Ecological environment protection method and system for power transmission and transformation project

Abstract

The invention belongs to the technical field of power transmission and transformation project construction, and provides an ecological environment protection method and system for a power transmission and transformation project, which can construct a frequency portrait of a bird flapping behavior in real time by fusing multiple technical means such as high-frame-rate visual perception, target tracking, individual flapping frequency extraction and frequency domain behavior modeling. The operation state of the rotating device is adaptively adjusted according to a frequency interference similarity principle, so that high-frequency phase dislocation disturbance with the most interference effect is triggered at the moment when birds enter a key critical area, the flight rhythm and the path recognition capability of the birds are effectively damaged, and the birds are promoted to deviate from a flight path of a power transmission equipment area; birds can be prevented from approaching a lead area in time, and the risks of short-circuit tripping, biological casualties and other events caused by the fact that the birds strike the lead are remarkably reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of power transmission and transformation project construction, and in particular relates to an ecological environment protection method and system for power transmission and transformation projects. Background Art

[0002] Ecological and environmental protection for power transmission and transformation projects refers to identifying and assessing the potential impacts of transmission lines and substations on natural ecosystems throughout their planning, construction, operation, and maintenance, and implementing scientific and rational engineering design and management measures to minimize interference with and damage to vegetation, water bodies, soil, wildlife, and their habitats, thereby ensuring the stability and sustainability of regional ecosystems. In the existing ecological and environmental protection system for power transmission and transformation projects, to prevent and control the dual risks to the ecological and power systems, such as conductor short circuits, equipment failures, and bird deaths, that may arise during the migration of migratory birds, the inhabitation of wild birds, or the flight of wild birds over transmission corridors, rotating bird-proofing and lightning protection devices are commonly deployed in high-voltage transmission line areas. These devices create visual disturbances to repel birds and reduce the probability of collisions. These devices typically utilize a mechanical rotating structure with a constant speed and rely on a low-speed electric drive to reflect natural light or create dynamic visual differences to interfere with the bird's visual recognition system, thereby achieving a repellent effect while also balancing energy consumption control and extending the life of the device.

[0003] Existing rotating bird-proofing and lightning protection devices are unable to respond and adjust based on the dynamic behavioral characteristics of birds when approaching the transmission area. During the period of rapid bird flight or migration peak, the visual disturbance caused by its constant rotation rate often lags behind the bird approach process, resulting in an insufficient visual interference time window, poor intervention accuracy and real-time performance, and difficulty in preventing birds from approaching the conductor area in time. As a result, the risk of incidents such as birds striking conductors, causing short circuit tripping and biological casualties remains high, seriously affecting the continuity, safety and ecological compatibility of transmission line operation. Summary of the Invention

[0004] In order to solve the above problems, the present invention proposes an ecological environment protection method and system for power transmission and transformation projects. The present invention integrates multiple technical means such as high-frame rate visual perception, target tracking, individual wingbeat frequency extraction and frequency domain behavior modeling, and can construct a frequency portrait of bird wingbeat behavior in real time. It also adaptively adjusts the operating state of the rotating device based on the principle of similar frequency interference, thereby triggering the most effective high-frequency phase shift disturbance at the moment the bird enters the critical area, effectively disrupting its flight rhythm and path recognition ability, causing it to deviate from the flight path of the transmission equipment area, and can promptly prevent birds from approaching the conductor area, significantly reducing the risk of birds colliding with the conductor, causing short circuit tripping and biological casualties.

[0005] In order to achieve the above object, the present invention is implemented through the following technical solutions: In a first aspect, the present invention provides an ecological environment protection method for power transmission and transformation projects, comprising: Acquire image data of a preset area on a power transmission and transformation project line; Identifying bird targets in the image data, generating a flight trajectory sequence of the target individual, and establishing an edge contour identification area for each bird; Based on the flight trajectory sequence and edge contour recognition area, the bird's morphological change range is extracted and the dynamic interest area is delineated; In the dynamic region of interest, the vertical pixel position changes of the bird's wings are continuously modeled, the pixel displacement data on the time axis is extracted, and the pixel displacement-time curve is generated; Perform fast Fourier transform on the pixel displacement-curve, map the original trajectory signal from the time domain to the frequency domain, extract the main frequency component in the spectrum and establish an analysis set; Extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set. Analyze the extracted key features within the monitoring window to assess the frequency and changing trends of bird wing movements and quantify bird strike behavior. The rotation rate of the rotating bird-proof and lightning-proof device is adjusted according to the activity frequency and change trend of bird wings and the principle of co-frequency interference.

[0006] Furthermore, bird target recognition includes: preprocessing the image data; based on the preprocessed image data, extracting feature vectors in the image through a deep convolutional neural network and generating candidate regions, classifying and judging the candidate regions and performing bounding box regression to obtain a predicted frame containing the bird target and its confidence score, and using a non-maximum suppression algorithm to remove duplicate detection frames.

[0007] Furthermore, a flight trajectory sequence of the target individual is generated, and an edge contour recognition area of ​​each bird is established, including: identifying the bounding box information of the bird target in each frame image, constructing a state vector for each detected bird target, and predicting the motion position in the next frame; matching the prediction result with the detection result of the current frame, assigning a stable ID to each target, and realizing individual tracking; in each frame, the bird contour area is secondary extracted based on the bounding box coordinates of the matched target, and the bird edge contour is refined in combination with edge detection; the bird edge information with unique ID is combined in chronological order to generate a stable flight trajectory sequence and a corresponding edge recognition area set.

