Wide-area monitoring method and related device for aerial hidden dangers in power corridors

Abstract

The present invention provides a method and related devices for wide-area monitoring of aerial hidden dangers in power corridors, and relates to the technical field of power systems. By acquiring low-orbit satellite echo signals; performing distance-dimensional and Doppler-dimensional Fourier transforms according to low-orbit satellite echo signals to generate range-Doppler images; preprocessing the range-Doppler images to build a range-Doppler image database; training based on the range-Doppler image database to generate a wide-area monitoring network for aerial hidden dangers; determining the type of hidden danger and two-dimensional coordinate information according to the low-orbit satellite echo signals; determining the hidden danger image information through the two-dimensional coordinate information combined with a preset image acquisition device; determining the three-dimensional coordinates according to the hidden danger image information and tracking the trajectory of the low-orbit satellite echo signal according to the hidden danger type. Through the collaborative analysis of the target type, position and motion trajectory, the wide-area situational awareness of aerial hidden dangers in the power corridor business area is achieved to ensure the safe operation of the power system.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method and related device for wide-area monitoring of aerial hazards in power corridors. Background Art

[0002] With the continuous expansion of power transmission networks around the world, the monitoring and prevention of aerial hidden dangers has become an important issue for the safe operation of power grids. Especially in remote mountainous areas and power corridors with complex terrain, due to the complex ground conditions, traditional ground monitoring methods are limited in coverage and real-time response, and it is difficult to fully capture the dynamic changes of aerial hidden dangers. Therefore, how to efficiently achieve wide-area monitoring of aerial hidden dangers has become an important challenge facing power grid management.

[0003] At present, aerial hidden danger monitoring mainly relies on ground cameras and drone inspections. Although these methods are somewhat feasible, they usually have problems such as limited monitoring range, insufficient inspection frequency, and difficulty in dealing with bad weather. In particular, for sudden hidden dangers that may be caused by fast-moving aerial obstacles or bad weather conditions, traditional methods are difficult to provide timely warnings, which in turn poses a potential threat to the safety of power corridors.

[0004] Existing methods for monitoring aerial hazards mainly include ground camera monitoring, drone inspection and radar monitoring. Among them, ground camera monitoring is suitable for continuous monitoring at fixed locations, but its monitoring range is limited and it is difficult to cover remote and complex terrain areas. Drone inspection is flexible and can be used to check for hidden dangers in specific areas, but due to battery life limitations, the inspection frequency is low and it is difficult to achieve long-term continuous monitoring. Radar monitoring can detect aerial obstacles to a certain extent and is suitable for environments with good weather conditions, but its performance is significantly reduced under severe weather conditions. In addition, these methods do not respond promptly to fast-moving aerial hazards and it is difficult to provide effective early warning and emergency treatment. By utilizing the full coverage characteristics of low-orbit satellites and the advantages of rapid changes in geometric figures within the signal coverage range to the rapid convergence of precise positioning, continuous and rapid monitoring of aerial hazards in the power corridor business area is achieved. In addition, combined with the range-Doppler images received by the satellite terminal, the motion characteristics of aerial hazards can be further extracted, providing effective support for early warning and emergency treatment of hazards.

[0005] However, the current use of low-orbit satellites for aerial hazard monitoring has the following defects: (1) It is difficult to accurately locate hidden dangers without the assistance of ground cameras. Satellite monitoring can only provide rough two-dimensional location information and cannot meet the needs of fine-grained tracking of hidden danger targets; (2) Satellite signals are greatly affected by environmental factors, especially in complex terrain and harsh weather conditions, where signal quality is easily disturbed, resulting in unstable monitoring results; (3) Although low-orbit satellites have a wide coverage, due to their high-speed operation, monitoring is somewhat intermittent, and continuous, real-time and detailed monitoring cannot be achieved. The above defects limit the actual application effect of relying solely on satellite monitoring.

[0006] Therefore, how to achieve wide-area situational awareness of aerial hazards in the power corridor business area and effectively ensure the safe operation of the power system has become a technical problem that needs to be solved urgently. Summary of the invention

[0007] In order to achieve wide-area situational awareness of aerial hazards in the power corridor business area and effectively ensure the safe operation of the power system, the present application provides a wide-area monitoring method and related devices for aerial hazards in the power corridor.

[0008] In the first aspect, the present application provides a method for wide-area monitoring of aerial hazards in power corridors using the following technical solutions:

[0009] A wide-area monitoring method for aerial hidden dangers in power corridors, comprising:

[0010] Acquire low-orbit satellite echo signals through a preset communication device;

[0011] Performing range-dimensional and Doppler-dimensional Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image;

[0012] Preprocessing the range-Doppler image to construct a range-Doppler image database;

[0013] Obtaining a preset fusion model and training it based on the range-Doppler image database to generate a wide-area monitoring network for air hazards;

[0014] Inputting the low-orbit satellite echo signal into the wide-area monitoring network for airborne hidden dangers to determine the hidden danger type and two-dimensional coordinate information;

[0015] Determine hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device;

[0016] The three-dimensional coordinates are determined according to the hidden danger image information and the trajectory of the low-orbit satellite echo signal is tracked according to the hidden danger type.

