A method and system for bird detection on power lines based on digital twins

Abstract

The present invention discloses a method and system for detecting birds on power transmission lines based on digital twins, comprising: collecting flight dynamics and behavioral data of birds; constructing a digital twin model of the power transmission line, introducing a bird behavior simulation algorithm into the digital twin model, and simulating the activities of birds around the power transmission line; inputting the collected video data into the digital twin model and fusing it with the simulated bird behavior; fusing the monitored bird activity data with the power transmission line information in the digital twin model to obtain the spatial relationship and interaction information between the birds and the power transmission line; based on the fusion results of the monitoring data and the digital twin model, determining the location and behavior of the birds and the power transmission line, assessing the bird damage risk, and identifying potential line tripping risk areas. The present invention can effectively address the limitations of traditional methods such as low data accuracy and time-consuming large-scale monitoring projects, thereby improving monitoring coverage and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission line monitoring, and in particular to a power transmission line bird detection method and system based on digital twins. Background Art

[0002] The presence and activities of birds around power transmission lines can pose a risk of bird damage. For example, behaviors such as roosting, nesting, resting, and droppings can cause short circuits, power outages, and even fires. Therefore, monitoring birds and understanding their flight dynamics and behavior can assist power companies in developing appropriate management and protection measures. Furthermore, due to their diverse migration patterns and behavioral patterns, some birds may conflict with power transmission lines. Therefore, bird monitoring can help predict areas of high bird damage risk and identify key areas requiring protective measures. Traditional bird monitoring methods have several limitations, including limited observers, high professional requirements for monitoring personnel, subjective judgment that can lead to inaccurate data, and the time-consuming nature of large-scale monitoring projects. Summary of the Invention

[0003] In order to solve the above problems, the present invention provides a method and system for detecting birds on power transmission lines based on digital twins.

[0004] To achieve the above object, the technical solutions adopted by the present invention are as follows:

[0005] In one aspect, the present invention discloses a method for detecting birds on power transmission lines based on digital twins, comprising the following steps:

[0006] Step 1: Conduct real-time monitoring of the area around the transmission line from both high-altitude and ground-based perspectives, collect bird flight dynamics and behavior data, and pre-process the collected video data;

[0007] Step 2: Build a digital twin model of the transmission line, including the line's geometry, geographic environment, and meteorological conditions. Introduce a bird behavior simulation algorithm into the digital twin model to simulate bird activity around the transmission line.

[0008] Step 3: Input the collected video data into the digital twin model and integrate it with the simulated bird behavior to achieve real-time monitoring of bird activity around the transmission lines. The monitored bird activity data is integrated with the transmission line information in the digital twin model to obtain the spatial relationship and interaction information between birds and transmission lines.

[0009] Step 4: Based on the fusion results of monitoring data and digital twin models, use computer vision and deep learning technologies to detect and identify birds, determine the location and behavior of birds and transmission lines, and combine the location and behavior of birds and transmission lines with the transmission line information in the digital twin model to assess the bird damage risk and identify potential line tripping risk areas.

[0010] Further: Step 1 includes:

[0011] The preprocessing of the collected video data includes:

[0012] Video stabilization: Use digital image processing technology to stabilize the collected video data and eliminate image jitter caused by factors such as equipment vibration or wind;

[0013] Denoising: Use image processing algorithms to denoise video data and eliminate interference and noise in the image;

[0014] Image segmentation: Use image segmentation algorithms to separate birds from the background in the video for subsequent bird identification and tracking.

[0015] Further: Step 2 includes:

[0016] Building a digital twin model of a transmission line includes:

[0017] The collected geometric structure, geographical environment, and meteorological condition data are processed and integrated to construct the input data required for the digital twin model. The digital twin model of the transmission line is constructed using digital twin technology. The geometric structure includes the location and height of the towers and conductors, the geographical environment includes the surrounding terrain and vegetation cover, and the meteorological conditions include wind speed and temperature.

[0018] The bird behavior simulation algorithm includes the following formula:

[0019] X t+1 =X t +δ x

[0020] X t+1 =Y t +δ y

[0021] Among them, (X t , Y t ) represents the position coordinates of the bird at time t, (δ x , δ y ) represents the horizontal and vertical displacement of the bird per unit time.