[0008] Furthermore, delineating the dynamic region of interest includes: extracting the bounding box area of ​​the target bird in each frame as the initial candidate area based on the target detection and tracking results; extracting the bird foreground mask within the initial candidate area; identifying the dynamically changing area based on the pixel displacement within the bird mask in consecutive frames; clustering analysis and spatial calibration of the high-frequency activity areas of the left and right wings using the pixel motion intensity map, and delineating the dynamic region of interest within the edge of the bird.

[0009] Furthermore, modeling the vertical pixel displacement of bird wings includes: frame-by-frame registration of the same dynamic region of interest in consecutive video frames; calculating the pixel grayscale difference between each pair of adjacent frames; selecting vertical scan lines or key points with a grayscale change frequency higher than a preset value within the dynamic region of interest, and accumulating and statistically analyzing the grayscale difference values ​​along the time series to form a pixel position change sequence of wing movement; mapping the pixel position change sequence to a pixel displacement-time curve, with the time frame as the horizontal axis and the pixel offset as the vertical axis.

[0010] Furthermore, extracting the main frequency component includes: normalizing the constructed pixel displacement sequence, removing the offset baseline and unifying the amplitude scale; applying the fast Fourier transform algorithm to map the time series signal from the time domain to the frequency domain to generate complex spectrum data; performing modulus calculation on the spectrum to obtain the amplitude spectrum, and determining the main wingbeat frequency of the bird in the current time window by identifying the frequency peak with the highest energy concentration; and storing the main wingbeat frequency and its corresponding spectrum energy as a set of feature points in the analysis set.

[0011] Furthermore, within the monitoring window, the extracted key features are analyzed, including: generating a frequency transition index after an in-depth analysis of the extension of the tail after the main frequency in the spectrum: in the set frequency window, first identifying the tail frequency segment after the main frequency point of the spectrum, and extracting the frequency transition amplitude by constructing an energy difference sequence between adjacent frequency points in the tail area; inputting all the frequency transition amplitudes of the tail into a nonlinear amplification function, using a high-order hyperbolic function to enhance the responsiveness to the mutation frequency, and constructing a frequency transition index.

[0012] Furthermore, adjusting the rotation rate of the rotating bird-proof lightning protection device includes: constructing a dynamic target interference frequency according to the bird's wingbeat frequency and its corresponding frequency transition index, which is used to drive the speed adjustment of the rotating bird-proof lightning protection device. The formula is: ; in: is the current main wingbeat frequency of the bird; is the frequency transition index; is the nonlinear perturbation amplification function; is the disturbance coefficient; is the dynamic target interference frequency; The calculated dynamic target interference frequency Convert it into the rotational angular velocity of the rotating bird-proof lightning protection device and synchronously generate a phase disturbance excitation signal to achieve dynamic speed control. The calculation formula of the rotational angular velocity is as follows: ; in: is the current target rotation angular velocity of the rotating bird-proof lightning protection device; is the phase perturbation intensity factor; is the disturbance phase.

[0013] In a second aspect, the present invention further provides an ecological environment protection system for power transmission and transformation projects, comprising: The image acquisition module is configured to: acquire image data of a preset area on the power transmission and transformation project line; The target recognition and tracking module is configured to: recognize the bird targets in the image data, generate a flight trajectory sequence of the target individual, and establish an edge contour recognition area for each bird; The dynamic interest region extraction module is configured to: extract the range of bird morphological changes based on the flight trajectory sequence and edge contour recognition area, and delineate the dynamic interest region; The wing-flapping trajectory modeling module is configured to: continuously model the vertical pixel position changes of the bird's wings within the dynamic region of interest, extract the pixel displacement data on the time axis, and generate a pixel displacement-time curve; The frequency feature extraction module is configured to: perform fast Fourier transform on the pixel displacement-curve, map the original trajectory signal from the time domain to the frequency domain, extract the main frequency component in the spectrum and establish an analysis set; The behavior analysis and risk assessment module is configured to extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set, analyze the extracted key features within the monitoring window, evaluate the frequency and changing trends of bird wing movements, and quantify bird strike behavior; The adaptive linkage control module is configured to adjust the rotation rate of the rotating bird-proof and lightning-proof device according to the activity frequency and change trend of the bird's wings and the principle of co-frequency interference.

[0014] In a third aspect, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the ecological environment protection method for power transmission and transformation projects described in the first aspect.

[0015] In a fourth aspect, the present invention also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, the steps of the ecological environment protection method for power transmission and transformation projects described in the first aspect are implemented.

[0016] In a fifth aspect, the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the steps of the ecological environment protection method for power transmission and transformation projects described in the first aspect.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention identifies bird targets in image data, generates a flight trajectory sequence of the target individual, and establishes an edge contour identification area for each bird; based on the flight trajectory sequence and the edge contour identification area, the range of bird morphological changes is extracted and a dynamic region of interest is defined; within the dynamic region of interest, vertical pixel position changes of the bird's wings are continuously modeled, pixel displacement data on the time axis is extracted, and a pixel displacement-time curve is generated; a fast Fourier transform is performed on the pixel displacement-curve, the original trajectory signal is mapped from the time domain to the frequency domain, and the main frequency component in the spectrum is extracted to establish an analysis set; key features reflecting the high-frequency wing flapping behavior of the bird are extracted from the analysis set, and the extracted key features are analyzed within the monitoring window to evaluate the activity frequency and change trend of the bird's wings and quantify the bird strike behavior; the rotation rate of the rotary bird-proofing and lightning protection device is adjusted according to the activity frequency and change trend of the bird's wings and the principle of co-frequency interference. By integrating multiple technical means such as high-frame rate visual perception, target tracking, individual wingbeat frequency extraction and frequency domain behavior modeling, it is possible to construct a frequency portrait of bird wingbeat behavior in real time, and adaptively adjust the operating state of the rotating device based on the principle of similar frequency interference, thereby triggering the most effective high-frequency phase shift disturbance at the moment the bird enters the critical area, effectively destroying its flight rhythm and path recognition ability, causing it to deviate from the flight path of the transmission equipment area, and being able to prevent birds from approaching the conductor area in time, significantly reducing the risk of birds colliding with conductors, causing short circuit tripping and biological casualties.