[0017] Optionally, the step of performing range-dimensional and Doppler-dimensional Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image includes: defining the low-orbit satellite echo signal as a matrix s(k, l) of size K×L:

[0018] ;

[0019] in, represents the signal amplitude, e is the natural exponent, and π represents pi; The Doppler frequency representing the speed information of the target movement, T represents the period interval, l represents a period of the data frame, and L represents the number of frequency modulation periods contained in a frame of the terminal receiving signal data frame; The distance frequency represents the distance information between the terminal and the target, k represents a certain sampling point, K represents the number of sampling points in a frequency modulation cycle, s(k) represents the distance dimension Fourier transform function, and s(l) represents the Doppler dimension transform function;

[0020] The distance dimension Fourier transform is performed on the low-orbit satellite echo signal, each frequency modulation period is taken as a basic unit, all sampling points in the period are regarded as time domain representation, and then Fourier transform is performed:

[0021] ;

[0022] Where w(k) represents the use of a window function for K points, u represents the index after the distance-dimensional Fourier transform, and j is a complex unit;

[0023] In a frame of ground satellite terminal receiving signal data, the data at the same distance is windowed with the frequency modulation period as the coordinate, and the Doppler Fourier transform is performed to obtain the range-Doppler map:

[0024] ;

[0025] Wherein, w(l) indicates the use of a window function for L points, and v indicates the index after the distance-dimensional Fourier transform;

[0026] After two-dimensional Fourier transform processing, a frame of signal data received by the terminal is converted into a data matrix containing the distance dimension and the Doppler dimension:

[0027] ;

[0028] A distance unit The signal strength in the range is converted to the logarithmic domain to obtain the range-Doppler map .

[0029] Optionally, the step of preprocessing the range-Doppler image to construct a range-Doppler image database includes:

[0030] Acquire actual environmental characteristics of the current power corridor, and determine a set of aerial hidden dangers according to the actual environmental characteristics;

[0031] The range-Doppler image is converted into a grayscale image using a logarithmic conversion function:

[0032] Assume the grayscale image is , obtained by the following formula:

[0033] ;

[0034] in is the set of real numbers, and They control the dynamic range and gray level respectively, Z represents the range-Doppler matrix, and |Z|max represents the maximum modulus of the range-Doppler matrix Z;

[0035] A range-Doppler image database is constructed according to the grayscale image and the aerial hidden danger set.

[0036] Optionally, the preset fusion model includes: a GCN and CNN fusion model.

[0037] Optionally, the step of obtaining a preset fusion model and training based on the range-Doppler image database to generate a wide-area monitoring network for air hazards includes:

[0038] Inputting the range-Doppler image database into the CNN to extract local features through multi-layer convolution operations;

[0039] Convert the range-Doppler image into an adjacency matrix and input it into the GCN to extract global features;

[0040] The local features and global features are combined to obtain feature H result ;

[0041] Predicting category results through the Softmax function ;

[0042] ;

[0043] in, is the weight matrix, d is the bias vector, and the trained preset fusion model is used as the wide-area monitoring network for aerial hazards.

[0044] Optionally, the step of determining hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device includes:

[0045] Acquire image information of low-orbit satellite echo signals through a preset image acquisition device;

[0046] Processing the image information by image dilation and image erosion strategies to generate first image information;

[0047] Among them, g(x) and b(x) are used to represent two discrete functions in two-dimensional space, and g(x) describes the grayscale image, and b(x) represents the structural element;

[0048] The expansion process is expressed as: ;

[0049] The corrosion process is expressed as: ;

[0050] Separating the background and foreground of the first image information to generate second image information;

[0051] Assume there is no noise Interference, then track the image sequence By background image With sports goals jointly constitute;

[0052] ;

[0053] image , is a combination of moving target and noise:

[0054] Use threshold segmentation algorithm:

[0055] ;

[0056] In the formula, m represents the threshold;

[0057] Based on the results of threshold segmentation, the color components are normalized using the normalized RGB difference algorithm, r=R / (R+G+B), g=G / (R+G+B), b=B / (R+G+B);

[0058] Calculate the Gaussian model parameters for each pixel in the background image , and u and σ represent the mean and variance of the corresponding points in the background respectively;

[0059] Based on the established background model, a difference operation is performed to obtain a binary image , foreground and background pixels are represented as 1 and 0 respectively;

[0060] ;

[0061] In the formula, Representing pixels Measured value, It belongs to the threshold parameter. If the r, g, b components of a pixel are changed, it is regarded as a foreground pixel to generate the second image information;

[0062] fusing the second image information by a panoramic stitching algorithm;

[0063] Assume that the width of the image is W, the height of the image is H, the focal length f taken by the preset image acquisition device, (x, y) is the coordinate of any point p in the viewing plane, and (xc, yc, zc) is the projection coordinate of point p on the cylindrical surface;

[0064] ;

[0065] The fused image information is used as the target image information corresponding to the low-orbit satellite echo signal.