[0022] Further: Step 3 includes:

[0023] In the digital twin model, real-time input video data is processed, including video stabilization, target detection, and tracking, to extract bird targets and their motion trajectories in the video. By calculating the distance between the positions of the real-time monitored bird targets and the positions of the bird behaviors in the simulated data, the simulated bird behavior data generated from the digital twin model is fused with the real-time monitored bird target data:

[0024]

[0025] Among them, (x1, y1) represents the position coordinates of the bird target monitored in real time, and (x2, y2) represents the position coordinates of the simulated bird behavior data;

[0026] This is achieved through spatial position matching, where the coordinate positions in the bird activity data correspond to the line positions in the transmission line model;

[0027] By calculating the distance and relative position between birds and transmission lines, a spatial relationship model between birds and transmission lines is established;

[0028] The established spatial relationship model is integrated with bird activity data to obtain the spatial relationship and interaction information between birds and transmission lines.

[0029] Further: Step 4 includes:

[0030] Apply the target detection algorithm to identify the bird targets in the image. For the identified bird targets, use the convolutional neural network model to extract features to obtain the feature representation of the target;

[0031] Based on the acquired feature representation, a pre-trained convolutional neural network model is used to classify and identify birds. The identified bird targets are combined with the spatial information in the digital twin model to obtain the location and behavior of the birds and the power transmission lines.

[0032] Combining the location and behavior of birds and transmission lines with the transmission line information in the digital twin model, the spatial relationship between birds and transmission lines is calculated, and a regression model is used to predict the bird damage risk. Based on the established risk assessment model, potential bird damage risk areas are identified.

[0033] On the other hand, the present invention discloses a digital twin-based transmission line bird detection system, comprising:

[0034] Data acquisition and preprocessing module: This module conducts real-time monitoring of the area around the power transmission lines from both high altitude and ground perspectives, collects bird flight dynamics and behavior data, and preprocesses the collected video data.

[0035] Digital twin model construction module: Builds a digital twin model of the transmission line, including the line's geometry, geographical environment, and meteorological conditions. It also incorporates a bird behavior simulation algorithm into the digital twin model to simulate bird activity around the transmission line.

[0036] Real-time monitoring and data fusion module: This module inputs the collected video data into the digital twin model and integrates it with simulated bird behavior to achieve real-time monitoring of bird activity around transmission lines. The monitored bird activity data is then integrated with the transmission line information in the digital twin model to obtain spatial relationships and interaction information between birds and transmission lines.

[0037] Bird detection and risk assessment module: Based on the fusion results of monitoring data and digital twin models, computer vision and deep learning technologies are used to detect and identify birds, determine the location and behavior of birds and transmission lines, and combine the location and behavior of birds and transmission lines with the transmission line information in the digital twin model to assess bird damage risks and identify potential line tripping risk areas.

[0038] Compared with the prior art, the present invention has the following technical advances:

[0039] Compared with existing technologies, the digital twin-based bird detection method for power transmission lines has the following advantages:

[0040] Improved monitoring coverage and accuracy: Traditional methods are limited by the observer's location and subjective judgment, and cannot capture bird activity far from power lines or high in the sky. However, digital twin technology can achieve real-time monitoring of a wide area around power lines, improving monitoring coverage and accuracy.

[0041] Reduced labor costs and improved efficiency: Traditional methods require on-site bird detection experts to observe and record bird activity, which is labor-intensive and time-consuming. Digital twin-based methods, however, reduce the need for specialized personnel and improve monitoring efficiency through automated monitoring and data processing.

[0042] Improved data accuracy and consistency: Traditional methods are subject to subjective judgment and human error, resulting in low data accuracy and consistency. However, digital twin technology can automate the monitoring and data recording of bird activities, reducing human interference and improving data accuracy and consistency.

[0043] Support for large-scale monitoring projects: Traditional methods are time-consuming for large-scale monitoring projects, requiring waiting for on-site monitoring personnel to provide data for processing and analysis. However, digital twin-based methods can achieve rapid response and efficient processing for large-scale monitoring projects, thus supporting large-scale monitoring activities.

[0044] In summary, the present invention can effectively solve the limitations of traditional methods such as low data accuracy and time-consuming large-scale monitoring projects, improve monitoring coverage and accuracy, reduce labor costs and improve efficiency, while improving data accuracy and consistency, and supporting the implementation of large-scale monitoring projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0046] In the attached figure:

[0047] Figure 1 Flowchart of the present invention. DETAILED DESCRIPTION

[0048] The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments. The embodiments of the present invention will be described below with reference to the accompanying drawings.

[0049] Example 1

[0050] like Figure 1 As shown, the present invention discloses a method for detecting birds on power transmission lines based on digital twins, comprising the following steps:

[0051] Step 1: Conduct real-time monitoring of the area around the power transmission lines from both high-altitude and ground-based perspectives, collect bird flight dynamics and behavior data, and pre-process the collected video data.

