Electric power safety monitoring method and system based on infrared imaging technology

Abstract

The invention discloses an electric power safety monitoring method and system based on an infrared imaging technology, and relates to the technical field of electric power monitoring, and the method comprises the steps: carrying an infrared imaging device by an unmanned aerial vehicle, defining an electric power system detection node, and setting a flight path of the unmanned aerial vehicle; the method comprises the following steps: acquiring an infrared image of a power system detection node through an infrared imaging device, and processing the infrared image to extract a node infrared vector; constructing a power system detection model, analyzing node safety based on a node infrared vector, and implementing processing measures according to an analysis result; and the unmanned aerial vehicle stores the infrared image to a database of the unmanned aerial vehicle in real time, and copies the database of the unmanned aerial vehicle to a cloud for storage after flight is finished. According to the method, the precision and accuracy of the infrared image are greatly improved, the convenience of power system detection is improved, the node infrared vectors are synchronously extracted and input into the power system detection model for anomaly detection, and efficient and accurate fault detection of the power system is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power monitoring, and in particular to an electric power safety monitoring method and system based on infrared imaging technology. Background Art

[0002] In recent years, with the continuous expansion and complexity of the scale of power systems, the safety monitoring of power equipment has become the key to ensuring the stable operation of power systems. However, traditional power equipment detection methods mainly rely on manual inspections and fixed monitoring devices. These methods have significant limitations in widely distributed power networks. Manual inspections require personnel to enter complex and dangerous environments, which is not only inefficient, but also has high safety risks. Although fixed monitoring devices can achieve continuous monitoring of certain key parts, their installation and maintenance costs are high. At the same time, in the case of dynamic changes in equipment failures, they cannot be adjusted and responded to in real time, which makes it difficult for traditional methods to meet the growing monitoring needs of modern power systems. Therefore, with the rapid development of infrared imaging technology, the use of infrared equipment to capture the thermal radiation data of equipment in real time has made it possible for efficient monitoring of power systems. However, the application of this technology in power systems still has shortcomings. Existing technologies usually rely on single pulse or continuous wave infrared detection, lack of in-depth analysis of complex power equipment structures, and low image resolution and accuracy, resulting in reduced reliability of fault detection. Summary of the invention

[0003] In view of the problems existing in the above-mentioned existing power safety monitoring method and system based on infrared imaging technology, the present invention is proposed.

[0004] Therefore, the problem to be solved by the present invention is that the existing technology usually relies on a single pulse or continuous wave infrared detection, lacks in-depth analysis of the complex structure of power equipment, and has low image resolution and accuracy, resulting in reduced reliability of fault detection.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: a power safety monitoring method based on infrared imaging technology, which includes: a drone equipped with an infrared imaging device, defining a power system detection node, and setting a flight path of the drone; obtaining an infrared image of the power system detection node through the infrared imaging device, and processing the infrared image to extract the node infrared vector; constructing a power system detection model, analyzing the node safety based on the node infrared vector, and implementing processing measures according to the analysis results; the drone stores the infrared image in real time in its own database, and copies its own database to cloud storage after the flight.

[0006] As a preferred solution of the power safety monitoring method based on infrared imaging technology described in the present invention, the drone is equipped with infrared imaging equipment, power system detection nodes are defined, and the drone flight path is set to collect power system information, define detection nodes in the power system, obtain location information of all detection nodes in the power system, and obtain the connection relationship between the detection nodes to form a power system connection diagram, use genetic algorithm to optimize the drone flight path, the drone is equipped with infrared imaging equipment to fly according to the drone flight path, and hovers when arriving at each detection node to detect the detection node using the infrared imaging equipment.