[0066] Optionally, the step of acquiring a low-orbit satellite echo signal by a preset communication device includes:

[0067] By presetting the communication strategy and combining the dual sensors of the global navigation satellite system and the inertial measurement unit, the position and attitude information of the satellite communication terminal is obtained as the low-orbit satellite echo signal;

[0068] The preset communication strategy includes: obtaining orbital data of low-orbit satellites in combination with ephemeris to determine the position of visible satellites within the power corridor;

[0069] Based on the position and attitude of the satellite communication terminal, the motion trajectory of the low-orbit satellite is estimated, and the satellite position is converted from the earth's fixed coordinate system to the terminal antenna coordinate system through multiple coordinate system conversion;

[0070] Control satellite communication terminal antennas to accurately point to visible satellites in azimuth, elevation and polarization;

[0071] A satellite switching algorithm is applied to switch to the next visible satellite when the currently tracked low-orbit satellite is not visible, stops transmitting, or the signal is below the set threshold.

[0072] In a second aspect, the present application provides a wide-area monitoring system for aerial hidden dangers in power corridors, the wide-area monitoring system for aerial hidden dangers in power corridors comprising:

[0073] A signal acquisition module, used to acquire low-orbit satellite echo signals through a preset communication device;

[0074] An image generation module, used for performing range dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image;

[0075] A database generation module, used for preprocessing the range-Doppler image to construct a range-Doppler image database;

[0076] A network training module, used for acquiring a preset fusion model and performing training based on the range-Doppler image database to generate a wide-area monitoring network for air hazards;

[0077] A coordinate information acquisition module, used for inputting the low-orbit satellite echo signal into the wide-area monitoring network for airborne hidden dangers to determine the hidden danger type and two-dimensional coordinate information;

[0078] A hidden danger image determination module, used to determine hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device;

[0079] The trajectory tracking module is used to determine the three-dimensional coordinates according to the hidden danger image information and to track the trajectory of the low-orbit satellite echo signal according to the hidden danger type.

[0080] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, wherein the processor executes the method described above when running computer instructions stored in the memory.

[0081] In a fourth aspect, the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a computer, enable the computer to execute the method described above.

[0082] In summary, the present application includes the following beneficial technical effects:

[0083] This application obtains low-orbit satellite echo signals through a preset communication device; performs distance dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signals to generate distance-Doppler images; pre-processes the distance-Doppler images to build a distance-Doppler image database; obtains a preset fusion model and trains based on the distance-Doppler image database to generate a wide-area monitoring network for air hazards; inputs low-orbit satellite echo signals into the wide-area monitoring network for air hazards to determine the type of hazard and two-dimensional coordinate information; determines the hazard image information through the two-dimensional coordinate information combined with the preset image acquisition device; determines the three-dimensional coordinates according to the hazard image information and tracks the trajectory of the low-orbit satellite echo signals according to the hazard type. Through the collaborative analysis of target type, location and motion trajectory, the wide-area situational awareness of air hazards in the power corridor business area is achieved, which effectively guarantees the safe operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 It is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0085] Figure 2 It is a flow chart of the first embodiment of the wide-area monitoring method for aerial hazards in power corridors of the present application.

[0086] Figure 3 It is a flow chart of processing data frames into range Doppler graphs in the wide-area monitoring method for aerial hidden dangers in power corridors of the present application;

[0087] Figure 4 It is a flow chart of the wide-area monitoring method for aerial hidden dangers in power corridors based on the fusion of GCN and CNN in the wide-area monitoring method for aerial hidden dangers in the present application;

[0088] Figure 5 It is a structural block diagram of the first embodiment of the wide-area monitoring system for aerial hazards in power corridors of the present application. DETAILED DESCRIPTION

[0089] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below through the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0090] Reference Figure 1 , Figure 1 A schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0091] like Figure 1 As shown, the computer device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0092] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0093] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a wide-area monitoring program for aerial hazards in power corridors.

[0094] exist Figure 1 In the computer device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the present application can be set in the computer device, and the computer device calls the power corridor aerial hidden danger wide-area monitoring program stored in the memory 1005 through the processor 1001, and executes the power corridor aerial hidden danger wide-area monitoring method provided in the embodiment of the present application.

[0095] The present application embodiment provides a method for wide-area monitoring of aerial hazards in power corridors, referring to Figure 2 , Figure 2 This is a flow chart of the first embodiment of the method for wide-area monitoring of aerial hazards in power corridors of the present application.

[0096] In this embodiment, the wide-area monitoring method for aerial hidden dangers in power corridors includes the following steps:

[0097] Step S10: acquiring a low-orbit satellite echo signal through a preset communication device.

