Power grid system disaster damage detection method and device, terminal and medium
Through real-time detection and model identification of drones, the location and type of disaster loss of power grids are quickly positioned, which solves the problem of low efficiency of disaster loss detection in the existing technology, and realizes efficient resource allocation and rapid recovery of the power grid system.
Patent Information
- Application Number
- CN202411097616.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-12
- Publication Date
- 2024-11-15
AI Technical Summary
The existing technology is inefficient in power grid disaster loss detection, making it difficult to achieve comprehensive and real-time analysis and evaluation. Manual inspections are easily affected by the environment, and the fault indicator cannot intuitively present the fault type and occurrence time.
The drone flight path is planned using a preset path planning algorithm, and real-time detection is performed using the carrying camera and infrared sensor. Combined with the disaster loss identification model and the ARIMA model, it can quickly locate the disaster loss location, identify the disaster loss type and predict the occurrence time.
It improves the efficiency of disaster loss detection in the power grid, shortens the response time, helps optimize resource allocation and operation and maintenance decisions, and improves the grid recovery efficiency and emergency management capabilities.
Smart Images

Figure CN118965116A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of power grid disaster damage detection, and in particular to a power grid system disaster damage detection method, device, terminal and medium. Background Art
[0002] Power grid system damage refers to the phenomenon that the related facilities of the power grid system (including transmission lines, distribution facilities, substations, etc.) are damaged or destroyed under natural disasters or other adverse conditions, resulting in power transmission interruption or serious impact on power supply. Power grid system damage may be caused by a variety of reasons, the most common of which is extreme weather conditions such as typhoons, heavy rains, blizzards, lightning, etc. These natural disasters may cause physical damage to power grid facilities, such as line breakage, tower collapse, equipment damage, etc.
[0003] At present, in the process of power grid disaster management, the fault and disaster detection of the power grid system is mainly achieved by means such as regular manual inspections and fault indicators built into the equipment. However, manual inspections are inefficient and easily affected by environmental conditions. Equipment at some nodes is difficult to detect manually, and the detection accuracy is low. The fault indicator can only provide the fault location, but cannot intuitively present the fault type and the specific time when the fault occurs. Therefore, it is difficult to conduct a comprehensive and real-time analysis and evaluation of the power grid system based on regular manual inspections and fault indicators built into the equipment, which is not conducive to managers or intelligent systems to make targeted maintenance decisions in a timely manner. Summary of the invention
[0004] The embodiments of the present invention provide a method, device, terminal and medium for detecting disaster damage in a power grid system, which can quickly locate the disaster damage position of the power grid system, and accurately determine the type of disaster damage and the time when the disaster damage occurred, thereby improving the efficiency of disaster damage detection.
[0005] In order to solve the above technical problems, an embodiment of the present invention provides a power grid system damage detection method, comprising:
[0006] Use the preset path planning algorithm to plan the flight path of the drone and generate a path plan;
[0007] Using the drone carrying the camera and the infrared sensor, the power grid system in the target area is photographed and detected in real time according to the path plan to obtain the power grid data to be identified;
[0008] Inputting the to-be-identified power grid data into a disaster damage identification model, so that the disaster damage identification model extracts and identifies features of the to-be-identified power grid data and outputs power grid disaster damage type and disaster damage location information corresponding to the target area;
[0009] The to-be-identified power grid data is input into the ARI MA model, so that the ARI MA model predicts the power grid load change for the to-be-identified power grid data and outputs the disaster occurrence time corresponding to the target area.