[0052] Specifically, they include:

[0053] When collecting data, first determine the appropriate time and location to ensure coverage of all key areas around the transmission lines. In this embodiment, drones or surveillance cameras are used to monitor the area around the transmission lines in real time from both high altitude and ground perspectives. Drones can provide more comprehensive coverage, while surveillance cameras can observe bird behavior in more detail.

[0054] Preprocessing of the collected video data includes:

[0055] 1. Video stabilization: Use digital image processing technology to stabilize the collected video data, eliminate image jitter caused by factors such as equipment vibration or wind, and ensure a stable video image.

[0056] 2. Denoising: Use image processing algorithms to denoise video data, eliminating interference and noise in the image to improve the accuracy of subsequent analysis.

[0057] 3. Image Segmentation: Use image segmentation algorithms to effectively separate the birds from the background in the video, facilitating subsequent bird identification and tracking. Common image segmentation methods include pixel-based threshold segmentation and region-based segmentation.

[0058] During data preprocessing, the following formula can be used to evaluate video stability:

[0059]

[0060] Where N represents the number of video frames; x i Indicates the pixel value of the i-th frame image; Var(x i ) represents the variance of the pixel values ​​of the image frame.

[0061] Through the above data collection and preprocessing steps, high-quality and stable video data can be obtained, and effective noise removal and image segmentation processing can be performed on it, laying the foundation for subsequent bird behavior analysis and identification.

[0062] Step 2: Build a digital twin model of the transmission line, including the line's geometric structure, geographical environment, and meteorological conditions. Introduce a bird behavior simulation algorithm into the digital twin model to simulate bird activities around the transmission line.

[0063] Specifically, they include:

[0064] First, collect relevant data of the transmission lines, including information such as the geometric structure of the lines, geographical environment, and meteorological conditions. These data can be obtained through field measurements, satellite remote sensing, geographic information systems, etc.

[0065] The collected data is processed and integrated to construct the input data required for the digital twin model, including data cleaning, format conversion, coordinate system conversion and other operations to ensure data consistency and availability.

[0066] Based on the processed data, digital twin technology is used to construct a digital twin model of the transmission line. The model includes the geometric structure information of the line (such as the location and height of poles and conductors), geographical environment information (such as surrounding terrain, vegetation cover, etc.), meteorological conditions information (such as wind speed, temperature, etc.), and other factors affecting bird activities.

[0067] During the construction of the digital twin model, the following formula can be used to calculate the impact of the terrain factors of the transmission line on bird activities:

[0068]

[0069] Among them, TI represents the terrain index (TerrainIndex); A irepresents the area of ​​the i-th region; I i Represents the terrain index of the i-th region.

[0070] A bird behavior simulation algorithm is introduced into the digital twin model to simulate bird activities around transmission lines, including:

[0071] First, select appropriate bird behavior patterns based on the bird species and seasonal characteristics. For example, different bird species may have different flight altitudes, activity ranges, and nesting habits. Determine the bird's activity range and behavioral characteristics based on the geographical environment of the transmission line. For example, wetland areas may have more waterfowl activities, while mountainous areas may have more raptor activities. Use a random walk algorithm to simulate bird flight paths. Considering that bird flight behavior is generally random, a random walk model can be used to simulate bird flight paths. Adjust bird activity patterns based on seasonal changes. For example, in spring, there may be more bird nesting and breeding activities, while in autumn, there may be more bird migration activities. Generate bird behavior data based on the simulation algorithm and output it to the corresponding data format for subsequent use in the digital twin model.

[0072] The random walk algorithm is expressed as follows:

[0073] X t+1 =X t +δ x

[0074] X t+1 =Y t +δ y

[0075] Among them, (X t , Y t ) represents the position coordinates of the bird at time t, (δ x , δ y ) represents the horizontal and vertical displacement of the bird per unit time.

[0076] Step 3: Input the collected video data into the digital twin model and integrate it with the simulated bird behavior to achieve real-time monitoring of bird activities around the transmission lines; integrate the monitored bird activity data with the transmission line information in the digital twin model to obtain the spatial relationship and interaction information between birds and transmission lines.

[0077] Specifically, they include:

[0078] Real-time video data is transmitted to the digital twin model. In this embodiment, this can be achieved through equipment such as drones and surveillance cameras, ensuring real-time and continuous data. The digital twin model processes the real-time input video data, including video stabilization, target detection, and tracking. This allows the extraction of bird targets and their movement trajectories in the video.