[0098] It can be understood that the step of obtaining the low-orbit satellite echo signal through the preset communication device includes: obtaining the position and attitude information of the satellite communication terminal as the low-orbit satellite echo signal through the preset communication strategy and combining the global navigation satellite system and the inertial measurement unit dual sensors; wherein the preset communication strategy includes: obtaining the orbital data of the low-orbit satellite in combination with the ephemeris, and determining the orientation of the visible satellite within the power corridor; estimating the motion trajectory of the low-orbit satellite based on the position and attitude of the satellite communication terminal, and converting the satellite position from the earth's fixed coordinate system to the terminal antenna coordinate system through multiple coordinate system conversions; controlling the satellite communication terminal antenna to accurately point to the visible satellite in azimuth, elevation and polarization directions; and applying the satellite switching algorithm to switch to the next visible satellite when the currently tracked low-orbit satellite is invisible, stops transmitting, or the signal is lower than the set threshold.

[0099] It should be noted that in order to achieve efficient wide-area monitoring of aerial hazards in power corridors, it is necessary to control the attitude of satellite terminals on transmission towers and maintain continuous and uninterrupted communication with each visible satellite. This provides reliable technical support for wide-area monitoring of power corridors while ensuring stable transmission of power signals.

[0100] GNSS provides high-precision geographic location measurement, and IMU is used to perceive changes in terminal attitude in real time, including tilt, pitch and yaw angles.

[0101] Based on the position and attitude of the satellite communication terminal, the motion trajectory of the low-orbit satellite is estimated, and through multiple coordinate system conversions, the satellite position is converted from the earth's fixed coordinate system to the terminal antenna coordinate system, achieving unified and accurate conversion between different coordinate systems.

[0102] The satellite communication terminal antenna is then controlled to accurately point to the visible satellite in azimuth, elevation and polarization to ensure the continuity and reliability of communication.

[0103] Finally, the satellite switching algorithm is applied. When the currently tracked low-orbit satellite is not visible, stops transmitting, or the signal is below the set threshold, it automatically switches to the next visible satellite to ensure uninterrupted communication. By combining the low-orbit satellite switching protocol with the electronic steerable antenna, seamless switching is achieved to continuously ensure uninterrupted communication between the ground and the satellite.

[0104] Step S20: Performing range-dimensional and Doppler-dimensional Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image.

[0105] It should be noted that the step of performing range-dimensional and Doppler-dimensional Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image includes:

[0106] The low-orbit satellite echo signal is defined as a matrix s(k, l) of size K×L:

[0107] ;

[0108] in, represents the signal amplitude, e is the natural exponent, and π represents pi; The Doppler frequency representing the speed information of the target movement, T represents the period interval, l represents a period of the data frame, and L represents the number of frequency modulation periods contained in a frame of the terminal receiving signal data frame; The distance frequency represents the distance information between the terminal and the target, k represents a certain sampling point, K represents the number of sampling points in a frequency modulation cycle, s(k) represents the distance dimension Fourier transform function, and s(l) represents the Doppler dimension transform function;

[0109] The distance dimension Fourier transform is performed on the low-orbit satellite echo signal, each frequency modulation cycle is taken as the basic unit, all sampling points in the cycle are regarded as time domain representation, and then the Fourier transform is performed:

[0110] ;

[0111] Where w(k) represents the use of a window function for K points, u represents the index after the distance-dimensional Fourier transform, and j is a complex unit;

[0112] In a frame of ground satellite terminal receiving signal data, the data at the same distance is windowed with the frequency modulation period as the coordinate, and the Doppler Fourier transform is performed to obtain the range-Doppler map:

[0113] ;

[0114] Wherein, w(l) indicates the use of a window function for L points, and v indicates the index after the distance-dimensional Fourier transform;

[0115] After two-dimensional Fourier transform processing, a frame of signal data received by the terminal is converted into a data matrix containing the distance dimension and the Doppler dimension:

[0116] ;

[0117] A distance unit The signal strength in the range is converted to the logarithmic domain to obtain the range-Doppler map .

[0118] It is understandable that if Figure 3 As shown in the figure, the range-Doppler map contains the distance information and Doppler dimension information between the aerial hidden danger target and the ground satellite terminal during its movement, and the correlation of the time dimension is implicit between adjacent maps.

[0119] Step S30: pre-processing the range-Doppler image to construct a range-Doppler image database.

[0120] It should be noted that based on the actual environmental characteristics of the power corridor, the aerial hazards that need to be closely monitored should be identified. These hazards mainly include but are not limited to birds or other flying animals nesting, roosting or flying on power facilities, which may cause short circuits or equipment damage; in severe weather conditions such as strong winds and lightning, there may be risks such as equipment loosening, wire breakage, and insulation damage; illegal intrusion of flying objects such as aircraft and drones in neighboring areas, posing a potential threat to power facilities; and lightweight objects such as plastic bags and kite lines floating in the air, which may hang on overhead lines or equipment and cause faults. In addition, aerial hazards may also include excessive vegetation growth invading power lines, etc. These factors may affect the safe operation of the power corridor. The above hazard information is entered as the actual environmental characteristics of the current power corridor.