[0010] In the implementation of the embodiment of the present invention, in the data collection stage, a preset path planning algorithm is first used to plan the flight path of the drone and generate a path plan. Then, a drone carrying a camera and an infrared sensor is used to perform real-time shooting and detection of the power grid system in the target area according to the path plan, so that the power grid data to be identified in the power grid system can be quickly obtained, so that the abnormal conditions and potential disasters of the power grid system terminals can be discovered in time based on the power grid data to be identified in the power grid system, shortening the overall response time and improving the disaster damage processing efficiency. Then, through the disaster damage identification model, the power grid data to be identified in the power grid system is feature extracted and identified to obtain the power grid disaster damage type and disaster damage location information corresponding to the target area, and the power grid data to be identified is input into the ARI MA model at the same time, so that the ARI MA model predicts the power grid load change of the power grid data to be identified and outputs the disaster damage occurrence time corresponding to the target area. Based on the above-mentioned power grid disaster damage type, disaster damage location information and disaster damage occurrence time, it can help dispatchers or intelligent management platforms optimize the resource allocation and operation and maintenance decisions of the power grid system, centrally solve disasters, and improve the recovery efficiency and emergency management capabilities of the power grid system.
[0011] As a preferred solution, the drone carrying the camera and the infrared sensor is used to perform real-time photography and detection of the power grid system in the target area according to the path plan to obtain the power grid data to be identified, specifically:
[0012] Using the drone to patrol the power grid system according to the path plan, and during the patrol process, using the camera carried by the drone to photograph the power grid system in real time to obtain image data of the power grid system, and using the infrared sensor carried by the drone to perform infrared detection on the power grid system to obtain temperature data of the power grid system;
[0013] Using a sensor network deployed in the power grid system, the voltages of multiple power grid nodes in the target area are detected to generate a voltage distribution diagram of the power grid system; wherein the sensor network includes multiple voltage sensors;
[0014] The image data, the temperature data and the voltage distribution diagram are fused and preprocessed to obtain the data to be transmitted, and then the data to be transmitted is transmitted to the central control module. Then, the central control module analyzes and processes the received data to be transmitted to obtain the power grid data to be identified.
[0015] The preferred scheme of implementing the embodiment of the present invention collects image and temperature data through the camera and infrared sensor of the drone, and uses the sensor network deployed in the power grid system to detect the voltage data of multiple power grid nodes in the target area. It is possible to obtain various operating status information of the power grid system from multiple aspects, and then fuse and pre-process the data of the above multiple dimensions, so as to comprehensively analyze the overall operating status of the power grid system, and then the central control module identifies potential problems in the power grid system.
[0016] As a preferred solution, the image data, the temperature data and the voltage distribution diagram are fused and preprocessed to obtain the data to be transmitted, and then the data to be transmitted is transmitted to the central control module, and then the central control module analyzes and processes the received data to be transmitted to obtain the power grid data to be identified, specifically:
[0017] fusing the image data, the temperature data and the voltage distribution map to obtain fused data;
[0018] Using compressed sensing technology and encryption technology, compressing and encrypting the fused data to obtain data to be transmitted;
[0019] The data to be transmitted is divided into a plurality of first data, and then the first data are transmitted to a central control module through a plurality of network paths according to a preset multi-path transmission strategy;
[0020] When the central control module receives the first data, it decrypts and reorganizes all the received first data to obtain the to-be-identified power grid data corresponding to the target area.
[0021] The preferred solution of the embodiment of the present invention is to apply compressed sensing technology and encryption technology to compress and encrypt the fused data obtained by fusing the image data, temperature data and voltage distribution map before transmitting the data, thereby reducing the amount of transmitted data and preventing the data from being tampered with or stolen during the transmission process. In addition, for the transmission of data, a multi-path transmission strategy is designed to divide the data into multiple first data, and then synchronously transmit each first data to the central control module through multiple different network paths to ensure reliable transmission of data in the event of a disaster.
[0022] As a preferred solution, the power grid system damage detection method further includes:
[0023] Storing the to-be-identified power grid data output by the central control module into a distributed database;
[0024] By means of a data backup module, the to-be-identified power grid data output by the central control module is backed up regularly according to a preset data backup cycle;
[0025] When data loss is detected in the distributed database, the power grid data to be identified stored in the distributed database is compared with the power grid data to be identified in the data backup module, and based on the comparison result, the power grid data to be identified stored in the distributed database is filled and restored.