[0018] 2. The method of the present invention not only has strong advantages of intelligence, dynamism and precision, but also can achieve the balance between bird-repelling behavior and the goals of power grid security and ecological protection, and promote the technological upgrade of power transmission and transformation projects from passive protection to active ecological intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings constituting a part of the specification of this embodiment are used to provide a further understanding of this embodiment. The schematic embodiments and descriptions of this embodiment are used to explain this embodiment and do not constitute an improper limitation on this embodiment.

[0020] Figure 1 This is a flow chart of the method of Example 1 of the present invention; Figure 2 This is a schematic diagram of the system modules of Example 1 of the present invention. DETAILED DESCRIPTION

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0022] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0023] Example 1: Ecological and environmental protection for power transmission and transformation projects involves systematically identifying and assessing the potential impacts on natural ecosystems throughout the planning, construction, operation, and maintenance of transmission lines and substations. Scientific and rational engineering design and management measures are then implemented to minimize disruption and damage to vegetation, water bodies, soil, wildlife, and their habitats, thereby ensuring the stability and sustainability of regional ecosystems. Specific measures include rationally selecting line routes to avoid nature reserves, wetlands, and drinking water source protection areas; employing miniaturized tower bases and minimal excavation construction techniques to minimize surface disturbance; conducting post-construction ecological restoration projects, such as vegetation restoration and soil and water conservation; and implementing ecological monitoring and dynamic assessment mechanisms during operation. This protection concept not only helps comply with environmental regulations but is also a key component in achieving green and low-carbon development for power transmission and transformation projects.

[0024] In the existing ecological and environmental protection system for power transmission and transformation projects, to prevent and control the dual ecological and power system risks such as conductor short circuits, equipment failures, and bird deaths that may arise from migratory birds, wild birds inhabiting or flying over transmission corridors, rotating bird-proofing and lightning protection devices are commonly deployed along high-voltage transmission lines. These devices create visual disturbances to repel birds and reduce the probability of collisions. These devices typically utilize a mechanical rotating structure with a constant speed and rely on a low-speed electric drive to reflect natural light or create dynamic visual differences to disrupt the bird's visual recognition system, thereby achieving a repellent effect while also balancing energy consumption and extending the device's lifespan.

[0025] However, long-term operational monitoring and analysis of bird behavior patterns have revealed significant technical limitations in practical applications of existing fixed-speed rotating devices. Specifically, existing rotating bird-proofing and lightning protection devices are unable to respond and adjust based on the dynamic behavioral characteristics of birds approaching power transmission areas. During periods of rapid bird flight or peak migration, the visual disturbance caused by their constant rotation rate often lags behind the bird's approach, resulting in an insufficient visual interference time window, poor intervention accuracy and real-time performance, and difficulty in preventing birds from approaching the conductor area. This results in a high risk of bird strikes, short-circuit tripping, and biological casualties, seriously impacting the continuity, safety, and ecological compatibility of transmission line operations.

[0026] In order to solve at least one of the above problems, the present embodiment provides an ecological environment protection method for power transmission and transformation projects, which can sense the approaching dynamics of birds and adjust the rotation disturbance strategy in real time accordingly, so as to improve the pertinence and effectiveness of bird damage intervention; specifically, by integrating multiple technical means such as high-frame rate visual perception, target tracking, individual wing flapping frequency extraction and frequency domain behavior modeling, it is possible to construct a frequency portrait of bird wing flapping behavior in real time, and adaptively adjust the operating state of the rotating device according to the "frequency interference similarity principle", thereby triggering the most effective high-frequency phase shift disturbance at the moment when the bird enters the critical zone, effectively destroying its flight rhythm and path recognition ability, causing it to deviate from the flight path of the transmission equipment area, and significantly reducing the risks of bird strikes, arc breakdown and tripping faults, so as to solve the problems in the above background technology. The method in this embodiment includes the following steps: S1. Deploy an industrial visual camera system with high frame rate acquisition capability (frame rate ≥ 120 fps) at key intervention nodes in the transmission channel to capture the rapid flight behavior of birds within the effective intervention area of ​​the rotating bird-proofing and lightning protection device with high temporal resolution, forming a continuous intervention airspace image data stream; and realize image data acquisition of preset areas on the transmission and transformation project line.

[0027] Optionally, the industrial vision camera system should have wide-angle monitoring capabilities and be equipped with a low-latency image cache unit to support high-speed object recognition and real-time processing. By establishing a three-dimensional monitoring perspective of the airspace above and around the rotating bird-proof lightning protection device, all-weather image acquisition is performed on the effective intervention area of ​​the rotating bird-proof lightning protection device (e.g., the spatial area within a radius of 30 meters and an elevation angle of 45°), forming a continuous image data stream of the intervention airspace. This step provides a high-precision, low-latency image foundation for capturing bird flight paths, ensuring that subsequent image processing has sufficient temporal and spatial data support. Key intervention nodes can be identified as substations, transmission towers, or other nodes requiring bird repellent. Rotating bird-proof lightning protection devices can utilize integrated bird-proofing and lightning protection devices, new multifunctional lightning rods, movable stainless steel windmills, or other rotatable devices with bird-proofing capabilities.

