

8. A device detection system based on narrowband communication, characterized in that, The system includes: a device positioning unit, a data acquisition unit, a semantic compression encoding unit, a communication transmission unit, and a detection unit; The device positioning unit is used to determine the power grid equipment to be tested that is not covered by the public network within the target area; The data acquisition unit is used to acquire images or video streams of the operating status of the power grid equipment; The semantic compression coding unit is used to perform semantic compression coding on the operating status image or video stream of the power grid equipment to generate low-bandwidth compressed data. The communication transmission unit transmits the low-bandwidth compressed data to the remote monitoring center via a satellite link through narrowband satellite communication equipment. The remote monitoring center is used to decode the received low-bandwidth compressed data and reconstruct the monitoring image of the power grid equipment. The detection unit is used to detect the operating status of power grid equipment based on the reconstructed monitoring images.

9. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the processor performs the steps of the method as described in any one of claims 1 to 7.

10. A computer device, comprising a memory and a processor, characterized in that, The memory stores a computer program that, when executed by the processor, causes the processor to perform the steps of the method as described in any one of claims 1 to 7.

Images