[0121] It can be understood that the steps of preprocessing the range-Doppler image to construct the range-Doppler image database include: obtaining the actual environmental characteristics of the current power corridor, determining the set of aerial hidden dangers according to the actual environmental characteristics; converting the range-Doppler image into a grayscale image using a logarithmic conversion function: assuming that the grayscale image is , obtained by the following formula:

[0122] ;

[0123] in is the set of real numbers, and The dynamic range and grayscale level are controlled respectively, Z represents the range-Doppler matrix, and |Z|max represents the maximum modulus of the range-Doppler matrix Z; a range-Doppler image database is constructed based on the grayscale image and the set of aerial hazards.

[0124] Step S40: Obtain a preset fusion model and perform training based on a range-Doppler image database to generate a wide-area monitoring network for aerial hazards.

[0125] It should be noted that the preset fusion models include: GCN and CNN fusion models such as Figure 4 The network input shown is the RD diagram of the low-orbit satellite signal received by the terminal under different hidden danger conditions obtained in the second part, and the network output is the hidden danger category obtained by pattern recognition and the two-dimensional coordinate position interval information corresponding to the hidden danger.

[0126] In a specific implementation, the steps of obtaining a preset fusion model and training it based on a range-Doppler image database to generate a wide-area monitoring network for air hazards include:

[0127] The range-Doppler image database is input into CNN to extract local features through multi-layer convolution operations; the range-Doppler image is converted into an adjacency matrix and input into GCN to extract global features; the local features and global features are fused to obtain the feature H result ;

[0128] Predicting category results through the Softmax function ;

[0129] ;

[0130] in, is the weight matrix, d is the bias vector, and the trained preset fusion model is used as the wide-area monitoring network for aerial hazards.

[0131] In the specific implementation, in this embodiment, CNN is first used to process the distance-Doppler map of the satellite terminal receiving signal. CNN uses the distance-Doppler map as input and extracts local features of the image through multi-layer convolution operations. At the same time, in order to better extract global features, the GCN module begins to construct the graph data. Since GCN can only act on graph structure data, it is necessary to convert the RD image into a topological graph structure. This process includes dividing the distance-Doppler map into multiple subgraphs of H×H pixel size, treating each subgraph as an independent node, and calculating the connection relationship between nodes to generate an adjacency matrix.

[0132] After completing the graph data construction, these nodes and their corresponding adjacency matrices are input into the GCN module to further extract global features. The GCN process is summarized as follows:

[0133] 1) Node feature aggregation: Synchronize the information of the node itself and its adjacent nodes to achieve deep integration and convergence of feature information.

[0134] 2) Global weighting: Multiply the fused information with the globally unified weight matrix, and adjust the influence of each node feature through weight distribution, so as to more accurately characterize the mutual relationship between nodes.

[0135] 3) Nonlinear mapping: Introducing activation functions to implement nonlinear mapping, giving the model the ability to learn complex nonlinear associations.

[0136] In this way, the CNN and GCN modules each focus on extracting local and global features, and the two complement each other. Next, the features extracted by the two are fused to obtain the feature Hresult, which is input into a fully connected network, and then the category result is predicted through the Softmax function to achieve end-to-end recognition of air hazards, as shown in the following formula:

[0137] ;

[0138] in and d are the weight matrix and bias vector respectively.

[0139] The GCN and CNN joint model is trained to enable the network to accurately identify different types of aerial hazards and the two-dimensional coordinate intervals where the hazards are located. Finally, the model is fine-tuned based on the environmental characteristics of the power corridor and historical monitoring data to improve the accuracy and adaptability of recognition.

[0140] Step S50: input the low-orbit satellite echo signal into the wide-area monitoring network for air hazards to determine the hazard type and two-dimensional coordinate information.

[0141] Step S60: Determine hidden danger image information through two-dimensional coordinate information combined with a preset image acquisition device.

[0142] It should be noted that in order to improve image quality and solve the image quality differences caused by factors such as lighting changes, climate environment and equipment vibration, the initial image is first subjected to morphological filtering to balance the image brightness and contrast, thereby improving the multi-camera image fusion effect and enhancing the accuracy of identifying aerial hidden danger targets, ensuring that hidden danger targets in complex scenes can be accurately identified.

[0143] In a specific implementation, the step of determining the hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device includes: acquiring the image information of the low-orbit satellite echo signal by the preset image acquisition device; processing the image information by image expansion and image erosion strategies to generate first image information; wherein g(x) and b(x) are used to represent two discrete functions in the two-dimensional space, and g(x) describes the grayscale image, and b(x) represents the structural element;

[0144] The expansion process is expressed as: ;

[0145] The corrosion process is expressed as: ;

[0146] Separating the background and foreground of the first image information to generate second image information;

[0147] Assume there is no noise Interference, then track the image sequence By background image With sports goals jointly constitute;

[0148] ;

[0149] image , is the combination of moving target and noise:

[0150] ;

[0151] Use threshold segmentation algorithm:

[0152] ;

[0153] In the formula, m represents the threshold;

[0154] Based on the results of threshold segmentation, the color components are normalized using the normalized RGB difference algorithm, r=R / (R+G+B), g=G / (R+G+B), b=B / (R+G+B);

[0155] Calculate the Gaussian model parameters for each pixel in the background image , and u and σ represent the mean and variance of the corresponding points in the background respectively;

