Wind power plant bird identification and protection method, device and equipment
By predicting the flight trajectory of birds and adjusting the speed of the wind motor, the problem of birds hitting the blades in the wind farm was solved, and the dual protection of birds and blades was achieved.
Patent Information
- Application Number
- CN202510184017.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-19
- Publication Date
- 2025-06-10
AI Technical Summary
Birds in wind farms are easily damaged by wind motor blades when flying. The existing bird repelling methods are not environmentally friendly and have poor results. Birds may damage the blades when impacting the blades.
Multiple sets of detection data of birds are obtained by collecting sensors, target feature information is extracted, birds are predicted, and the speed of the wind motor is adjusted according to the trajectory to prevent birds from hitting the blades.
It protects bird safety while protecting wind turbine blades, avoids bird damage and blade damage, and does not affect the normal flight of birds.
Smart Images
Figure CN120125883A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of bird identification, and particularly relates to a method, device and equipment for bird identification and protection in a wind farm. Background Art
[0002] With the progress and development of human society, the awareness of human beings for the protection of the natural ecological environment has been continuously enhanced. Birds, as friends of human beings, have been listed as protected objects. However, since human beings entered modern civilization, great changes have taken place in fields such as transportation and energy. For example, birds are clearly prohibited from activities in places such as aircraft carriers, airplanes, and power grids.
[0003] Among them, wind farms are located in vast areas, and there are often birds flying around. However, birds cannot distinguish the activity areas. When flying near a wind farm, they are easily affected by the airflow formed by the rotation of the blades and are injured by the blades, resulting in serious consequences. In order to protect birds and reduce the harm, bird repelling methods are often used. However, the bird repelling methods damage the activity space of birds, are not environmentally friendly, and the bird repelling effect is not very ideal. Especially in the wind farm areas involving the migration routes of birds, using the bird repelling method is likely to damage the original migration routes and breeding areas of birds. At the same time, when birds hit the fan blades, it may cause damage to the blades, thereby reducing the service life of the fan blades. Summary of the Invention
[0004] In view of this, the present invention provides a method, device and equipment for bird identification and protection in a wind farm, which can protect the safety of birds in the wind farm and the fan blades at the same time.
[0005] In a first aspect, the present invention provides a method for bird identification and protection in a wind farm. The method for bird identification and protection in a wind farm includes: continuously obtaining multiple groups of detection data of a target bird by using a collection sensor; extracting features from the multiple groups of detection data to obtain multiple groups of target feature information of the target bird, where the feature information includes target biological feature information, target position information, and target distance information; predicting the target flight trajectory of the target bird based on the multiple groups of target feature information; and setting the rotation speed of the wind farm fan based on the target flight trajectory and the blade position information of the wind farm fan, so that the wind farm fan rotates according to the rotation speed.
[0006] In this implementation manner, the present application collects and calculates multiple feature information of birds, and uses the multiple feature information to predict the flight trajectory of the target bird, which can realize the tracking and prediction of birds. On this basis, the running speed of the wind farm fan is adjusted according to whether the predicted flight trajectory of the bird is close to the fan blades of the wind farm. This method does not drive away birds and does not affect the normal flight trajectory of birds. Instead, it adjusts the wind farm fan based on the normal flight of birds, avoiding birds hitting the fan blades and causing blade damage, and can protect the safety of birds in the wind farm and the fan blades at the same time.
[0007] In an alternative embodiment, the acquisition sensors include a camera, an infrared thermal imager, and a Doppler radar device; continuously acquiring multiple sets of detection data of the target bird by using the acquisition sensors includes: acquiring image data based on the camera and the infrared thermal imager, and acquiring point cloud data based on the Doppler radar device; extracting features from the multiple sets of detection data to obtain multiple sets of target feature information of the target bird includes: extracting features from the image data to obtain the target biological feature information of the target bird; calculating the target position information and the target distance information of the target bird by combining the image data and the point cloud data.
[0008] In this implementation, the method of comprehensively calculating based on images and radar to extract the position and distance information of the target bird can improve the detection accuracy.
[0009] In an alternative embodiment, after extracting features from the multiple sets of detection data to obtain multiple sets of target feature information of the target bird, it further includes: obtaining the basic bird feature information from the bird database; based on the deep learning algorithm, performing feature matching between the multiple sets of target biological feature information and the basic bird feature information to obtain the species of the target bird; based on the species of the target bird, obtaining the behavior feature information of the target bird from the bird database, and the behavior feature information includes flight feature information and living feature information.