[0007] As a preferred solution of the power safety monitoring method based on infrared imaging technology described in the present invention, wherein: the temperature field of the power system detection node obtained by the infrared imaging device refers to the drone hovering after arriving at the power system detection node according to the flight path, using image recognition technology to identify the specific position of the detection node and mark it, the drone uses the infrared imaging device to record the initial temperature of the detection node, and after the recording is completed, the detection node is pulse heated, and the pulse thermal imaging of the detection node is monitored by an infrared camera to calculate the pulse temperature T1 (z, t) at the depth z at time t; The temperature of the detection node is continuously monitored by infrared imaging equipment. When the temperature of the detection node returns to the initial temperature, a laser point source is used to laser heat the same position of the detection node to the same temperature and an infrared camera is used to monitor the laser temperature T2 (z, t); The pulse temperature T1 (z, t) and the laser temperature T2 (z, t) are converted into frequency domain representations using Fourier transform, and the real and imaginary parts of the pulse temperature frequency domain representation and the laser temperature frequency domain representation are extracted respectively. The phase delays of the pulse temperature and the laser temperature are calculated based on the extracted real and imaginary parts: in and are the phase delays of pulse temperature and laser temperature, and are the real and imaginary parts of the pulse temperature frequency domain representation, and are the real and imaginary parts of the frequency domain representation of the laser temperature, respectively; Use lock-in thermography to complement the phase delay to the pulse temperature and laser temperature respectively: in and are the enhanced pulse temperature and laser temperature, respectively; Based on the enhanced pulse temperature and laser temperature, a pulse temperature infrared image and a laser temperature infrared image are formed respectively, and Gaussian pyramids of the pulse temperature infrared image and the laser temperature infrared image are constructed respectively, and a Laplacian pyramid is constructed through the Gaussian pyramid; Combining Gaussian pyramid and Laplacian pyramid to construct edge enhancement pyramid: Where E k (x, y) is the pixel at the kth layer point (x, y) in the edge enhancement pyramid, L k (x, y) is the pixel of the k-th layer point (x, y) in the Laplacian pyramid, G k (x, y) is the pixel of the k-th level point (x, y) in the Laplacian pyramid, ∇ 2 G k (x, y) is the Laplacian operator of the Gaussian pyramid, is the edge enhancement coefficient; The weight of each layer in the edge enhancement pyramid for pulse temperature infrared images and laser temperature infrared images is calculated as: Where m k is the weight of the kth layer of the edge enhancement pyramid, including the weight of the kth layer of the edge enhancement pyramid of the pulse temperature infrared image and the weight of the kth layer of the edge enhancement pyramid of the laser temperature infrared image ; Based on the calculated weights, the edge enhancement pyramid of the pulse temperature infrared image and the edge enhancement pyramid of the laser temperature infrared image are fused using a weighted Laplace fusion algorithm to obtain a comprehensive edge enhancement pyramid after fusion of each layer. The comprehensive edge enhancement pyramids of all layers are reconstructed to form a comprehensive infrared image of the detection node.

[0008] As a preferred solution of the power safety monitoring method based on infrared imaging technology described in the present invention, wherein: the processing of infrared images to extract node infrared vectors refers to obtaining the comprehensive infrared images of all time points and then using Gaussian filters to perform filtering and wavelet transform processing, and performing dimensionality reduction processing on the comprehensive infrared images through principal component analysis; The pixel gradient in the integrated infrared image is calculated by the Sobel operator, and the edge of the integrated infrared image is extracted according to the pixel gradient; The change rate of each pixel in the comprehensive infrared image over time is calculated, and the change matrix is ​​constructed by integrating the change rates of all pixels; The integrated infrared image after dimensionality reduction, the edge of the integrated infrared image and the change matrix are formed into a set to construct a node infrared vector.

[0009] As a preferred solution of the power safety monitoring method based on infrared imaging technology described in the present invention, wherein: the construction of the power system detection model, the node safety analysis based on the node infrared vector refers to the use of a convolutional neural network to construct a power system detection model, including an input layer, a convolution layer, a pooling layer, a fully connected layer and an output layer, the input layer input format is set to a node infrared vector format, and the output layer output is a detection node state, including normal and abnormal; Collect historical data of the power system, extract infrared vectors of historical nodes of detection nodes to form a training set, and annotate the training set realistically; Substitute the training set into the power system detection model for model training, define the cross entropy loss function and the adam optimizer to iteratively optimize the model parameters, and stop iterating and output the optimized model parameters when the model loss no longer decreases significantly during continuous iterations; The optimized model parameters are substituted into the power system detection model, and the node infrared vector extracted by the UAV is input into the power system detection model to obtain the detection node status.

[0010] As a preferred solution of the power safety monitoring method based on infrared imaging technology described in the present invention, wherein: the implementation of processing measures according to the analysis results refers to implementing power system processing measures according to the detection node status after obtaining the detection node status through the power system detection model: If the detection node status is normal, the drone will record the detection node status, mark it in the power system connection diagram, and fly to the next detection node for detection according to the flight path; If the status of the detection node is abnormal, the drone will mark the detection node as abnormal in the power system connection diagram, and simultaneously send the location of the abnormal detection node to the power system maintenance personnel. At the same time, the drone will suspend the detection task and hover at the location of the abnormal detection node waiting for the arrival of the power system maintenance personnel. When the power system maintenance personnel arrive, the drone will send the infrared image of the abnormal detection node to the power system maintenance personnel through the wireless network to assist in determining the cause of the detection node failure and maintenance.