[0028] S2. Use a target detection algorithm to identify bird targets in the image stream in real time. Use a target tracking algorithm between consecutive frames to generate a sequence of individual target flight paths, and simultaneously establish an edge contour recognition region for each bird. The target tracking algorithm can use Kalman filtering combined with SORT (Simple Online and Realtime Tracking) or DeepSORT (Deep Simple Online and Realtime Tracking).

[0029] S2.1. Using a target detection algorithm to perform real-time recognition of bird targets in an image stream, including the following steps: S2.1.1. Preprocess the collected video image stream, including image size normalization, brightness adjustment and noise suppression, to improve subsequent recognition accuracy.

[0030] S2.1.2. Input the preprocessed image frames one by one into the trained object detection model. A deep convolutional neural network is used to extract feature vectors from the image and generate candidate regions. The object detection model can use the YOLOv5 model or the Faster R-CNN (Faster Region-based Convolutional Neural Network) model.

[0031] S2.1.3. The model performs classification and bounding box regression on the candidate regions, and outputs the predicted box containing the bird target and its confidence score.

[0032] S2.1.4. Use the non-maximum suppression (NMS) algorithm to remove duplicate detection frames, retaining only the best detection results, and pass the bird target's location information (center coordinates, width, height, and category label, etc.) to subsequent behavior analysis or tracking units to achieve real-time recognition and precise positioning of bird targets in consecutive frames.

[0033] This process can complete single-frame processing within milliseconds and is suitable for high-frequency bird damage monitoring scenarios in power transmission channels.

[0034] S2.2. Generate a flight trajectory sequence of the target individual through a continuous frame-to-frame target tracking algorithm and simultaneously establish an edge contour recognition area. This generally includes the following steps: S2.2.1. In each frame, call the target detection algorithm to identify the bounding box information of the bird target, including the position coordinates and size.

[0035] S2.2.2. Initialize the Kalman filter model and construct a state vector (including position, velocity, etc.) for each detected bird target to predict the moving position of the bird target in the next frame.

[0036] S2.2.3. Use the SORT or DeepSORT algorithm to perform data association, match the prediction results with the current frame detection results through the Hungarian algorithm, assign a stable ID to each bird target, and achieve individual tracking.

[0037] S2.2.4. In each frame, perform secondary extraction of the bird's contour area based on the bounding box coordinates of the matched bird target, and use the Canny edge detection or GrabCut algorithm to refine the extraction of the bird's edge contour.

[0038] S2.2.5. Combine the bird edge information with unique IDs in chronological order to generate a stable flight trajectory sequence and a corresponding set of edge recognition regions, providing structured input data for subsequent wingbeat frequency analysis and behavioral modeling.

[0039] S3. Through image segmentation and motion estimation algorithms, the range of bird morphological changes is preliminarily extracted, the dynamic region of interest (ROI) is delineated, and the activity areas of the left and right wings of the bird are focused on to achieve individualized recognition and dynamic tracking of flying birds, providing feature positioning support for subsequent wingbeat trajectory extraction and frequency analysis.

[0040] Optionally, image segmentation and motion estimation algorithms are used to preliminarily extract the range of bird morphological changes and delineate dynamic regions of interest (ROIs), which typically include the following four steps: S3.1. Based on the target detection and tracking results, the bounding box area of ​​the target bird in each frame is extracted as the initial candidate area.

[0041] S3.2. Apply an image segmentation algorithm (such as GrabCut or DeepLabV3+ based on semantic segmentation) within the initial candidate area to extract the bird foreground mask and eliminate background interference.

[0042] S3.3. Use optical flow methods (such as Farneback Optical Flow) or inter-frame difference methods to calculate the pixel displacement inside the bird mask in consecutive frames and identify areas with significant dynamic changes.

[0043] S3.4. Based on the pixel motion intensity map, cluster analysis and spatial calibration are performed on the high-frequency activity areas of the left and right wings, and the dynamic region of interest (ROI) is delineated within the edge of the bird's body.

[0044] This process can accurately extract the key areas reflecting the bird's wing-flapping behavior, greatly improving the accuracy and stability of subsequent frequency extraction.

[0045] S4. Within the identified and tracked ROI area, the inter-frame difference method is used to continuously model the vertical pixel position changes of the bird's wings, extract the pixel displacement data on the time axis, generate a "pixel displacement-time" curve, reflect the dynamic morphological changes of the wings in continuous shooting frames, and establish a fluctuation feature data sequence.

[0046] Optionally, the vertical pixel displacement of the bird's wings is modeled using inter-frame difference within the identified and tracked ROI region, which typically includes the following four steps: S4.1. Perform frame-by-frame registration of the same ROI region in consecutive video frames to ensure pixel-level alignment and reduce the impact of background disturbance on the analysis results.

[0047] S4.2. Calculate the pixel grayscale difference between each pair of adjacent frames and identify areas with significant grayscale changes to detect dynamic edges caused by the up and down vibration of the wings.

[0048] S4.3. Select vertical scan lines or key points with high grayscale change frequency (e.g., grayscale change frequency higher than a preset value) within the ROI, and perform cumulative statistics on their grayscale differences along the time series to form a pixel position change sequence of wing movement.

[0049] S4.4. Map the pixel position change sequence into a "pixel displacement-time" curve, with the time frame as the horizontal axis and the pixel offset as the vertical axis, to reflect the periodic movement trend of the bird's wings in the vertical direction, laying the foundation for subsequent frequency domain analysis.