[0156] Based on the established background model, a difference operation is performed to obtain a binary image , foreground and background pixels are represented as 1 and 0 respectively;

[0157] ;

[0158] In the formula, Representing pixels Measured value, It belongs to the threshold parameter. If the r, g, b components of a pixel are changed, it is regarded as a foreground pixel to generate the second image information;

[0159] fusing the second image information through a panoramic stitching algorithm;

[0160] Assume that the width of the image is W, the height of the image is H, the focal length f taken by the preset image acquisition device, (x, y) is the coordinate of any point p in the viewing plane, and (xc, yc, zc) is the projection coordinate of point p on the cylindrical surface;

[0161] ;

[0162] The fused image information is used as the target image information corresponding to the low-orbit satellite echo signal.

[0163] It can be understood that this embodiment captures the target's moving path in real time through the linkage control of the camera, so as to accurately record the flight trajectory of the aerial hidden dangers and ensure continuous monitoring and position updating of the target.

[0164] Step S70: Determine the three-dimensional coordinates according to the hidden danger image information and track the trajectory of the low-orbit satellite echo signal according to the hidden danger type.

[0165] In the specific implementation, the two-dimensional rough coordinate information obtained in the previous step and the high-resolution cameras arranged on the transmission towers are used to further determine the three-dimensional position information of the aerial hidden dangers and track their real-time trajectory.

[0166] Secondly, a comprehensive alarm mechanism is established. When the system detects that an aerial hazard enters the power corridor area, an alarm is triggered immediately and alarm information is sent to the control center, including the target's real-time location, flight speed and related image data.

[0167] To reduce false alarms, the system has a built-in multi-level target verification mechanism that performs multiple confirmations through Doppler frequency shift analysis and imaging features to ensure the accuracy and reliability of the alarm.

[0168] Finally, to ensure the safe operation of the power transmission line, after the alarm is triggered, the system can automatically dispatch drones to closely track the hidden dangers in the air and obtain more detailed flight data. At the same time, ground inspection personnel can be dispatched to the potential risk area to inspect and assess the risks of the power transmission equipment.

[0169] In addition, this embodiment can analyze the type and threat level of aerial hazards and decide whether to take additional defensive measures, such as expulsion or isolation. This collaborative response mechanism can quickly respond to aerial hazards and ensure that threats to power transmission lines are promptly and effectively controlled and handled.

[0170] It should be noted that this embodiment involves using the multipath effect of low-orbit satellite communication to generate a range-Doppler image (RD map), and fusing the graph neural network (GCN) and the convolutional neural network (CNN) to achieve preliminary identification and positioning of aerial hidden dangers in the power corridor business area, and at the same time, the preset high-definition camera on the ground pole tower is controlled in linkage to capture the precise position and motion trajectory of the target. First, the data frame of the terminal receiving signal is subjected to the Fourier transform of the distance dimension and the Doppler dimension by using the communication channel information between the low-orbit satellite Internet and the ground power communication terminal, and the RD map for air hidden danger monitoring is generated. Then, the generated RD image is analyzed by the pattern recognition network fused with GCN and CNN to preliminarily determine the type of hidden danger target and its two-dimensional position information. Then, the system linkage controls the ground high-definition camera deployed on the adjacent pole tower. According to the preliminary recognition results, the shooting angle of the camera is adjusted and the panoramic image fusion is performed to accurately obtain the three-dimensional position and motion trajectory of the target, so as to achieve accurate tracking of the target. The system realizes the wide-area situational awareness of aerial hidden dangers in the power corridor business area through the collaborative analysis of the target type, position and motion trajectory, and effectively guarantees the safe operation of the power system.

[0171] This embodiment obtains the low-orbit satellite echo signal through a preset communication device; performs distance dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signal to generate a distance-Doppler image; pre-processes the distance-Doppler image to build a distance-Doppler image database; obtains a preset fusion model and trains based on the distance-Doppler image database to generate an aerial hidden danger wide-area monitoring network; inputs the low-orbit satellite echo signal into the aerial hidden danger wide-area monitoring network to determine the hidden danger type and two-dimensional coordinate information; determines the hidden danger image information through the two-dimensional coordinate information combined with the preset image acquisition device; determines the three-dimensional coordinates according to the hidden danger image information and tracks the trajectory of the low-orbit satellite echo signal according to the hidden danger type. Through the collaborative analysis of the target type, position and motion trajectory, the wide-area situational awareness of the aerial hidden dangers in the power corridor business area is realized, which effectively guarantees the safe operation of the power system.

[0172] In addition, an embodiment of the present application also proposes a computer-readable storage medium, on which is stored a program for wide-area monitoring of aerial hidden dangers in power corridors. When the program for wide-area monitoring of aerial hidden dangers in power corridors is executed by a processor, the steps of the method for wide-area monitoring of aerial hidden dangers in power corridors as described above are implemented.

[0173] Reference Figure 5 , Figure 5 This is a structural block diagram of the first embodiment of the wide-area monitoring system for aerial hazards in power corridors of the present application.