[0011] As a preferred solution of the power safety monitoring method based on infrared imaging technology described in the present invention, wherein: the drone stores the infrared image in real time in its own database, and copies its own database to cloud storage after the flight. It means that after obtaining the infrared image of the detection node, the drone stores the infrared image in real time in its own database and continues to collect the infrared image of the detection node. When the drone has detected all the detection nodes, it returns, and after returning, it uploads the infrared image of the detection node in its own database to the cloud for storage and backup.

[0012] Another object of the present invention is to provide a power safety monitoring system based on infrared imaging technology, which includes a drone control module for defining a drone flight path according to a power system detection node and controlling the drone to perform detection according to the flight path; The infrared monitoring module is used to collect the infrared temperature of the detection node and form an infrared image according to the infrared imaging device carried by the drone; A feature extraction module is used to extract infrared features of nodes in infrared images of detection nodes; The status analysis module is used to analyze the status of the detection node according to the infrared characteristics of the node and implement corresponding processing measures according to the status of the detection node; The storage module is used for the drone to store the infrared images of the detection nodes and upload the infrared images to the cloud after the drone returns.

[0013] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned power safety monitoring method based on infrared imaging technology when executing the computer program.

[0014] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned power safety monitoring method based on infrared imaging technology.

[0015] The beneficial effects of the present invention are as follows: the present invention detects the detection nodes of the power system by using an infrared imaging device carried by a drone, and detects the detection nodes respectively by using pulses and laser point sources, and finally constructs an edge-enhanced pyramid to fuse the detection results of the pulses and laser point sources to generate an infrared image, thereby greatly improving the precision and accuracy of the infrared image, and enhancing the convenience of power system detection. Simultaneously, the infrared vector of the node is extracted and input into the power system detection model for abnormality detection, thereby realizing efficient and accurate fault detection of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 It is a flowchart of the power safety monitoring method based on infrared imaging technology.

[0018] Figure 2 Schematic diagram of the process of acquiring infrared images.

[0019] Figure 3This is a structural diagram of the power safety monitoring system based on infrared imaging technology. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments. Example

[0023] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and which provides a power safety monitoring method based on infrared imaging technology. The power safety monitoring method based on infrared imaging technology includes: S1. The drone is equipped with infrared imaging equipment, defines the power system detection nodes, and sets the flight path of the drone; Specifically, the UAV is equipped with infrared imaging equipment, the power system detection nodes are defined, and the UAV flight path is set to collect power system information, define the detection nodes in the power system, obtain the location information of all detection nodes in the power system, and obtain the connection relationship between the detection nodes to form a power system connection diagram. The UAV flight path is optimized using a genetic algorithm. The UAV is equipped with infrared imaging equipment to fly according to the UAV flight path, and hovers when arriving at each detection node to detect the detection node using the infrared imaging equipment.

[0024] The steps of defining the detection nodes of the power system and forming a connection diagram not only provide basic data for the path planning of the UAV, but also ensure the pertinence and comprehensiveness of the monitoring. In a complex power system, the distribution of detection nodes is often uneven. Some key nodes play an important role in system security. By clarifying the location of these nodes and the connection relationship between them, more effective monitoring strategies can be formulated to ensure that key nodes are monitored first and avoid systemic chain reactions caused by the failure of a certain node. The use of genetic algorithms to optimize the flight path of the UAV significantly improves the detection efficiency of the system, which not only improves the accuracy of path planning, but also reduces the flight distance and energy consumption of the UAV, and extends the flight time of the UAV. The coverage of a single flight mission has been improved, which has important practical application value in the actual monitoring of large-scale power systems, especially in remote or complex terrain environments. The design of the UAV hovering for detection after reaching the detection node has significantly improved the accuracy and reliability of detection. By hovering at the detection node, it ensures that the infrared imaging equipment can obtain clear and stable images, thereby improving the accuracy of fault detection. The acquired infrared image data can be processed and analyzed to extract key features reflecting the health status of the equipment, such as temperature anomalies, hot spot locations, etc. This information is crucial for predictive maintenance of power systems and can help operation and maintenance personnel take preventive measures before a failure occurs to avoid greater losses.