[0050] To ensure the purity of the trajectory signal, the motion noise can be reduced by combining the morphological filtering algorithm to ensure that the final trajectory data has stable periodic fluctuation characteristics.

[0051] S5. Perform a fast Fourier transform on the "pixel displacement-curve" in the time domain, map the original trajectory signal from the time domain to the frequency domain, extract the main frequency component in the spectrum to establish an analysis set, where the main frequency component is the average wingbeat frequency of the current bird during the detection period.

[0052] Optionally, a Fast Fourier Transform (FFT) is performed on the "pixel displacement-time" curve in the time domain to extract the main frequency component, which generally includes the following four steps: S5.1. Normalize the constructed pixel displacement sequence to remove the offset baseline and unify the amplitude scale to ensure the comparability of the spectral results.

[0053] S5.2. Apply the fast Fourier transform algorithm to map the time series signal from the time domain to the frequency domain to generate complex spectrum data.

[0054] S5.3. Calculate the modulus of the frequency spectrum to obtain an amplitude spectrum, and determine the main wingbeat frequency of the bird in the current time window by identifying the frequency peak with the highest energy concentration.

[0055] S5.4. The main wing beat frequency and its corresponding spectrum energy are taken as a set of feature points and stored in the analysis set for subsequent high-frequency behavior recognition and dynamic matching analysis of the rotation device drive parameters.

[0056] Step S5 achieves quantitative frequency extraction of the wing vibration rhythm, serving as a crucial bridge from visual signals to intervention-linked control. This improves frequency recognition accuracy, identifies key frequency bands using a frequency energy spectrum distribution diagram, and eliminates abnormal or spurious signals by setting a frequency confidence interval (e.g., 6 Hz to 20 Hz).

[0057] S6. Extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set through feature engineering technology. After in-depth analysis of the extracted features within the monitoring window, evaluate the activity frequency and change trend of bird wings and quantify the risk of bird strike behavior.

[0058] A high degree of tail extension following the main frequency in the spectrum typically indicates significant high-frequency energy tailing during the bird's wingbeat. This is often related to air disturbances, muscle vibrations, or fine-tuning movements generated by the bird's high-intensity, high-speed flight. Predatory or high-speed migratory birds, in particular, may superimpose short, high-frequency oscillations on top of their main-frequency wingbeats when crossing power transmission corridors to maintain attitude stability or rapidly change course. This complex high-frequency wingbeat pattern manifests in the spectrum as a high-amplitude energy extension to the right of the main frequency, known as "spectral tail stretch." This stretching characteristic not only indicates the bird's high speed and strong ability to respond to interference, but also implies a very short response window to static or low-dynamic intervention devices. If the system fails to intervene in a timely manner, the bird is likely to directly impact a conductor or insulator without changing its flight path. Therefore, a high degree of spectral tail extension can be a significant predictor of high-risk bird strikes.

[0059] Optionally, key features reflecting the high-frequency wing-flapping behavior of birds are extracted from the analysis set through feature engineering technology, where the extracted key features are the extension degree of the tail after the main frequency in the spectrum. Within the monitoring window, after in-depth analysis of the extracted features, a frequency transition index is generated. The frequency transition index is used to evaluate the activity frequency and change trend of bird wings and quantify the risk of bird strike behavior.

[0060] Optionally, within the monitoring window, the specific steps for generating a frequency transition index after performing an in-depth analysis of the tail extension following the main frequency in the spectrum are as follows: S6.1, in the set frequency window First, identify the main frequency point of the spectrum In the subsequent tail frequency segment, the frequency transition amplitude is extracted by constructing an energy difference sequence between adjacent frequency points in the tail region. The extracted expression is: ; in: Indicates frequency point The spectral energy density at Indicates frequency point The spectral energy density at Adjacent frequencies and The frequency transition amplitude between .

[0061] The purpose of step S6.1 is to measure the severity of the tail frequency energy disturbance. A larger differential rate indicates that the bird has non-steady-state or rapid frequency modulation behavior in the high-frequency vibration of the wings, capturing its discontinuous transition characteristics.

[0062] S6.2. Frequency jump amplitude of all tails The input is fed into the nonlinear amplification function, and the high-order hyperbolic function is used to enhance the responsiveness to the sudden change frequency. The frequency transition index is constructed comprehensively. The construction expression of the frequency transition index is: ; in: It is a frequency transition index used to quantify the overall tail disturbance trend; is the jump amplification coefficient, which controls the amplification sensitivity to small disturbances. ; is a nonlinear regulation index used to enhance the response to high-order mutations. ; n represents the end frequency index of the tail in the spectrum; p represents the index position of the main frequency point in the spectrum, which is used to define the starting point of the tail analysis, that is, the frequency point after the main frequency To begin with, the “tail region” on the right side of the entire spectrum is analyzed; The hyperbolic tangent function makes the frequency transition index sensitive to significant disturbances, but has a certain inhibitory ability to small changes, thus constructing a mutation response enhancement mechanism.

[0063] Step S6.2 is used to enhance the tail energy disturbance through nonlinear mapping, generating a value that can be monotonically mapped to the intensity of wingbeat activity. A larger frequency transition index indicates a higher frequency of wingbeat activity and a more dramatic change.