[0079] The simulated bird behavior data generated from the digital twin model is fused with the bird target data monitored in real time. This step can be based on a matching algorithm to match the actual monitored bird targets with the simulated data to obtain more complete bird behavior information. Specifically, first, a matching algorithm is used to match the real-time monitored bird target data with the simulated bird behavior data. In this embodiment, this can be achieved by calculating the distance between the position of the real-time monitored bird target and the position of the bird behavior in the simulated data. For each bird target monitored in real time, the simulated bird behavior data closest to it is found and fused. In this embodiment, the distance can be weighted or the nearest simulated behavior can be simply selected as the fusion result. In this way, more complete bird behavior information can be obtained, including flight trajectories, perching behaviors, etc. in the simulated data. The matching algorithm is expressed by the following formula:

[0080]

[0081] Among them, (x1, y1) represents the position coordinates of the bird target monitored in real time, and (x2, y2) represents the position coordinates of the simulated bird behavior data.

[0082] The spatial relationship modeling of the monitored bird activity data and the transmission line information in the digital twin model was performed, including determining the distance, relative position, and interaction between the birds and the transmission lines. The established spatial relationship model was integrated with the bird activity data to obtain the spatial relationship and interaction information between the birds and the transmission lines.

[0083] Step 4: Based on the fusion results of monitoring data and digital twin models, use computer vision and deep learning technologies to detect and identify birds, determine the location and behavior of birds and transmission lines, and combine the location and behavior of birds and transmission lines with the transmission line information in the digital twin model to assess the bird damage risk and identify potential line tripping risk areas.

[0084] Specifically, they include:

[0085] The real-time monitoring data obtained in step 3, including video images and bird target data, is preprocessed using image enhancement and target detection to improve the accuracy and efficiency of bird detection. Based on the fusion of real-time monitoring data and the digital twin model, bird detection and identification are performed using computer vision and deep learning techniques. Combining the bird detection results with the power transmission line information in the digital twin model, the spatial relationship and interaction between birds and power transmission lines are analyzed. This includes determining the distance, location, behavior, and potential risk factors associated with these interactions.

[0086] Based on the analysis results, a bird damage risk assessment model is established, taking into account the factors affecting the stable operation of transmission lines caused by bird activities, such as bird density, distance between flight trajectory and line, and type of bird activity. In this embodiment, this can be achieved through statistical models, machine learning models, and other methods.

[0087] Based on the risk assessment model, potential bird damage risk areas are identified and warnings and suggestions are provided. In this embodiment, potential risk areas can be automatically identified by setting thresholds or using machine learning algorithms so that corresponding protective measures can be taken.

[0088] In the process of establishing the risk assessment model, the Bird Hazard Index (BHI) can be calculated using the following formula:

[0089]

[0090] Among them, N birds Indicates the number of birds monitored; A interaction represents the interaction area between birds and transmission lines; L line Indicates the length of the transmission line.

[0091] Through bird detection and risk assessment, computer vision and deep learning technologies can be used to process and analyze real-time monitoring data, thereby identifying potential bird damage risk areas and providing early warnings and recommendations.

[0092] Example 2

[0093] This embodiment discloses a digital twin-based power transmission line bird detection system, including:

[0094] Data acquisition and preprocessing module: This module conducts real-time monitoring of the area around the power transmission lines from both high altitude and ground perspectives, collects bird flight dynamics and behavior data, and preprocesses the collected video data.

[0095] Digital twin model construction module: Builds a digital twin model of the transmission line, including the line's geometry, geographical environment, and meteorological conditions. It also incorporates a bird behavior simulation algorithm into the digital twin model to simulate bird activity around the transmission line.

[0096] Real-time monitoring and data fusion module: This module inputs the collected video data into the digital twin model and integrates it with simulated bird behavior to achieve real-time monitoring of bird activity around transmission lines. The monitored bird activity data is then integrated with the transmission line information in the digital twin model to obtain spatial relationships and interaction information between birds and transmission lines.

[0097] Bird Detection and Risk Assessment Module: Based on the fusion of monitoring data and the digital twin model, computer vision and deep learning technologies are used to detect and identify birds, determine the location and behavior of birds and transmission lines, and combine these locations and behaviors with the transmission line information in the digital twin model to assess bird damage risks and identify areas with potential line tripping risks. These modules are used to implement the functions of Example 1.

[0098] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.