[0026] A preferred scheme for implementing the embodiment of the present invention is to perform regular backup of the power grid data to be identified output by the central control module, and when data loss is detected in the distributed database, the power grid data to be identified stored in the distributed database is compared with the power grid data to be identified in the data backup module, and based on the comparison result, the power grid data to be identified stored in the distributed database is filled and restored to achieve rapid data recovery and ensure the normal operation of the power grid system and the integrity of the data.
[0027] As a preferred solution, the preset path planning algorithm is used to plan the flight path of the UAV and generate a path plan, specifically:
[0028] The A* algorithm is used to plan the flight path of the UAV and generate a path plan in combination with the equipment distribution and obstacle characteristics of the power grid system;
[0029] Alternatively, the Dijkstra algorithm is used to plan the flight path of the UAV and generate a path plan in combination with the equipment distribution and obstacle characteristics of the power grid system.
[0030] In the preferred embodiment of the present invention, the A* algorithm can quickly generate the optimal flight path by combining the estimated cost of the current state and the target state through heuristic search, while the Dijkstra algorithm gradually expands to the shortest path when processing all possible paths, and is suitable for weightless or relatively simple environments, and can ensure the optimality of the path while finding the shortest path.
[0031] As a preferred solution, the construction of the disaster damage identification model is specifically as follows:
[0032] Based on the convolutional neural network, an initial model is constructed; wherein the initial model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer and an output layer, the convolutional layer is used to extract features from the input data, the pooling layer is used to perform maximum pooling processing on the feature extraction results, and the fully connected layer is used to perform classification and recognition processing on the feature extraction results after pooling;
[0033] The initial model is trained and optimized multiple times using a sample data set, and during the training and optimization process, the cross entropy loss function and the mean square error loss function are used to calculate the loss value of the current initial model in real time until the loss value of the current initial model is less than a first preset value or the current number of training optimizations is equal to a second preset value, and the current initial model is used as a disaster damage identification model; wherein the sample data set includes power grid data at several disaster damage moments, and power grid disaster damage types and disaster damage location information corresponding to each of the disaster damage moments, and the disaster damage moment is the moment when a disaster occurs in the power grid system.
[0034] A preferred solution for implementing the embodiment of the present invention adopts the cross entropy loss function and the mean square error loss function in the process of model training optimization to optimize the classification task and regression task of the model respectively, so that the model is more flexible and accurate in processing different types of tasks, and through continuous training and adjustment, the model can gradually learn the inherent laws and characteristics of power grid faults, thereby improving the accuracy of fault identification of the model.
[0035] In order to solve the same technical problem, an embodiment of the present invention further provides a power grid system damage detection device, comprising:
[0036] The path planning module is used to plan the flight path of the drone and generate a path plan using a preset path planning algorithm;
[0037] A shooting and detection module is used to use the drone carrying the camera and the infrared sensor to perform real-time shooting and detection of the power grid system in the target area according to the path plan to obtain the power grid data to be identified;
[0038] A disaster damage detection module is used to input the power grid data to be identified into a disaster damage identification model so that the disaster damage identification model extracts and identifies features of the power grid data to be identified and outputs the power grid disaster type and disaster damage location information corresponding to the target area; inputs the power grid data to be identified into an ARI MA model so that the ARI MA model predicts the power grid load changes for the power grid data to be identified and outputs the disaster occurrence time corresponding to the target area.
[0039] As a preferred solution, the shooting detection module specifically includes:
[0040] An inspection unit is used to use the drone to inspect the power grid system according to the path plan, and during the inspection process, use the camera carried by the drone to take real-time photos of the power grid system to obtain image data of the power grid system, and use the infrared sensor carried by the drone to perform infrared detection on the power grid system to obtain temperature data of the power grid system;
[0041] A voltage detection unit, configured to detect the voltages of a plurality of grid nodes in the target area using a sensor network deployed in the grid system, so as to generate a voltage distribution diagram of the grid system; wherein the sensor network includes a plurality of voltage sensors;
[0042] A fusion analysis unit is used to fuse and preprocess the image data, the temperature data and the voltage distribution diagram to obtain the data to be transmitted, and then transmit the data to be transmitted to the central control module, and then analyze and process the received data to be transmitted through the central control module to obtain the power grid data to be identified.