[0174] like Figure 5 As shown, the wide-area monitoring system for aerial hidden dangers in power corridors proposed in the embodiment of the present application includes:

[0175] The signal acquisition module 10 is used to acquire the low-orbit satellite echo signal through a preset communication device;

[0176] An image generation module 20 is used to perform range dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image;

[0177] A database generation module 30, for preprocessing the range-Doppler image to construct a range-Doppler image database;

[0178] A network training module 40 is used to obtain a preset fusion model and perform training based on a range-Doppler image database to generate an air hazard wide-area monitoring network;

[0179] A coordinate information acquisition module 50 is used to input the low-orbit satellite echo signal into the wide-area monitoring network for air hazards to determine the type of hazard and two-dimensional coordinate information;

[0180] A hidden danger image determination module 60 is used to determine hidden danger image information by combining two-dimensional coordinate information with a preset image acquisition device;

[0181] The trajectory tracking module 70 is used to determine the three-dimensional coordinates according to the hidden danger image information and to track the trajectory of the low-orbit satellite echo signal according to the hidden danger type.

[0182] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present application. In specific applications, technicians in this field can make settings as needed, and the present application does not impose any limitation on this.

[0183] This embodiment obtains the low-orbit satellite echo signal through a preset communication device; performs distance dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signal to generate a distance-Doppler image; pre-processes the distance-Doppler image to build a distance-Doppler image database; obtains a preset fusion model and trains based on the distance-Doppler image database to generate an aerial hidden danger wide-area monitoring network; inputs the low-orbit satellite echo signal into the aerial hidden danger wide-area monitoring network to determine the hidden danger type and two-dimensional coordinate information; determines the hidden danger image information through the two-dimensional coordinate information combined with the preset image acquisition device; determines the three-dimensional coordinates according to the hidden danger image information and tracks the trajectory of the low-orbit satellite echo signal according to the hidden danger type. Through the collaborative analysis of the target type, position and motion trajectory, the wide-area situational awareness of the aerial hidden dangers in the power corridor business area is realized, which effectively guarantees the safe operation of the power system.

[0184] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present application. In practical applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0185] In addition, for technical details not described in detail in this embodiment, please refer to the method for wide-area monitoring of aerial hazards in power corridors provided in any embodiment of the present application, which will not be repeated here.

[0186] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0187] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0188] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0189] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A wide-area monitoring method for aerial hidden dangers in power corridors, characterized in that: include: Acquire low-orbit satellite echo signals through a preset communication device; Performing range-dimensional and Doppler-dimensional Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image; Preprocessing the range-Doppler image to construct a range-Doppler image database; Obtaining a preset fusion model and training it based on the range-Doppler image database to generate a wide-area monitoring network for air hazards; Inputting the low-orbit satellite echo signal into the wide-area monitoring network for airborne hidden dangers to determine the hidden danger type and two-dimensional coordinate information; Determine hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device; The three-dimensional coordinates are determined according to the hidden danger image information and the trajectory of the low-orbit satellite echo signal is tracked according to the hidden danger type.

2. The method for wide-area monitoring of aerial hidden dangers in power corridors according to claim 1 is characterized in that: The step of performing range dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image comprises: The low-orbit satellite echo signal is defined as a matrix s(k, l) of size K×L: ； in, represents the signal amplitude, e is the natural exponent, and π represents pi; The Doppler frequency represents the speed information of the target movement, T represents the period interval, l represents a period of the data frame, and L represents the number of frequency modulation periods contained in a frame of the terminal receiving signal data frame; The distance frequency represents the distance information between the terminal and the target, k represents a certain sampling point, K represents the number of sampling points in a frequency modulation cycle, s(k) represents the distance dimension Fourier transform function, and s(l) represents the Doppler dimension transform function; The distance dimension Fourier transform is performed on the low-orbit satellite echo signal, each frequency modulation period is taken as a basic unit, all sampling points in the period are regarded as time domain representation, and then Fourier transform is performed: ； Where w(k) represents the use of a window function for K points, u represents the index after the distance-dimensional Fourier transform, and j is a complex unit; In a frame of ground satellite terminal receiving signal data, the data at the same distance is windowed with the frequency modulation period as the coordinate, and the Doppler Fourier transform is performed to obtain the range-Doppler map: ； Wherein, w(l) indicates the use of a window function for L points, and v indicates the index after the distance-dimensional Fourier transform; After two-dimensional Fourier transform processing, a frame of signal data received by the terminal is converted into a data matrix containing the distance dimension and the Doppler dimension: ； A distance unit The signal strength in the range is converted to the logarithmic domain to obtain the range-Doppler map .

3. The method for wide-area monitoring of aerial hidden dangers in power corridors according to claim 2 is characterized in that: The step of preprocessing the range-Doppler image to construct a range-Doppler image database comprises: Acquire actual environmental characteristics of the current power corridor, and determine a set of aerial hidden dangers according to the actual environmental characteristics; The range-Doppler image is converted into a grayscale image using a logarithmic conversion function: Assume the grayscale image is , obtained by the following formula: ； in is the set of real numbers, and They control the dynamic range and grayscale level respectively, Z represents the range-Doppler matrix, |Z| max Represents the maximum modulus of the range-Doppler matrix Z; A range-Doppler image database is constructed according to the grayscale image and the aerial hidden danger set.