[0064] The frequency transition index, generated through an in-depth analysis of the tail extension following the dominant frequency in the spectrum within the monitoring window, reveals that a higher value indicates a higher bird wingbeat frequency, indicating a greater risk of the bird striking power transmission equipment. Conversely, a lower value indicates a lower risk of the bird striking power transmission equipment. This is because when a bird is flying at high speed, swooping across, or avoiding obstacles, its wingbeat frequency not only increases but is also accompanied by complex attitude adjustments and nonlinear perturbations. This produces a large number of jump components in the tail region following the dominant frequency in the spectrum, manifesting as a significant increase in the frequency transition index. In these conditions, the bird's control accuracy decreases and its reaction time shortens, making it very likely to deviate from its path or fail to avoid obstacles, significantly increasing the probability of striking wires or power equipment. Therefore, the frequency transition index, as a quantitative indicator of high-frequency, unstable flight behavior, has a strong ability to indicate impact risk.

[0065] S7. The current wing-flapping frequency characteristics of the birds obtained through analysis are used as input parameters and input into the linkage control unit of the rotary bird-proofing and lightning protection device. The rotation rate of the rotary bird-proofing and lightning protection device is dynamically adjusted according to the principle of similar frequency interference (the principle of same-frequency interference) to form a high-frequency phase-shift disturbance. This provides the most effective behavioral intervention at the moment when the bird approaches the critical zone, reduces the risk of bird strikes, and realizes the integrated closed-loop bird-repelling control of "perception-judgment-response".

[0066] Optionally, the rotation rate of the rotating bird-proof lightning protection device is dynamically adjusted according to the principle of similar frequency interference to form a high-frequency phase shift disturbance. The specific steps are as follows: S7.1. Based on the bird's wingbeat frequency and its corresponding frequency transition index, a dynamic target interference frequency is constructed to drive the rate adjustment of the rotating bird-proof lightning protection device. The construction formula is: ; in: is the bird's current main wingbeat frequency, derived from the main frequency of the spectrum; It is a frequency transition index that quantifies the degree of high-frequency disturbance; It is a nonlinear perturbation amplification function that enhances the speed regulation response when the transition is severe; is the disturbance coefficient, which controls the disturbance frequency offset amplitude. ; is the dynamic target interference frequency.

[0067] Real-time calculation of the dynamic target interference frequency of the rotating device based on the flight status of the bird , making it close to the bird's main frequency but slightly offset, creating a phase shift phenomenon in the visual disturbance band, destroying the bird's stable perception of the environmental rhythm.

[0068] S7.2, calculate the dynamic target interference frequency , converted into the rotational angular velocity of the rotating bird-proof lightning protection device, and synchronously generates a phase disturbance excitation signal to achieve dynamic speed control. The calculation formula of the rotational angular velocity is as follows: ; in: is the current target rotation angular velocity of the rotating bird-proof lightning protection device; is the phase perturbation intensity factor, , controls the amplitude of periodic disturbance; To perturb the phase, the constant periodic rotation rhythm is disrupted by small amplitude jitters.

[0069] By adding dynamic target interference frequency The linkage control unit applied to the rotating bird-proof lightning protection device enables its rotation rate to produce an angular velocity output that is highly similar to the bird's wingbeat frequency but with a disturbance phase, further enhancing the interference effect and disrupting the movement synchronization of the bird's nervous system, thereby improving the avoidance success rate and bird-repelling efficiency.

[0070] The core function of step S7 is to implement an adaptive, coordinated intervention mechanism based on the perception of birds' dynamic behavioral characteristics. By using the analyzed bird's wingbeat frequency as a key input parameter, the rotating bird-proof lightning protection device dynamically adjusts its rate according to the "principle of similar frequency interference." This creates a strong, high-frequency, phase-shifted disturbance at the critical moment when the bird's flight path approaches the power transmission equipment. This disturbance not only creates a visually complex stimulus signal, preventing birds from achieving stable path recognition and visual adaptation in a short period of time, but also interferes with their neural rhythm control mechanisms, disrupting their instinctive flight rhythm synchronization and inducing them to deviate from their flight path or abandon their crossing attempts. This is especially true for high-frequency, high-speed birds of prey or migratory birds, which are most susceptible to blind crossings when their flight rhythm is stable. If the intervention device continues to rotate at a constant rate during this period, the interference effect is extremely limited. However, by precisely coupling the frequency and introducing a perturbation shift (perturbation phase), the intervention effect can be enhanced within a very short time window. In addition, this step is based on the "perception-judgment-response" integrated control closed-loop logic, ensuring that the system can not only react passively, but also make adaptive adjustments based on the identified comprehensive characteristics such as wing flapping frequency intensity, frequency transition trend, activity rhythm, etc., to achieve intelligent and forward-looking linkage control, effectively respond to birds of different flight states and types, significantly improve the pertinence, real-timeness and ecological safety of the bird-repellent system, and minimize safety risks such as bird collisions, power short circuits and equipment tripping. It is a key link in the evolution of traditional bird-repellent devices to intelligent perception-based active prevention and control systems.

[0071] The method in this embodiment, by integrating multiple technical means such as high-frame-rate visual perception, target tracking, individual wingbeat frequency extraction, and frequency-domain behavior modeling, can construct a frequency profile of bird wingbeat behavior in real time, and adaptively adjust the operating state of the rotating device based on the principle of similar frequency interference. This triggers the most effective high-frequency phase shift disturbance at the moment the bird enters the critical zone, effectively disrupting its flight rhythm and path recognition ability, causing it to deviate from the flight path of the transmission equipment area, and significantly reducing the risks of bird strikes, arc breakdown, and tripping faults. This method not only has strong advantages in intelligence, dynamism, and precision, but also can achieve a balance between the safety of the power grid and ecological protection goals in bird-repelling behavior, promoting the technical upgrade of power transmission and transformation projects from passive protection to active ecological intervention.