[0043] In order to solve the same technical problem, the present invention also provides a terminal, including a processor, a memory and a computer program stored in the memory; wherein the computer program can be executed by the processor to implement the power grid system disaster damage detection method.
[0044] In order to solve the same technical problem, the present invention also provides a computer-readable storage medium, which includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the method for detecting disaster damage in a power grid system. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] Figure 1 : A schematic diagram of a flow chart of a method for detecting damage in a power grid system provided in Embodiment 1 of the present invention;
[0046] Figure 2 : A schematic structural diagram of a power grid system disaster damage detection device provided in Example 1 of the present invention. DETAILED DESCRIPTION
[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0048] Embodiment one:
[0049] Please refer to Figure 1 , is a power grid system damage detection method provided by an embodiment of the present invention, the method includes steps S1 to S4, each step is specifically as follows:
[0050] Step S1, using a preset path planning algorithm to plan the flight path of the UAV and generate a path plan.
[0051] As a preferred solution, step S1 is step S11 or step S12, and each step is specifically as follows:
[0052] Step S11, using the A* algorithm, combined with the equipment distribution and obstacle characteristics of the power grid system, plans the flight path of the drone and generates a path plan.
[0053] Step S12, using the Dijkstra algorithm, combined with the equipment distribution and obstacle characteristics of the power grid system, plans the flight path of the UAV and generates a path plan.
[0054] In this embodiment, the optimal flight path can be planned in combination with the geographical location and surrounding environment of the power grid equipment to ensure that the data collection range of the camera and infrared sensor carried by the drone can cover all key areas.
[0055] Step S2, using a drone carrying a camera and an infrared sensor, according to a path plan, real-time photography and detection of the power grid system in the target area are performed to obtain power grid data to be identified.
[0056] As a preferred solution, step S2 includes step S21 to step S23, and each step is specifically as follows:
[0057] Step S21, using a drone to patrol the power grid system according to the path plan, and during the patrol process, using a camera with high-definition imaging capability carried by the drone to photograph the power grid system in real time to obtain image data of the power grid system, and at the same time using an infrared sensor carried by the drone to perform infrared detection on the power grid system to obtain temperature data of the power grid system.
[0058] It should be noted that drones carrying cameras and infrared sensors can also collect the following data: the structural status of power grid facilities, such as the inclination of towers and conductor tension; environmental parameters, such as wind speed, wind direction, temperature, humidity, etc.; disaster damage, such as the spread of fires, line interruptions caused by fallen trees, etc.; video and image data for intuitive disaster damage assessment and on-site condition recording. And based on the fusion of the above data and voltage distribution diagrams, the grid data to be identified for disaster damage detection is formed.
[0059] Step S22, using the sensor network deployed in the power grid system to detect the voltage of multiple power grid nodes in the target area to generate a voltage distribution map of the power grid system.
[0060] The sensor network includes a plurality of voltage sensors with self-calibration and self-learning functions.
[0061] In this embodiment, a sensor network is deployed in the power grid system, so that voltage sensors can exchange data and work together, and potential fault areas can be quickly located through the collaborative work.
[0062] Step S23, the image data, temperature data and voltage distribution diagram are integrated and preprocessed to obtain the data to be transmitted, and then the data to be transmitted is transmitted to the central control module, and then the central control module analyzes and processes the received data to be transmitted to obtain the power grid data to be identified.
[0063] In this embodiment, near field communication technology (NFC) is used in the data transmission process to improve the security and reliability of data transmission.
[0064] As a preferred solution, step S23 includes step S231 to step S234, and each step is specifically as follows:
[0065] Step S231, fusing the image data, the temperature data and the voltage distribution diagram to obtain fused data.
[0066] Step S232, using compressed sensing technology and encryption technology, compressing and encrypting the fused data to obtain data to be transmitted.