[0025] S2, obtaining infrared images of detection nodes of the power system through infrared imaging equipment, and processing the infrared images to extract infrared vectors of the nodes; Specifically, obtaining the temperature field of the power system detection node through infrared imaging equipment means that the drone arrives at the power system detection node according to the flight path and hovers, uses image recognition technology to identify the specific location of the detection node and mark it, and uses infrared imaging equipment to record the initial temperature of the detection node. After recording, the detection node is pulse heated, and the pulse thermal imaging of the detection node is monitored by an infrared camera to calculate the pulse temperature T1 (z, t) at the depth z at time t, as follows: Where z is the detection depth, is the thermal diffusion coefficient, , k is the thermal conductivity of the detection node material, is the density, c is the specific heat capacity, Q0 is the pulse heat source intensity, To detect the thermal effect coefficient of the node material, ; The temperature of the detection node is continuously monitored by infrared imaging equipment. When the temperature of the detection node returns to the initial temperature, a laser point source is used to laser heat the same position of the detection node to the same temperature and an infrared camera is used to monitor the laser temperature T2 (z, t), as follows: Among them I maxis the maximum power density of the laser point source, a is the radius of the laser beam, r is the radial coordinate of the laser point source heating, t ， is the time variable, p(tt ， ) The time distribution function of the laser point source, usually a Gaussian function; The pulse temperature T1 (z, t) and the laser temperature T2 (z, t) are converted into frequency domain representations using Fourier transform, and the real and imaginary parts of the pulse temperature frequency domain representation and the laser temperature frequency domain representation are extracted, respectively, and expressed as: in is the frequency domain representation, j is the imaginary number representation, f is the frequency, is the real part, is the imaginary part; The phase delays of the pulse temperature and the laser temperature are calculated based on the extracted real and imaginary parts: in and are the phase delays of pulse temperature and laser temperature, and are the real and imaginary parts of the pulse temperature frequency domain representation, and are the real and imaginary parts of the frequency domain representation of the laser temperature, respectively; Use lock-in thermography to complement the phase delay to the pulse temperature and laser temperature respectively: in and are the enhanced pulse temperature and laser temperature, respectively; Based on the enhanced pulse temperature and laser temperature, a pulse temperature infrared image and a laser temperature infrared image are formed respectively, and Gaussian pyramids of the pulse temperature infrared image and the laser temperature infrared image are constructed respectively, and a Laplacian pyramid is constructed through the Gaussian pyramid; Combining Gaussian pyramid and Laplacian pyramid to construct edge enhancement pyramid: Where E k (x, y) is the pixel at the kth layer point (x, y) in the edge enhancement pyramid, L k (x, y) is the pixel of the k-th layer point (x, y) in the Laplacian pyramid, G k (x, y) is the pixel of the k-th level point (x, y) in the Laplacian pyramid, ∇ 2 G k (x, y) is the Laplacian operator of the Gaussian pyramid, is the edge enhancement coefficient; The weight of each layer in the edge enhancement pyramid for pulse temperature infrared images and laser temperature infrared images is calculated as: Where m kis the weight of the kth layer of the edge enhancement pyramid, including the weight of the kth layer of the edge enhancement pyramid of the pulse temperature infrared image and the weight of the kth layer of the edge enhancement pyramid of the laser temperature infrared image ; Based on the calculated weights, the edge enhancement pyramid of the pulse temperature infrared image and the edge enhancement pyramid of the laser temperature infrared image are fused using a weighted Laplace fusion algorithm to obtain a comprehensive edge enhancement pyramid after fusion of each layer. The comprehensive edge enhancement pyramids of all layers are reconstructed to form a comprehensive infrared image of the detection node.

[0026] By using pulse heating and laser heating successively, the same detection node is thermally stimulated twice in different ways. This combination can not only detect the thermal response of the material surface, but also deeply analyze the thermal diffusion characteristics of the internal structure of the material. Pulse heating provides a fast and strong temperature response, which helps to reveal the internal defects of the material; while laser heating provides more detailed thermal distribution information through precise and controllable heat input. This combination greatly improves the ability to analyze the full range of thermal characteristics of the detection node. By converting the pulse temperature and laser temperature into frequency domain representation through Fourier transform, the frequency component of the temperature field can be analyzed more intuitively, thereby revealing the thermal effect characteristics that are difficult to observe in the time domain. Especially when there are complex heat conduction paths or multi-layer materials in the detection node, frequency domain analysis can provide clearer phase Information helps identify the thermal wave propagation characteristics inside the material. The phase-locked thermal imaging technology effectively integrates the phase delay information of the pulse and laser temperature into the temperature field, greatly improving the resolution and contrast of the infrared image. The temperature image after phase-locked processing can more clearly show the temperature abnormality area of ​​the detection node, especially those small thermal differences that are difficult to detect under traditional imaging methods. The edge enhancement pyramid combines the Gaussian pyramid and the Laplace pyramid to perform multi-scale and multi-level fusion processing on the infrared images of the pulse temperature and the laser temperature. This method not only enhances the edge details of the image, but also effectively suppresses noise, ensuring the clarity and resolution of the image. Especially in the edge feature analysis of the detection node, the edge enhancement pyramid can accurately capture subtle edge changes, thereby improving the detection accuracy.