[0010] In this implementation, using the bird database as a comparison benchmark to judge the feature information of the target bird can improve the accuracy of the feature information. At the same time, using the neural network deep learning algorithm to identify the species, flight features, and living features of the target bird can analyze the bird habits in the wind farm in multiple dimensions, which is convenient for subsequent prediction of the bird flight trajectory.
[0011] In an alternative embodiment, predicting the target flight trajectory of the target bird based on multiple sets of target feature information includes: calculating the historical flight trajectory of the target bird at a historical moment based on the multiple sets of target position information and the target distance information; predicting the target flight trajectory of the target bird at a future moment based on the historical flight trajectory and the flight features.
[0012] In this implementation, predicting the future flight trajectory based on the historical flight trajectory can further improve the prediction accuracy.
[0013] In an alternative embodiment, setting the rotational speed of the wind farm turbines based on the flight trajectory and the blade position information of the wind farm turbines includes: obtaining the blade position information of each wind farm turbine; determining the target area based on the blade position information, and the target area is the area affected by the operating airflow when the wind farm turbines are operating; when the flight trajectory passes through the target area, reducing the rotational speed of the corresponding wind farm turbine.
[0014] In an alternative embodiment, after reducing the rotational speed of the wind farm turbines corresponding when the flight trajectory passes through the target area, the following steps are further included: determining whether the target bird enters the target area based on the position information of the target bird and the target area; when the target bird enters the target area, calculating the distance between the target bird and the wind farm turbines based on the position information of the target bird and the blade position information of the wind farm turbines; and when the distance between the target bird and the wind farm turbines is less than the distance threshold, shutting down the corresponding wind farm turbines.
[0015] In this implementation, different response methods are adopted according to whether the birds are approaching the dangerous area and whether they enter the dangerous area. When it is predicted that the birds are about to enter the dangerous area, the rotational speed of the wind farm turbines is reduced to reduce the impact of the turbine airflow on the birds; when the birds are within the dangerous area and close to the turbine blades, shutting down the wind farm turbines can avoid damage to the birds by the wind farm turbines, prevent the birds from hitting the blades, and ensure the flight safety of the birds and the safety of the blades.
[0016] In a second aspect, the present invention provides a wind farm bird identification and protection device, which includes: a collection sensor for continuously collecting multiple groups of detection data of the target bird; a processing device connected to the collection sensor for extracting features from the multiple groups of detection data to obtain multiple groups of target feature information of the target bird, where the feature information includes target biological feature information, target position information, and target distance information; predicting the target flight trajectory of the target bird based on the multiple groups of target feature information; setting the rotational speed of the wind farm turbines based on the target flight trajectory and the blade position information of the wind farm turbines; and a wind farm turbine device, at least one blade of which includes a preset color area and is connected to the processing device for rotating according to the rotational speed.
[0017] In this implementation, adding a color area to the wind farm turbine blades and increasing the visibility of the blades can enable the birds to identify the operation of the turbines in advance, change their flight paths, and ensure flight safety.
[0018] In a third aspect, the present invention provides a computer device, which includes: a memory and a processor, which are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute the wind farm bird identification and protection method according to the first aspect or any corresponding embodiment thereof.
[0019] In a fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the wind farm bird identification and protection method according to the first aspect or any corresponding embodiment thereof.
[0020] Fifth aspect, the present invention provides a computer program product, including computer instructions for causing a computer to execute the wind farm bird identification and protection method according to the first aspect or any corresponding embodiment thereof as described above. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0022] Figure 1 is a schematic diagram of a wind farm bird identification and protection device according to an embodiment of the present invention;
[0023] Figure 2 is a schematic flowchart of a wind farm bird identification and protection method according to an embodiment of the present invention;
[0024] Figure 3 is a schematic flowchart of another wind farm bird identification and protection method according to an embodiment of the present invention;
[0025] Figure 4 is a schematic hardware structure diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.
[0027] In this embodiment, a wind farm bird identification and protection device is provided. Please refer to Figure 1 , Figure 1 is a schematic diagram of a wind farm bird identification and protection device according to an embodiment of the present invention. The wind farm bird identification and protection device includes a collection sensor 11, a processing device 22, and a wind farm fan device 33. Among them, the processing device 2 is respectively connected to the collection sensor 1 and the wind farm fan device 3.
[0028] The collection sensor 1 is used to continuously collect multiple groups of detection data of the target bird.