[0067] In this embodiment, compressed sensing technology is applied before data transmission to efficiently compress data and reduce the amount of transmitted data. Encryption technology is applied during data compression to encrypt data, which can ensure the security of data transmission. Among them, encryption technology includes but is not limited to: symmetric encryption algorithms such as AES (Advanced Encryption Standard) and DES (Data Encryption Standard), asymmetric encryption algorithms such as RSA (a public key cryptographic system) and ECC (elliptic curve cryptography), hash functions and hash algorithms such as SHA-256 (Secure Hash Algorithm) and MD5 (Message Digest Algorithm), and digital signature technology. In addition, Gzip can also be used to achieve data compression to save storage space and improve data management efficiency.
[0068] Step S233 , dividing the data to be transmitted into a plurality of first data, and then transmitting each first data to the central control module through a plurality of network paths according to a preset multi-path transmission strategy.
[0069] In this embodiment, the Internet, satellite communications, and dedicated networks may be used for redundant transmission to ensure that data can reach the destination under any circumstances.
[0070] Step S234, when the central control module receives the first data, it decrypts and reorganizes all the received first data to obtain the to-be-identified power grid data corresponding to the target area.
[0071] As a preferred solution, a power grid system disaster detection method provided in an embodiment of the present invention further includes a data backup and recovery process, which includes steps S01 to S03, and each step is specifically as follows:
[0072] Step S01, storing the to-be-identified power grid data output by the central control module into a distributed database.
[0073] In this embodiment, distributed storage technologies such as HDFS are used to implement data storage, which can improve the capacity and reliability of data storage.
[0074] Step S02: The data backup module performs a regular backup of the to-be-identified power grid data output by the central control module according to a preset data backup cycle.
[0075] Step S03, when data loss is detected in the distributed database, the power grid data to be identified stored in the distributed database is compared with the power grid data to be identified in the data backup module, and the power grid data to be identified stored in the distributed database is filled and restored according to the comparison result.
[0076] Step S3, inputting the power grid data to be identified into the disaster damage identification model, so that the disaster damage identification model extracts and identifies the features of the power grid data to be identified and outputs the power grid disaster damage type and disaster damage location information corresponding to the target area.
[0077] As a preferred solution, a power grid system disaster detection method provided in an embodiment of the present invention further includes a disaster identification model construction process, which includes steps S04 to S05, and each step is specifically as follows:
[0078] Step S04: construct an initial model based on a convolutional neural network.
[0079] Among them, the initial model includes input layer, convolution layer, pooling layer, fully connected layer and output layer. The convolution layer is used to extract features of the input data, the pooling layer is used to perform maximum pooling processing on the feature extraction results, and the fully connected layer is used to classify and recognize the feature extraction results after pooling.
[0080] In this embodiment, the initial model is ResNet50, which has a deep network structure and residual learning mechanism, and can efficiently extract the features of power grid data. The processing process of the convolution layer of the model is specifically shown in formula (1), and the processing process of the pooling layer of the model is specifically shown in formula (2).
[0081]
[0082] In the formula, f i,j,k is the value at position (i, j) of the convolutional layer output feature map, x is the input image, ω is the convolution kernel, and b k is bias.
[0083]
[0084] In the formula, p i,j,k It is the feature map after pooling. The pooling layer here is used to achieve maximum pooling.
[0085] Step S05, using the sample data set, the initial model is trained and optimized multiple times, and during the training and optimization process, the cross entropy loss function and the mean square error loss function are used to calculate the loss value of the current initial model in real time until the loss value of the current initial model is less than the first preset value or the current number of training optimizations is equal to the second preset value, then the current initial model is used as the disaster loss identification model.
[0086] The sample data set includes power grid data at several disaster damage moments, as well as power grid disaster damage type and disaster damage location information corresponding to each disaster damage moment. The disaster damage moment is the moment when the disaster occurs in the power grid system.
[0087] Step S4, inputting the to-be-identified power grid data into the ARIMA model, so that the ARIMA model predicts the power grid load change for the to-be-identified power grid data and outputs the disaster occurrence time corresponding to the target area.