[0027] Furthermore, the infrared image is processed to extract the node infrared vector to obtain the comprehensive infrared image of all time points, and then the Gaussian filter is used for filtering and wavelet transformation, and the dimension reduction of the comprehensive infrared image is performed through principal component analysis; The pixel gradient in the integrated infrared image is calculated by the Sobel operator, and the edge of the integrated infrared image is extracted according to the pixel gradient; The change rate of each pixel in the comprehensive infrared image over time is calculated, and the change matrix is ​​constructed by integrating the change rates of all pixels; The integrated infrared image after dimensionality reduction, the edge of the integrated infrared image and the change matrix are formed into a set to construct a node infrared vector.

[0028] Gaussian filter reduces the interference of noise by smoothing the image, making the subsequent image analysis more stable; while wavelet transform provides detailed representation of the image at multiple scales and retains important edge and texture information. This joint processing method not only improves the signal-to-noise ratio of the image, but also lays a good foundation for subsequent edge detection and feature extraction. When processing a large amount of infrared image data, excessively high data dimensions may lead to low processing efficiency. Dimensionality reduction through principal component analysis can effectively reduce the dimension of the data while retaining the main information features of the image. PCA projects the data into a new coordinate system through linear transformation, so that the data can also well represent the original information in low-dimensional space, which not only improves the computational efficiency, but also reduces the interference of redundant information. Sobel The operator can effectively extract the edge information of the image by calculating the gradient of pixels in the comprehensive infrared image. The edge feature plays an important role in image analysis and can reveal the details and defects on the surface of the device. In the present invention, the application of the Sobel operator enhances the extraction accuracy of the edge feature and provides accurate basic data for the subsequent construction of the node infrared vector. By calculating the change rate of each pixel in the comprehensive infrared image and constructing a change matrix, the dynamic characteristics of the image changing over time can be captured. This process reveals the thermal dynamic changes of the equipment during operation and provides rich information support for the generation of the node infrared vector. By combining the change matrix with the image features and edge information after dimensionality reduction, a node infrared vector with high accuracy and high robustness can be constructed to reflect the overall thermal state of the equipment.

[0029] S3, build a power system detection model, analyze node safety based on node infrared vectors, and implement treatment measures based on the analysis results; Specifically, a power system detection model is constructed, and node safety is analyzed based on node infrared vectors. A convolutional neural network is used to construct a power system detection model, including an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. The input layer input format is set to the node infrared vector format, and the output layer outputs the detection node status, including normal and abnormal; Collect historical data of the power system, extract infrared vectors of historical nodes of detection nodes to form a training set, and annotate the training set realistically; Substitute the training set into the power system detection model for model training, define the cross entropy loss function and the adam optimizer to iteratively optimize the model parameters, and stop iterating and output the optimized model parameters when the model loss no longer decreases significantly during continuous iterations; The optimized model parameters are substituted into the power system detection model, and the node infrared vector extracted by the UAV is input into the power system detection model to obtain the detection node status.

[0030] By constructing a power system detection model based on a convolutional neural network, the present invention can automatically extract representative features from node infrared vectors. The introduction of the convolutional layer enables the model to recognize complex spatial patterns, such as temperature changes and abnormal areas in infrared vectors. These features are crucial to determining the safety status of nodes. Compared with traditional feature extraction methods, the CNN model can automatically learn and extract key features without human intervention, greatly improving the degree of automation and accuracy of detection. By collecting historical data of the power system, a training set containing real annotations is formed. This step ensures the accuracy and robustness of the detection model. The historical data contains the operation of the power system under different working conditions. By extracting historical node infrared vectors and matching them with actual state annotations, the model can learn infrared feature patterns under different states. The real annotation of the training set is the key to ensuring that the model performs well in practical applications, and can effectively avoid the model from processing new data. In the event of misjudgment, during the model training process, the cross entropy loss function is defined and the adam optimizer is used for iterative optimization to ensure the accuracy and stability of the model parameters. The cross entropy loss function provides a clear optimization goal. By continuously minimizing the loss value, the model can fit the training data more accurately. The adam optimizer improves the convergence speed of the model by adaptively adjusting the learning rate, avoiding overfitting or underfitting problems. This optimization process enables the model to quickly reach the optimal state and has strong generalization ability. By substituting the optimized model parameters into the power system detection model and inputting the node infrared vector extracted by the drone, the model can judge the status of the detection node in real time. The node status output by the model includes normal and abnormal. Through this binary classification, operation and maintenance personnel can quickly identify equipment or areas that need attention. This process significantly improves the efficiency of power system monitoring, reduces errors in human judgment, and provides a reliable fault warning mechanism.