[0029] Specifically, set the acquisition area of the acquisition sensor 1, where the acquisition area is the surrounding area of the wind farm turbine equipment 3. Align the acquisition sensor 1 with the acquisition area, and use the acquisition sensor 1 to collect data of target birds flying or staying in the acquisition area in real time, obtaining multiple groups of detection data at consecutive moments.
[0030] In one implementation, the acquisition sensor 1 includes one or more of a camera, an infrared thermal imager, and a Doppler radar device 12.
[0031] Specifically, use the camera to capture multiple groups of visible light image data containing target birds in real time, use the infrared thermal imager to collect multiple groups of infrared image data containing target birds in real time, and use the Doppler radar device 12 to collect multiple groups of point cloud data of target birds in real time.
[0032] In another implementation, the acquisition sensor 1 includes an infrared-visible light integrated camera 11 that combines a camera and an infrared thermal imager, and a Doppler radar device 12. The infrared-visible light integrated camera 11 and the Doppler radar device 12 are respectively connected to the processing device 2.
[0033] Specifically, use the infrared-visible light integrated camera 11 to collect multiple groups of image data containing target birds in real time, and use the Doppler radar device 12 to collect multiple groups of point cloud data of target birds in real time.
[0034] Among them, the image data and the point cloud data contain time series data.
[0035] The acquisition sensor 1 sends the multiple groups of detection data collected to the processing device 2.
[0036] Specifically, the infrared-visible light integrated camera 11 sends multiple groups of image data to the processing device 2, and the Doppler radar device 12 sends multiple groups of point cloud data to the processing device 2.
[0037] The processing device 2 is used to receive multiple groups of detection data and perform feature extraction on the multiple groups of detection data to obtain multiple groups of target feature information of the target birds.
[0038] Among them, the target feature information includes target biological feature information, target position information, and target distance information.
[0039] Among them, the target coordinate system can be the image coordinate system of the infrared-visible light integrated camera 11, the radar coordinate system of the Doppler radar device 12, or the basic coordinate system.
[0040] The processing device 2 is also used to register the image data and the point cloud data before performing feature extraction on the multiple groups of detection data, and convert the image data and the point cloud data to the same coordinate system.
[0041] In one implementation, the image data is converted in the radar coordinate system according to the conversion parameters between the image coordinate system and the radar coordinate system; in another implementation, the point cloud data is converted in the image coordinate system according to the conversion parameters between the image coordinate system and the radar coordinate system; in another implementation, the image data is converted in the basic coordinate system according to the conversion parameters between the image coordinate system and the basic coordinate system, and the point cloud data is converted in the basic coordinate system according to the conversion parameters between the radar coordinate system and the basic coordinate system.
[0042] The processing device 2 is also used to extract features from the image data to obtain the target biometric information of the target bird.
[0043] Among them, the target biometric information is the biometric part information of the target bird. Such as the shape of the beak, the shape of the tail, the length of the legs, etc.
[0044] In one implementation, target recognition is performed on the image data to obtain the recognition frame of the target bird, and the target bird is segmented according to biological parts. Recognition is performed on the images of each segmented part. For example, beak recognition is performed on the image containing the bird's head to recognize the shape of the beak, and the beak shapes include hooked, flat, slender, etc.; leg length recognition is performed on the image containing the bird's body, etc.
[0045] In one implementation, target recognition is performed on multiple groups of image data based on a bird recognition algorithm, and the information of each biological part of the target bird is output. Among them, the bird recognition algorithm is a trained neural network algorithm. The specific training process is as follows: obtain multiple groups of bird image samples, divide the bird image samples into a training set and a test set, use the bird image samples in the training set as input data, and use the information of each part in the bird image samples as labels to train the neural network algorithm; then use the test set to test the trained neural network algorithm.
[0046] The processing device 2 is also used to calculate the target position information and target distance information of the target bird by combining the image data and the point cloud data.
[0047] Among them, the target position information is the position coordinates of the target bird based on the target coordinate system. The target distance information is the distance information between the target bird and the acquisition sensor 1.
[0048] In one implementation, based on the method combining Doppler ranging and monocular ranging, the distance information between the target bird and the acquisition sensor 1 is determined.
[0049] Specifically, preprocess the point cloud data, downsample the point cloud data, denoise the point cloud data using statistical filtering, and filter out the point cloud data that is not of interest using an irrelevant point filtering module.
[0050] The position of the target bird is detected in the image data using a target detection algorithm, and a rectangular bounding box of the target bird is obtained.
[0051] The processed point cloud data is clustered using a clustering algorithm to obtain multiple point cloud clusters. Each point cloud cluster is projected onto the image data, and the distance between the target bird and the acquisition sensor 1 is calculated by combining the point cloud clusters within the rectangular bounding box and the image data.