[0088] In this embodiment, the model parameters of the above two models are updated in real time using the online stochastic gradient descent (SGD) algorithm to improve the processing accuracy. In addition, by introducing a reinforcement learning algorithm (such as the Q-learning algorithm) in the central control module, the optimal decision is made based on historical data and current status to ensure that the power grid resumes normal operation in the shortest time and reduce the impact of disasters on the power grid system.
[0089] It should be noted that after obtaining the type of power grid damage, location information and time of damage in the target area, visualization tools such as D3.js or Tableau are used to provide an intuitive data display interface to intuitively present the relevant data of power grid system damage to assist in decision-making. In addition, natural language processing technology is introduced to support users to query and analyze data through natural language.
[0090] Please refer to Figure 2 , is a schematic diagram of the structure of a power grid system damage detection device provided by an embodiment of the present invention, the device includes a path planning module M1, a shooting detection module M2 and a damage detection module M3, and the specific details of each module are as follows:
[0091] The path planning module M1 is used to plan the flight path of the UAV and generate a path plan using a preset path planning algorithm;
[0092] The shooting and detection module M2 is used to use a drone carrying a camera and an infrared sensor to perform real-time shooting and detection of the power grid system in the target area according to the path plan to obtain the power grid data to be identified;
[0093] The disaster damage detection module M3 is used to input the power grid data to be identified into the disaster damage identification model, so that the disaster damage identification model can extract and identify the features of the power grid data to be identified and output the power grid disaster type and disaster location information corresponding to the target area; the power grid data to be identified is input into the ARI MA model, so that the ARI MA model can predict the power grid load changes of the power grid data to be identified and output the disaster occurrence time corresponding to the target area.
[0094] As a preferred solution, the shooting detection module M2 specifically includes a patrol unit 21, a voltage detection unit 22 and a fusion analysis unit 23, and the specific details of each unit are as follows:
[0095] The inspection unit 21 is used to inspect the power grid system using a drone according to a path plan, and during the inspection process, use a camera carried by the drone to take real-time photos of the power grid system to obtain image data of the power grid system, and use an infrared sensor carried by the drone to perform infrared detection on the power grid system to obtain temperature data of the power grid system;
[0096] A voltage detection unit 22 is used to detect the voltage of multiple grid nodes in a target area using a sensor network deployed in the grid system to generate a voltage distribution diagram of the grid system; wherein the sensor network includes multiple voltage sensors;
[0097] The fusion analysis unit 23 is used to fuse and preprocess the image data, temperature data and voltage distribution diagram to obtain the data to be transmitted, and then transmit the data to be transmitted to the central control module. Then, the central control module analyzes and processes the received data to be transmitted to obtain the power grid data to be identified.
[0098] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0099] An embodiment of the present invention further provides a computer-readable storage medium, the computer-readable storage medium including a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a power grid system disaster detection method of embodiment 1.
[0100] An embodiment of the present invention further provides a terminal, including a processor, a memory, and a computer program stored in the memory; wherein the computer program can be executed by the processor to implement a power grid system disaster damage detection method of embodiment 1.
[0101] Preferably, the computer program can be divided into one or more modules / units (such as computer program, computer program), one or more modules / units are stored in the memory and executed by the processor to complete the present invention. One or more modules / units can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal.
[0102] The processor can be a central processing unit (CPU), or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can also be any conventional processor. The processor is the control center of the terminal, and various parts of the terminal are connected using various interfaces and lines.
[0103] The memory mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (Smart Med iaCard, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory can also be other volatile solid-state storage devices.
[0104] It should be noted that the above-mentioned terminal may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the above-mentioned terminal is merely an example and does not constitute a limitation on the terminal. It may include more or fewer components, or a combination of certain components, or different components.