4. The wide-area monitoring method for aerial hidden dangers in power corridors according to claim 1 is characterized in that: The preset fusion model includes: GCN and CNN fusion model.

5. The method for wide-area monitoring of aerial hidden dangers in power corridors according to claim 4 is characterized in that: The step of obtaining a preset fusion model and training based on the range-Doppler image database to generate a wide-area monitoring network for air hazards includes: Inputting the range-Doppler image database into the CNN to extract local features through multi-layer convolution operations; Convert the range-Doppler image into an adjacency matrix and input it into the GCN to extract global features; The local features and global features are combined to obtain feature H result ; Predicting category results through the Softmax function ; ； in, is the weight matrix, d is the bias vector, and the trained preset fusion model is used as the wide-area monitoring network for aerial hazards.

6. The wide-area monitoring method for aerial hidden dangers in power corridors according to claim 1 is characterized in that: The step of determining hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device includes: Acquire image information of low-orbit satellite echo signals through a preset image acquisition device; Processing the image information by image dilation and image erosion strategies to generate first image information; Among them, g(x) and b(x) are used to represent two discrete functions in two-dimensional space, and g(x) describes the grayscale image, and b(x) represents the structural element; The expansion process is expressed as: ; The corrosion process is expressed as: ; Separating the background and foreground of the first image information to generate second image information; Assume there is no noise Interference, then track the image sequence By background image With sports goals jointly constitute; ； image , is a combination of moving target and noise: ； Use threshold segmentation algorithm: ； In the formula, m represents the threshold; Based on the results of threshold segmentation, the color components are normalized using the normalized RGB difference algorithm, r=R / (R+G+B), g=G / (R+G+B), b=B / (R+G+B); Calculate the Gaussian model parameters for each pixel in the background image , and u and σ represent the mean and variance of the corresponding points in the background respectively; Based on the established background model, a difference operation is performed to obtain a binary image , foreground and background pixels are represented as 1 and 0 respectively; ； In the formula, Representing pixels Measured value, It belongs to the threshold parameter. If the r, g, b components of a pixel are changed, it is regarded as a foreground pixel to generate the second image information; fusing the second image information by a panoramic stitching algorithm; Assume that the width of the image is W, the height of the image is H, the focal length f taken by the preset image acquisition device, (x, y) is the coordinate of any point p in the viewing plane, (x c , y c , z c ) is the projection coordinate of point p on the cylindrical surface; ； The fused image information is used as the target image information corresponding to the low-orbit satellite echo signal.

7. The method for wide-area monitoring of aerial hidden dangers in power corridors according to any one of claims 1 to 6, characterized in that: The step of obtaining the low-orbit satellite echo signal through the preset communication device includes: By presetting the communication strategy and combining the dual sensors of the global navigation satellite system and the inertial measurement unit, the position and attitude information of the satellite communication terminal is obtained as the low-orbit satellite echo signal; The preset communication strategy includes: obtaining orbital data of low-orbit satellites in combination with ephemeris to determine the position of visible satellites within the power corridor; Based on the position and attitude of the satellite communication terminal, the motion trajectory of the low-orbit satellite is estimated, and the satellite position is converted from the earth's fixed coordinate system to the terminal antenna coordinate system through multiple coordinate system conversion; Control satellite communication terminal antennas to accurately point to visible satellites in azimuth, elevation and polarization; A satellite switching algorithm is applied to switch to the next visible satellite when the currently tracked low-orbit satellite is not visible, stops transmitting, or the signal is below the set threshold.

8. A wide-area monitoring system for aerial hidden dangers in power corridors, characterized in that: The power corridor aerial hidden danger wide-area monitoring system includes: A signal acquisition module, used to acquire low-orbit satellite echo signals through a preset communication device; An image generation module, used for performing range dimension and Doppler dimension Fourier transform according to the low-orbit satellite echo signal to generate a range-Doppler image; A database generation module, used for preprocessing the range-Doppler image to construct a range-Doppler image database; A network training module, used for acquiring a preset fusion model and performing training based on the range-Doppler image database to generate a wide-area monitoring network for air hazards; A coordinate information acquisition module, used for inputting the low-orbit satellite echo signal into the wide-area monitoring network for air hazards to determine the type of hazard and two-dimensional coordinate information; A hidden danger image determination module, used to determine hidden danger image information by combining the two-dimensional coordinate information with a preset image acquisition device; The trajectory tracking module is used to determine the three-dimensional coordinates according to the hidden danger image information and to track the trajectory of the low-orbit satellite echo signal according to the hidden danger type.

9. A computer device, characterized in that: The device comprises: a memory and a processor, and when the processor runs the computer instructions stored in the memory, the processor executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The method comprises instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 7.

Images