[0031] Further, implementing treatment measures according to the analysis results refers to implementing power system treatment measures according to the detection node status after obtaining the detection node status through the power system detection model: If the detection node status is normal, the drone will record the detection node status, mark it in the power system connection diagram, and fly to the next detection node for detection according to the flight path; If the status of the detection node is abnormal, the drone will mark the detection node as abnormal in the power system connection diagram, and simultaneously send the location of the abnormal detection node to the power system maintenance personnel. At the same time, the drone will suspend the detection task and hover at the location of the abnormal detection node waiting for the arrival of the power system maintenance personnel. When the power system maintenance personnel arrive, the drone will send the infrared image of the abnormal detection node to the power system maintenance personnel through the wireless network to assist in determining the cause of the detection node failure and maintenance.

[0032] The node status is automatically detected through the power system detection model, and the corresponding processing measures are triggered in real time according to the detection results. When the detection node status is normal, the drone continues to fly along the predetermined path, thereby ensuring efficient monitoring coverage of the entire power system; when the detection node status is abnormal, the drone can respond immediately, suspend the current task and mark the abnormal node. This instant response capability significantly improves the efficiency of power system fault handling, and can discover and locate problems in the first place to avoid the spread and escalation of problems. When a node abnormality is detected, the drone not only marks the node in the power system connection diagram, but also sends the abnormal information to the power system maintenance personnel simultaneously. This dual notification mechanism ensures the timely transmission of information. Through delivery and sharing, maintenance personnel can quickly locate abnormal nodes, shortening the fault response time. At the same time, marking abnormal nodes helps the maintenance team to understand the fault location and distribution more intuitively, and effectively organize and dispatch resources for repairs. After detecting the abnormality, the drone suspends the mission and hovers at the abnormal node location, waiting for the maintenance personnel to arrive. This design ensures that the drone can continuously monitor the abnormal node and provide real-time data support when the maintenance personnel arrive. Through the wireless network, the drone can transmit key data such as infrared images to maintenance personnel, helping them to determine the cause of the fault faster and more accurately. This auxiliary function not only improves the accuracy of fault diagnosis, but also reduces the workload and uncertainty of maintenance personnel on site.

[0033] S4, the drone stores the infrared image in real time in its own database, and copies its own database to cloud storage after the flight; Specifically, the drone stores the infrared image in real time in its own database, and copies its own database to cloud storage after the flight. This means that after obtaining the infrared image of the detection node, the drone stores the infrared image in real time in its own database and continues to collect infrared images of the detection node. When the drone has detected all the detection nodes, it returns, and after returning, it uploads the infrared image of the detection node in its own database to the cloud for storage and backup.

[0034] By storing the infrared image in the database of the drone in real time, the safe preservation of each frame of detection data is ensured. The real-time storage mechanism avoids the risk of data loss during flight, especially during long-term inspections, the stability and integrity of the data are guaranteed. In addition, real-time storage also allows the drone to access and analyze the collected data at any time, which helps to make instant fault judgments and early warnings during flight. This mechanism greatly improves the reliability of the system and ensures the effectiveness of the detection task. By configuring a local database on the drone, the present invention can effectively manage and store infrared image data. The use of the database not only improves the organization of data storage, but also optimizes the efficiency of data retrieval and access. During the inspection process, the drone can Previously stored infrared images can be retrieved at any time as needed for comparative analysis, thereby improving the accuracy of fault detection. At the same time, the use of the database also makes data transmission more flexible, and data can be uploaded to the cloud in a centralized manner after returning, reducing the communication burden during flight. After the drone completes the inspection task, the infrared images in the local database are uploaded to the cloud to achieve data storage and backup. Cloud storage not only provides a larger storage space, but also enhances data security and accessibility. By backing up data to the cloud, maintenance personnel can access inspection data anytime and anywhere for remote analysis and diagnosis. At the same time, cloud storage also supports data sharing and collaboration, and multiple teams can access and process the same data at the same time, greatly improving the efficiency of fault handling. Example