[0105] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0106] The present invention provides a method, device, terminal and medium for detecting disaster damage in a power grid system. In the data collection stage, a preset path planning algorithm is first used to plan the flight path of a drone and generate a path plan. Then, a drone carrying a camera and an infrared sensor is used to perform real-time shooting and detection of the power grid system in the target area according to the path plan, so that the power grid data to be identified in the power grid system can be quickly obtained, so that the abnormal conditions and potential disasters of the power grid system terminal can be discovered in time based on the power grid data to be identified in the power grid system, shortening the overall response time and improving the disaster damage processing efficiency. Then, through the disaster damage identification model, the power grid data to be identified in the power grid system is feature extracted and identified to obtain the power grid disaster damage type and disaster damage location information corresponding to the target area, and the power grid data to be identified is input into the ARIMA model at the same time, so that the ARIMA model predicts the power grid load change of the power grid data to be identified and outputs the disaster damage occurrence time corresponding to the target area. Based on the above-mentioned power grid disaster damage type, disaster damage location information and disaster damage occurrence time, it can help dispatchers or intelligent management platforms optimize the resource allocation and operation and maintenance decision-making of the power grid system, centrally solve disasters, and improve the recovery efficiency and emergency management capabilities of the power grid system.
[0107] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A method for detecting power grid system damage, characterized in that: include: Use the preset path planning algorithm to plan the flight path of the drone and generate a path plan; Using the drone carrying the camera and the infrared sensor, the power grid system in the target area is photographed and detected in real time according to the path plan to obtain the power grid data to be identified; Inputting the to-be-identified power grid data into a disaster damage identification model, so that the disaster damage identification model extracts and identifies features of the to-be-identified power grid data and outputs power grid disaster damage type and disaster damage location information corresponding to the target area; The to-be-identified power grid data is input into the ARIMA model, so that the ARIMA model predicts the power grid load change for the to-be-identified power grid data and outputs the disaster occurrence time corresponding to the target area.
2. A power grid system damage detection method according to claim 1, characterized in that: The drone carrying the camera and the infrared sensor is used to perform real-time photography and detection of the power grid system in the target area according to the path plan to obtain the power grid data to be identified, specifically: Using the drone to patrol the power grid system according to the path plan, and during the patrol process, using the camera carried by the drone to photograph the power grid system in real time to obtain image data of the power grid system, and using the infrared sensor carried by the drone to perform infrared detection on the power grid system to obtain temperature data of the power grid system; Using a sensor network deployed in the power grid system, the voltages of multiple power grid nodes in the target area are detected to generate a voltage distribution diagram of the power grid system; wherein the sensor network includes multiple voltage sensors; The image data, the temperature data and the voltage distribution diagram are fused and preprocessed to obtain the data to be transmitted, and then the data to be transmitted is transmitted to the central control module. Then, the central control module analyzes and processes the received data to be transmitted to obtain the power grid data to be identified.
3. A power grid system damage detection method as claimed in claim 2, characterized in that: The image data, the temperature data and the voltage distribution diagram are fused and preprocessed to obtain the data to be transmitted, and then the data to be transmitted is transmitted to a central control module, and then the central control module analyzes and processes the received data to be transmitted to obtain the power grid data to be identified, specifically: fusing the image data, the temperature data and the voltage distribution map to obtain fused data; Using compressed sensing technology and encryption technology, compressing and encrypting the fused data to obtain data to be transmitted; The data to be transmitted is divided into a plurality of first data, and then the first data are transmitted to a central control module through a plurality of network paths according to a preset multi-path transmission strategy; When the central control module receives the first data, it decrypts and reorganizes all the received first data to obtain the to-be-identified power grid data corresponding to the target area.
4. A power grid system damage detection method as claimed in claim 2, characterized in that: Also includes: Storing the to-be-identified power grid data output by the central control module into a distributed database; By means of a data backup module, the to-be-identified power grid data output by the central control module is backed up regularly according to a preset data backup cycle; When data loss is detected in the distributed database, the power grid data to be identified stored in the distributed database is compared with the power grid data to be identified in the data backup module, and based on the comparison result, the power grid data to be identified stored in the distributed database is filled and restored.