[0072] This embodiment also provides an ecological environment protection system for power transmission and transformation projects. The working method of the system is the same as the ecological environment protection method for power transmission and transformation projects in this embodiment. The system includes an image acquisition module, a target recognition and tracking module, a dynamic interest region extraction module, a wing flapping trajectory modeling module, a frequency feature extraction module, a behavior analysis and risk assessment module, and an adaptive linkage control module.

[0073] The image acquisition module captures the flight behavior of birds within the effective intervention area of ​​the rotating bird-proof lightning protection device through the industrial visual camera system of the power transmission channel, forming a continuous intervention airspace image data stream; The target recognition and tracking module uses a target detection algorithm to perform real-time recognition of bird targets in the image stream, and generates a flight trajectory sequence of the target individual through a continuous inter-frame target tracking algorithm, and simultaneously establishes an edge contour recognition area for each bird; The dynamic region of interest extraction module uses image segmentation and motion estimation algorithms to preliminarily extract the range of bird morphological changes, delineate dynamic regions of interest, and focus on the active areas of the left and right wings of the bird, thereby achieving individual recognition and dynamic tracking of flying birds. The wing-flapping trajectory modeling module uses the inter-frame difference method to continuously model the vertical pixel position changes of the bird's wings within the identified and tracked ROI area, extracts its pixel displacement data on the time axis, generates a "pixel displacement-time" curve, and establishes a fluctuation feature data sequence; The frequency feature extraction module performs a fast Fourier transform on the "pixel displacement-curve" in the time domain, maps the original trajectory signal from the time domain to the frequency domain, extracts the main frequency component in the spectrum to establish an analysis set, where the main frequency component is the average wingbeat frequency of the current bird during the detection period; The behavior analysis and risk assessment module uses feature engineering technology to extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set. After in-depth analysis of the extracted features within the monitoring window, it evaluates the frequency and changing trends of bird wing movements and quantifies the risk of bird strike behavior. The adaptive linkage control module inputs the analyzed current wing-flapping frequency characteristics of the bird as input parameters to the linkage control unit of the rotary bird-proof and lightning-proof device, and dynamically adjusts the rotation rate of the rotary bird-proof and lightning-proof device according to the principle of similar frequency interference, forming a high-frequency phase shift disturbance and reducing the risk of bird strikes.

[0074] Example 2: This embodiment provides an ecological environment protection system for power transmission and transformation projects, including: The image acquisition module is configured to: acquire image data of a preset area on the power transmission and transformation project line; The target recognition and tracking module is configured to: recognize the bird targets in the image data, generate a flight trajectory sequence of the target individual, and establish an edge contour recognition area for each bird; The dynamic interest region extraction module is configured to: extract the range of bird morphological changes based on the flight trajectory sequence and edge contour recognition area, and delineate the dynamic interest region; The wing-flapping trajectory modeling module is configured to: continuously model the vertical pixel position changes of the bird's wings within the dynamic region of interest, extract the pixel displacement data on the time axis, and generate a pixel displacement-time curve; The frequency feature extraction module is configured to: perform fast Fourier transform on the pixel displacement-curve, map the original trajectory signal from the time domain to the frequency domain, extract the main frequency component in the spectrum and establish an analysis set; The behavior analysis and risk assessment module is configured to extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set, analyze the extracted key features within the monitoring window, evaluate the frequency and changing trends of bird wing movements, and quantify bird strike behavior; The adaptive linkage control module is configured to adjust the rotation rate of the rotating bird-proof and lightning-proof device according to the activity frequency and change trend of the bird's wings and the principle of co-frequency interference.

[0075] The working method of the system is the same as the ecological environment protection method for power transmission and transformation projects in Example 1, and will not be repeated here.

[0076] Example 3: This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the ecological environment protection method for power transmission and transformation projects described in Example 1 are implemented.

[0077] Example 4: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, the steps of the ecological environment protection method for power transmission and transformation projects described in Example 1 are implemented.

[0078] Example 5: This embodiment provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the ecological environment protection method for power transmission and transformation projects described in Example 1 are implemented.

[0079] The above description is merely a preferred embodiment of this embodiment and is not intended to limit this embodiment. Those skilled in the art will readily appreciate that this embodiment may be modified and varied in various ways. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this embodiment shall be within the scope of protection of this embodiment.

Claims

1. A method for protecting the ecological environment in power transmission and transformation projects, characterized in that: include: Acquire image data of a preset area on a power transmission and transformation project line; Identifying bird targets in the image data, generating a flight trajectory sequence of the target individual, and establishing an edge contour identification area for each bird; Based on the flight trajectory sequence and edge contour recognition area, the bird's morphological change range is extracted and the dynamic interest area is delineated; In the dynamic region of interest, the vertical pixel position changes of the bird's wings are continuously modeled, the pixel displacement data on the time axis is extracted, and the pixel displacement-time curve is generated; Perform fast Fourier transform on the pixel displacement-curve, map the original trajectory signal from the time domain to the frequency domain, extract the main frequency component in the spectrum and establish an analysis set; Extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set. Analyze the extracted key features within the monitoring window to assess the frequency and changing trends of bird wing movements and quantify bird strike behavior. The rotation rate of the rotating bird-proof lightning protection device is adjusted according to the activity frequency and change trend of bird wings and the principle of co-frequency interference.

2. The ecological environment protection method for power transmission and transformation projects according to claim 1, characterized in that: The bird target recognition includes: preprocessing the image data; extracting feature vectors in the image and generating candidate regions through a deep convolutional neural network based on the preprocessed image data, classifying and judging the candidate regions and performing bounding box regression to obtain a predicted frame containing the bird target and its confidence score, and using a non-maximum suppression algorithm to remove duplicate detection frames.