[0035] Reference Figure 3 , which is the second embodiment of the present invention, and which is different from the previous embodiment, provides a power safety monitoring system based on infrared imaging technology, which includes: The UAV control module is used to define the UAV flight path according to the power system detection node and control the UAV to perform detection according to the flight path; The infrared monitoring module is used to collect the infrared temperature of the detection node and form an infrared image according to the infrared imaging device carried by the drone; A feature extraction module is used to extract infrared features of nodes in infrared images of detection nodes; The status analysis module is used to analyze the status of the detection node according to the infrared characteristics of the node and implement corresponding processing measures according to the status of the detection node; The storage module is used for the drone to store the infrared images of the detection nodes and upload the infrared images to the cloud after the drone returns.

[0036] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0037] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0038] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0039] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0040] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A power safety monitoring method based on infrared imaging technology, characterized in that: include, The drone is equipped with infrared imaging equipment to define the power system detection nodes and set the flight path of the drone; Obtain infrared images of power system detection nodes through infrared imaging equipment, and process the infrared images to extract node infrared vectors; Construct a power system detection model, analyze node safety based on node infrared vectors, and implement treatment measures based on the analysis results; The drone stores infrared images in real time in its own database, and copies its own database to cloud storage after the flight.

2. The power safety monitoring method based on infrared imaging technology according to claim 1, characterized in that: The drone is equipped with infrared imaging equipment, and the detection nodes of the power system are defined. The flight path of the drone is set to collect power system information, define the detection nodes in the power system, obtain the location information of all detection nodes in the power system, and obtain the connection relationship between the detection nodes to form a power system connection diagram. The flight path of the drone is obtained by using a genetic algorithm to optimize. The drone is equipped with infrared imaging equipment to fly according to the flight path of the drone, and hovers when arriving at each detection node to detect the detection node using the infrared imaging equipment.

3. The power safety monitoring method based on infrared imaging technology as claimed in claim 2 is characterized in that: The method of obtaining the temperature field of the power system detection node by infrared imaging equipment means that the drone arrives at the power system detection node according to the flight path and hovers, uses image recognition technology to identify the specific location of the detection node and marks it, and uses infrared imaging equipment to record the initial temperature of the detection node. After recording, the detection node is pulse heated, and the pulse thermal imaging of the detection node is monitored by an infrared camera to calculate the pulse temperature T1 (z, t) at the depth z at time t. The temperature of the detection node is continuously monitored by infrared imaging equipment. When the temperature of the detection node returns to the initial temperature, a laser point source is used to laser heat the same position of the detection node to the same temperature and an infrared camera is used to monitor the laser temperature T2 (z, t); The pulse temperature T1 (z, t) and the laser temperature T2 (z, t) are converted into frequency domain representations using Fourier transform, and the real and imaginary parts of the pulse temperature frequency domain representation and the laser temperature frequency domain representation are extracted respectively. The phase delays of the pulse temperature and the laser temperature are calculated based on the extracted real and imaginary parts: in and are the phase delays of pulse temperature and laser temperature, and are the real and imaginary parts of the pulse temperature frequency domain representation, and are the real and imaginary parts of the frequency domain representation of the laser temperature, respectively; Use lock-in thermography to complement the phase delay to the pulse temperature and laser temperature respectively: in and are the enhanced pulse temperature and laser temperature, respectively; Based on the enhanced pulse temperature and laser temperature, a pulse temperature infrared image and a laser temperature infrared image are formed respectively, and Gaussian pyramids of the pulse temperature infrared image and the laser temperature infrared image are constructed respectively, and a Laplacian pyramid is constructed through the Gaussian pyramid; Combining Gaussian pyramid and Laplacian pyramid to construct edge enhancement pyramid: Where E k (x, y) is the pixel at the kth layer point (x, y) in the edge enhancement pyramid, L k (x, y) is the pixel of the k-th layer point (x, y) in the Laplacian pyramid, G k (x, y) is the pixel of the k-th level point (x, y) in the Laplacian pyramid, ∇ 2 G k (x, y) is the Laplacian operator of the Gaussian pyramid, is the edge enhancement coefficient; The weight of each layer in the edge enhancement pyramid for pulse temperature infrared images and laser temperature infrared images is calculated as: Where m k is the weight of the kth layer of the edge enhancement pyramid, including the weight of the kth layer of the edge enhancement pyramid of the pulse temperature infrared image and the weight of the kth layer of the edge enhancement pyramid of the laser temperature infrared image ; Based on the calculated weights, the edge enhancement pyramid of the pulse temperature infrared image and the edge enhancement pyramid of the laser temperature infrared image are fused using a weighted Laplace fusion algorithm to obtain a comprehensive edge enhancement pyramid after fusion of each layer. The comprehensive edge enhancement pyramids of all layers are reconstructed to form a comprehensive infrared image of the detection node.