5. A power grid system damage detection method as claimed in claim 1, characterized in that: The preset path planning algorithm is used to plan the flight path of the UAV and generate a path plan, specifically: The A* algorithm is used to plan the flight path of the UAV and generate a path plan in combination with the equipment distribution and obstacle characteristics of the power grid system; Alternatively, the Dijkstra algorithm is used to plan the flight path of the UAV and generate a path plan in combination with the equipment distribution and obstacle characteristics of the power grid system.
6. A power grid system damage detection method as claimed in claim 1, characterized in that: The construction of the disaster damage identification model is specifically as follows: Based on the convolutional neural network, an initial model is constructed; wherein the initial model includes an input layer, a convolutional layer, a pooling layer, a fully connected layer and an output layer, the convolutional layer is used to extract features from the input data, the pooling layer is used to perform maximum pooling processing on the feature extraction results, and the fully connected layer is used to perform classification and recognition processing on the feature extraction results after pooling; The initial model is trained and optimized multiple times using a sample data set, and during the training and optimization process, the cross entropy loss function and the mean square error loss function are used to calculate the loss value of the current initial model in real time until the loss value of the current initial model is less than a first preset value or the current number of training optimizations is equal to a second preset value, and the current initial model is used as a disaster damage identification model; wherein the sample data set includes power grid data at several disaster damage moments, and power grid disaster damage types and disaster damage location information corresponding to each of the disaster damage moments, and the disaster damage moment is the moment when a disaster occurs in the power grid system.
7. A power grid system damage detection device, characterized in that: include: The path planning module is used to plan the flight path of the drone and generate a path plan using a preset path planning algorithm; A shooting and detection module is used to use the drone carrying the camera and the infrared sensor to perform real-time shooting and detection of the power grid system in the target area according to the path plan to obtain the power grid data to be identified; The disaster damage detection module is used to input the power grid data to be identified into the disaster damage identification model, so that the disaster damage identification model extracts and identifies the features of the power grid data to be identified and outputs the power grid disaster type and disaster damage location information corresponding to the target area; inputs the power grid data to be identified into the ARIMA model, so that the ARIMA model predicts the power grid load changes for the power grid data to be identified and outputs the disaster occurrence time corresponding to the target area.
8. A power grid system damage detection device as claimed in claim 7, characterized in that: The shooting detection module specifically includes: An inspection unit is used to use the drone to inspect the power grid system according to the path plan, and during the inspection process, use the camera carried by the drone to take real-time photos of the power grid system to obtain image data of the power grid system, and use the infrared sensor carried by the drone to perform infrared detection on the power grid system to obtain temperature data of the power grid system; A voltage detection unit, configured to detect the voltages of a plurality of grid nodes in the target area using a sensor network deployed in the grid system, so as to generate a voltage distribution diagram of the grid system; wherein the sensor network includes a plurality of voltage sensors; A fusion analysis unit is used to fuse and preprocess the image data, the temperature data and the voltage distribution diagram to obtain the data to be transmitted, and then transmit the data to be transmitted to the central control module, and then analyze and process the received data to be transmitted through the central control module to obtain the power grid data to be identified.
9. A terminal, characterized in that: It comprises a processor, a memory and a computer program stored in the memory; wherein the computer program can be executed by the processor to implement a power grid system disaster damage detection method as described in any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a power grid system disaster detection method as described in any one of claims 1 to 6.
Citation Information
Patent Citations
Unmanned aerial vehicle aerial photography technology-based electric power disaster situation acquisition system and method
CN105373134A
Power transmission line comprehensive fault detection method and system, and storage medium
CN117807558A
Power failure information prediction method based on external abnormal information influence
CN118115901A
Power distribution network potential safety hazard identification method and device
CN118226198A
Method and system for detecting disaster damage information based on artificial intelligence using drone
KR102203135B1
Cited By
Steel structure health monitoring and maintenance method based on multiple intelligent algorithms
CN120087787A
Electric power facility disaster damage monitoring method and system fusing gray release and multi-source heterogeneous data
CN121459510A