3. The ecological environment protection method for power transmission and transformation projects according to claim 2, characterized in that: Generate a flight trajectory sequence of the target individual and establish the edge contour recognition area of ​​each bird, including: identifying the bounding box information of the bird target in each frame image, constructing a state vector for each detected bird target to predict the motion position in the next frame; matching the prediction result with the detection result of the current frame, assigning a stable ID to each target to achieve individual tracking; in each frame, perform a secondary extraction of the bird contour area based on the bounding box coordinates of the matched target, and combine edge detection to refine the extraction of the bird edge contour; combine the bird edge information with unique IDs in chronological order to generate a stable flight trajectory sequence and a corresponding set of edge recognition areas.

4. The ecological environment protection method for power transmission and transformation projects according to claim 1, characterized in that: Delineating dynamic regions of interest includes: extracting the bounding box area of ​​the target bird in each frame as the initial candidate area based on target detection and tracking results; extracting the bird foreground mask within the initial candidate area; identifying dynamically changing areas based on the pixel displacement within the bird mask in consecutive frames; clustering analysis and spatial calibration of the high-frequency activity areas of the left and right wings using the pixel motion intensity map, and delineating the dynamic region of interest within the edge of the bird.

5. The ecological environment protection method for power transmission and transformation projects according to claim 1, characterized in that: Modeling the vertical pixel displacement of bird wings includes: frame-by-frame registration of the same dynamic region of interest in consecutive video frames; calculating the pixel grayscale difference between each pair of adjacent frames; selecting vertical scan lines or key points with a grayscale change frequency higher than a preset value within the dynamic region of interest, and accumulating and statistically analyzing their grayscale difference values ​​along the time series to form a pixel position change sequence of wing movement; mapping the pixel position change sequence to a pixel displacement-time curve, with the time frame as the horizontal axis and the pixel offset as the vertical axis.

6. The ecological environment protection method for power transmission and transformation projects according to claim 1, characterized in that: Extracting the main frequency component includes: normalizing the constructed pixel displacement sequence, removing the offset baseline and unifying the amplitude scale; applying the fast Fourier transform algorithm to map the time series signal from the time domain to the frequency domain to generate complex spectrum data; performing modulus calculation on the spectrum to obtain the amplitude spectrum, and determining the main wingbeat frequency of the bird in the current time window by identifying the frequency peak with the highest energy concentration; and storing the main wingbeat frequency and its corresponding spectrum energy as a set of feature points in the analysis set.

7. The ecological environment protection method for power transmission and transformation projects according to claim 1, characterized in that: Within the monitoring window, the extracted key features are analyzed, including: generating a frequency transition index after an in-depth analysis of the extension of the tail after the main frequency in the spectrum: in the set frequency window, first identify the tail frequency segment after the main frequency point of the spectrum, and extract the frequency transition amplitude by constructing an energy difference sequence between adjacent frequency points in the tail area; input the frequency transition amplitudes of all tails into a nonlinear amplification function, and use a high-order hyperbolic function to enhance the responsiveness to the sudden change frequency to construct a frequency transition index.

8. The ecological environment protection method for power transmission and transformation projects according to claim 7, characterized in that: Adjusting the rotation rate of the rotating bird-proof lightning protection device includes: constructing a dynamic target interference frequency according to the bird's wingbeat frequency and its corresponding frequency transition index, which is used to drive the speed adjustment of the rotating bird-proof lightning protection device. The formula is: ； in: is the current main wingbeat frequency of the bird; is the frequency transition index; is the nonlinear perturbation amplification function; is the disturbance coefficient; is the dynamic target interference frequency; The calculated dynamic target interference frequency Convert it into the rotational angular velocity of the rotating bird-proof lightning protection device and synchronously generate a phase disturbance excitation signal to achieve dynamic speed control. The calculation formula of the rotational angular velocity is as follows: ； in: is the current target rotation angular velocity of the rotating bird-proof lightning protection device; is the phase perturbation intensity factor; is the disturbance phase.

9. An ecological environment protection system for power transmission and transformation projects, characterized in that: include: The image acquisition module is configured to: acquire image data of a preset area on the power transmission and transformation project line; The target recognition and tracking module is configured to: recognize the bird targets in the image data, generate a flight trajectory sequence of the target individual, and establish an edge contour recognition area for each bird; The dynamic interest region extraction module is configured to: extract the range of bird morphological changes based on the flight trajectory sequence and edge contour recognition area, and delineate the dynamic interest region; The wing-flapping trajectory modeling module is configured to: continuously model the vertical pixel position changes of the bird's wings within the dynamic region of interest, extract the pixel displacement data on the time axis, and generate a pixel displacement-time curve; The frequency feature extraction module is configured to: perform fast Fourier transform on the pixel displacement-curve, map the original trajectory signal from the time domain to the frequency domain, extract the main frequency component in the spectrum and establish an analysis set; The behavior analysis and risk assessment module is configured to extract key features reflecting the high-frequency wing flapping behavior of birds from the analysis set, analyze the extracted key features within the monitoring window, evaluate the frequency and changing trends of bird wing movements, and quantify bird strike behavior; The adaptive linkage control module is configured to adjust the rotation rate of the rotating bird-proof and lightning-proof device according to the activity frequency and change trend of the bird's wings and the principle of co-frequency interference.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of the ecological environment protection method for power transmission and transformation projects according to any one of claims 1 to 8 are implemented.

Images