Claims

1. A method for detecting birds on power lines based on digital twins, characterized in that: The steps include: Step 1: Conduct real-time monitoring of the area around the transmission line from both high-altitude and ground-based perspectives, collect bird flight dynamics and behavior data, and pre-process the collected video data; Step 2: Build a digital twin model of the transmission line, including the line's geometry, geographic environment, and meteorological conditions. Introduce a bird behavior simulation algorithm into the digital twin model to simulate bird activity around the transmission line. Step 3: Input the collected video data into the digital twin model and integrate it with the simulated bird behavior to achieve real-time monitoring of bird activities around the transmission lines; The monitored bird activity data is integrated with the transmission line information in the digital twin model to obtain the spatial relationship and interaction information between birds and transmission lines; Step 4: Based on the fusion results of monitoring data and digital twin models, use computer vision and deep learning technology to detect and identify birds, determine the location and behavior of birds and transmission lines, and combine the location and behavior of birds and transmission lines with the transmission line information in the digital twin model to assess the bird damage risk and identify potential line tripping risk areas. The step 1 comprises: The preprocessing of the collected video data includes: Video stabilization: Use digital image processing technology to stabilize the collected video data and eliminate image jitter caused by factors such as equipment vibration or wind; Denoising: Use image processing algorithms to denoise video data and eliminate interference and noise in the image; Image segmentation: Use image segmentation algorithms to separate birds from the background in the video for subsequent bird identification and tracking; The step 2 includes: Building a digital twin model of a transmission line includes: The collected geometric structure, geographical environment, and meteorological condition data are processed and integrated to construct the input data required for the digital twin model. The digital twin model of the transmission line is constructed using digital twin technology. The geometric structure includes the location and height of the towers and conductors, the geographical environment includes the surrounding terrain and vegetation cover, and the meteorological conditions include wind speed and temperature. The bird behavior simulation algorithm includes the following formula: in,( , ) represents the position coordinates of the bird at time t, ( , ) represents the horizontal and vertical displacement of the bird per unit time; The step 3 comprises: In the digital twin model, real-time input video data is processed, including video stabilization, target detection, and tracking, to extract bird targets and their motion trajectories in the video. By calculating the distance between the positions of the real-time monitored bird targets and the positions of the bird behaviors in the simulated data, the simulated bird behavior data generated from the digital twin model is fused with the real-time monitored bird target data: in,( ) represents the position coordinates of the bird target monitored in real time, ( ) represents the position coordinates of the simulated bird behavior data; This is achieved through spatial position matching, where the coordinate positions in the bird activity data correspond to the line positions in the transmission line model; By calculating the distance and relative position between birds and transmission lines, a spatial relationship model between birds and transmission lines is established; The established spatial relationship model is integrated with bird activity data to obtain the spatial relationship and interaction information between birds and transmission lines.

2. The method for detecting birds on power lines based on digital twins according to claim 1, characterized in that: The step 4 comprises: Apply the target detection algorithm to identify the bird targets in the image. For the identified bird targets, use the convolutional neural network model to extract features to obtain the feature representation of the target; Based on the acquired feature representation, a pre-trained convolutional neural network model is used to classify and identify birds. The identified bird targets are combined with the spatial information in the digital twin model to obtain the location and behavior of the birds and the power transmission lines. Combining the location and behavior of birds and transmission lines with the transmission line information in the digital twin model, the spatial relationship between birds and transmission lines is calculated, and a regression model is used to predict the bird damage risk. Based on the established risk assessment model, potential bird damage risk areas are identified.

3. A digital twin-based power line bird detection system, using the power line bird detection method according to any one of claims 1-2, characterized in that: include: Data acquisition and preprocessing module: This module conducts real-time monitoring of the area around the power transmission lines from both high altitude and ground perspectives, collects bird flight dynamics and behavior data, and preprocesses the collected video data. Digital twin model construction module: Builds a digital twin model of the transmission line, including the line's geometry, geographical environment, and meteorological conditions. It also incorporates a bird behavior simulation algorithm into the digital twin model to simulate bird activity around the transmission line. Real-time monitoring and data fusion module: This module inputs the collected video data into the digital twin model and integrates it with simulated bird behavior to achieve real-time monitoring of bird activity around transmission lines. The monitored bird activity data is then integrated with the transmission line information in the digital twin model to obtain spatial relationships and interaction information between birds and transmission lines. Bird detection and risk assessment module: Based on the fusion results of monitoring data and digital twin models, computer vision and deep learning technologies are used to detect and identify birds, determine the location and behavior of birds and transmission lines, and combine the location and behavior of birds and transmission lines with the transmission line information in the digital twin model to assess the bird damage risk and identify potential line tripping risk areas.

Images