[0052] Among them, the method of comprehensively calculating based on images and radar to extract the position and distance information of the target bird can improve the detection accuracy.
[0053] The processing device 2 is also used to determine the behavioral characteristic information of the target bird.
[0054] Specifically, the processing device 2 obtains the basic bird characteristic information from the bird database.
[0055] Among them, the bird database is a collection database recording the basic data of birds and the knowledge data of local bird experts.
[0056] Among them, the basic bird characteristic information includes bird biological characteristic information, bird flight characteristic information, and bird living characteristic information. The bird biological characteristic information includes bird species, body size, biological part characteristics, such as the shape of the beak, the shape of the tail, the length of the legs, etc., such as migratory birds, resident birds, etc.; the bird flight characteristic information includes flight altitude, flight time, flight habits, etc.; the bird living characteristic information includes predation information, reproduction information, activity area information, activity time and season information, etc.
[0057] In one implementation, the above-mentioned extracted multiple sets of target biological characteristic information are matched with the basic bird characteristic information to obtain the species of the target bird. The flight characteristics and living characteristics corresponding to this type of bird are extracted from the bird database.
[0058] In one implementation, feature matching is performed based on a deep learning algorithm. Specifically, the deep learning algorithm is used to match multiple sets of target biological characteristic information with the basic bird characteristic information, and the bird species with the highest or greater than the matching degree threshold in the bird database is used as the species of the target bird. The flight characteristics and living characteristics corresponding to this type of bird are extracted from the bird database.
[0059] The processing device 2 is also used to predict the target flight trajectory of the target bird based on multiple sets of target characteristic information.
[0060] In one implementation, the target flight trajectory of the target bird is predicted based on the historical flight trajectory and flight characteristics.
[0061] Specifically, obtain multiple groups of target position information and corresponding target distance information of the target bird, and fit the historical flight trajectory of the target bird at historical moments in chronological order. Combine the historical flight trajectory with information such as the flight time and flight habits of the target bird to predict information such as the flight direction and flight speed of the target bird at the next moment, and obtain the position information at the next moment. Use the above method to predict the position information of the target bird at multiple future moments to obtain the target flight trajectory at future moments.
[0062] It can be understood that, in order to improve the accuracy of the target flight trajectory, predict the target flight trajectory within a certain future time, and at the same time, adjust the target flight trajectory according to the real-time position information of the target bird.
[0063] In one implementation, a neural network model is used to implement the prediction of the target flight trajectory of the target bird.
[0064] In this implementation, taking the bird database as a comparison benchmark to judge the characteristic information of the target bird can improve the accuracy of the characteristic information. At the same time, using the neural network deep learning algorithm to identify the species, flight characteristics, and living characteristics of the target bird can analyze the bird habits in the wind farm from multiple dimensions, facilitating the prediction of the bird flight trajectory. Predicting the future flight trajectory based on the historical flight trajectory can further improve the prediction accuracy.
[0065] The processing device 2 is also used to set the rotation speed of the wind farm fan device 3 based on the flight trajectory and the blade information of the wind farm fan.
[0066] Specifically, determine the target area according to the blade information, and when it is predicted that the flight trajectory of the target bird is about to pass through the target area, adjust the rotation speed of the wind farm fan device 3 corresponding to the target area.
[0067] In one implementation, for each wind farm fan in the wind farm fan device 3, set the target area.
[0068] Specifically, for a wind farm fan, obtain the blade information. Among them, the blade information includes the blade position information and the current blade rotation speed information. Calculate the target area affected by the fan air flow when the current wind farm fan is operating according to the blade position information and the rotation speed information. It can be understood that the areas affected by blades at different positions and blades with different rotation speeds are different.
[0069] When it is predicted that the target flight trajectory of the target bird is about to pass through the target area, reduce the rotation speed of the wind farm fan corresponding to the target area.
[0070] Further, based on the position information of the target bird and the target area, it is determined whether the target bird enters the target area. When the target bird enters the target area, the distance or distance change between the target bird and the blades of the wind farm turbines is calculated based on the position information of the target bird and the position information of the blades of the wind farm turbines. When the distance between the target bird and the wind farm turbines is less than the distance threshold or the distance decreases, the corresponding wind farm turbine is shut down.
[0071] In one implementation, after adjusting the rotational speed of the wind farm turbines corresponding to the target area, the target area is recalculated, and the target flight trajectory of the bird is detected again.