4. The power safety monitoring method based on infrared imaging technology as claimed in claim 3 is characterized in that: Processing the infrared image to extract the node infrared vector means obtaining the comprehensive infrared image of all time points, filtering it with a Gaussian filter and performing wavelet transformation, and performing dimensionality reduction processing on the comprehensive infrared image through principal component analysis; The pixel gradient in the integrated infrared image is calculated by the Sobel operator, and the edge of the integrated infrared image is extracted according to the pixel gradient; The change rate of each pixel in the comprehensive infrared image over time is calculated, and the change matrix is ​​constructed by integrating the change rates of all pixels; The integrated infrared image after dimensionality reduction, the edge of the integrated infrared image and the change matrix are formed into a set to construct a node infrared vector.

5. The power safety monitoring method based on infrared imaging technology as claimed in claim 4 is characterized in that: The construction of the power system detection model and the node safety analysis based on the node infrared vector refers to the use of a convolutional neural network to construct a power system detection model, including an input layer, a convolution layer, a pooling layer, a fully connected layer and an output layer, setting the input layer input format to the node infrared vector format, and the output layer output is the detection node status, including normal and abnormal; Collect historical data of the power system, extract infrared vectors of historical nodes of detection nodes to form a training set, and annotate the training set realistically; Substitute the training set into the power system detection model for model training, define the cross entropy loss function and the adam optimizer to iteratively optimize the model parameters, and stop iterating and output the optimized model parameters when the model loss no longer decreases significantly during continuous iterations; The optimized model parameters are substituted into the power system detection model, and the node infrared vector extracted by the UAV is input into the power system detection model to obtain the detection node status.

6. The power safety monitoring method based on infrared imaging technology as claimed in claim 5, characterized in that: The implementation of treatment measures according to the analysis results refers to implementing power system treatment measures according to the detection node status after the detection node status is obtained through the power system detection model: If the detection node status is normal, the drone will record the detection node status, mark it in the power system connection diagram, and fly to the next detection node for detection according to the flight path; If the status of the detection node is abnormal, the drone will mark the detection node as abnormal in the power system connection diagram, and simultaneously send the location of the abnormal detection node to the power system maintenance personnel. At the same time, the drone will suspend the detection task and hover at the location of the abnormal detection node waiting for the arrival of the power system maintenance personnel. When the power system maintenance personnel arrive, the drone will send the infrared image of the abnormal detection node to the power system maintenance personnel through the wireless network to assist in determining the cause of the detection node failure and maintenance.

7. The power safety monitoring method based on infrared imaging technology according to claim 6 is characterized in that: The drone stores the infrared image in real time in its own database, and copies its own database to cloud storage after the flight. This means that after obtaining the infrared image of the detection node, the drone stores the infrared image in real time in its own database and continues to collect the infrared images of the detection node. When the drone has detected all the detection nodes, it returns, and after returning, it uploads the infrared images of the detection nodes in its own database to the cloud for storage and backup.

8. An electric power safety monitoring system based on infrared imaging technology according to any one of claims 1 to 7, characterized in that: include, The UAV control module is used to define the UAV flight path according to the power system detection node and control the UAV to perform detection according to the flight path; The infrared monitoring module is used to collect the infrared temperature of the detection node and form an infrared image according to the infrared imaging device carried by the drone; A feature extraction module is used to extract infrared features of nodes in infrared images of detection nodes; The status analysis module is used to analyze the status of the detection node according to the infrared characteristics of the node and implement corresponding processing measures according to the status of the detection node; The storage module is used for the drone to store the infrared images of the detection nodes and upload the infrared images to the cloud after the drone returns.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the power safety monitoring method based on infrared imaging technology described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power safety monitoring method based on infrared imaging technology described in any one of claims 1 to 7 are implemented.

Images