[0072] In one implementation, the target bird that enters the target area is continuously detected. When it is detected that the target bird is away from the target area, the wind speed of the wind farm turbine device 3 is set to the normal operating wind speed.
[0073] In this implementation, different response methods are adopted according to whether the bird is approaching the dangerous area and whether it enters the dangerous area. When it is predicted that the bird is about to enter the dangerous area, the rotational speed of the wind farm turbine is reduced to reduce the impact of the turbine airflow on the bird; when the bird is within the dangerous area and close to the turbine blade, the wind farm turbine is shut down, which can avoid the wind farm turbine from damaging the bird, avoid the bird hitting the blade, and ensure the flight safety of the bird and the safety of the blade.
[0074] The processing device 2 is also used to send the rotational speed to the wind farm turbine device 3.
[0075] The wind farm turbine device 3 is used to receive the rotational speed information and rotate according to the rotational speed.
[0076] In one implementation, a preset color area is selected in at least one blade of each wind farm turbine, and the preset color area is colored according to the preset color. Exemplarily, please refer to Figure 1 , a first preset color area 31 and a second preset color area 32 are provided on a single blade of the wind farm turbine device 3.
[0077] Exemplarily, the first preset color area 32 is colored black, and the second preset color area 32 is colored red.
[0078] In this implementation, by adding a color area to the wind farm turbine blade and increasing the visibility of the blade, birds can recognize the operation of the turbine in advance, change their flight paths, and ensure flight safety.
[0079] In one implementation, the bird identification and protection device for a wind farm further includes a battery device 4. The battery device is respectively connected to the processing device 2, the infrared and visible light integrated camera 11, and the Doppler radar device 12, and powers the processing device 2, the infrared and visible light integrated camera 11, and the Doppler radar device 12.
[0080] Among them, the battery device 4 can be a storage battery, a photovoltaic cell, or a device combining a photovoltaic module and a storage battery.
[0081] According to an embodiment of the present invention, an embodiment of a method for identifying and protecting birds in a wind farm is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.
[0082] In this embodiment, a method for identifying and protecting birds in a wind farm is provided. Figure 2 It is a flowchart of the method for identifying and protecting birds in a wind farm according to an embodiment of the present invention. It should be noted that if there are substantially the same results, this embodiment is not limited to Figure 2 the process order shown. As Figure 2 shown, the process includes the following steps:
[0083] Step S201, continuously collect multiple groups of detection data of the target bird using a collection sensor.
[0084] Step S202, perform feature extraction on the multiple groups of detection data to obtain multiple groups of target feature information of the target bird.
[0085] The feature information includes target biological feature information, target position information, and target distance information.
[0086] Step S203, predict the target flight trajectory of the target bird based on the multiple groups of target feature information.
[0087] Step S204, set the rotation speed of the wind farm fan based on the target flight trajectory and the blade position information of the wind farm fan, so that the wind farm fan rotates according to the rotation speed.
[0088] The method for identifying and protecting birds in a wind farm provided in this embodiment collects and calculates multiple characteristic information of birds, and predicts the flight trajectory of the target bird by using the multiple characteristic information, which can realize the tracking and prediction of birds. On this basis, according to whether the predicted flight trajectory of the bird is close to the wind turbine blades of the wind farm, the operating speed of the wind turbine of the wind farm is adjusted. This method does not drive away the birds and does not affect the normal flight trajectory of the birds. Instead, it adjusts the wind turbines of the wind farm based on the normal flight of the birds, avoiding the birds hitting the wind turbine blades and causing damage to the blades, and can protect the safety of the birds in the wind farm while protecting the wind turbine blades.
[0089] In this embodiment, a method for identifying and protecting birds in a wind farm is provided, which can be used for the above-mentioned mobile terminals, such as mobile phones, tablet computers, etc. Figure 3 It is a flowchart of the method for identifying and protecting birds in a wind farm according to an embodiment of the present invention. It should be noted that if there are substantially the same results, this embodiment does not Figure 3 be limited by the process sequence shown. As Figure 3 shown, the process includes the following steps:
[0090] Step S301, continuously obtain multiple groups of detection data of the target bird by using a collection sensor.
[0091] Specifically, the above step S301 includes:
[0092] Step S3011, collect image data based on a camera and an infrared thermal imager.
[0093] Step S3012, collect point cloud data based on a Doppler radar device.
[0094] Step S302, perform feature extraction on multiple groups of detection data to obtain multiple groups of target feature information of the target bird.
[0095] The feature information includes target biological feature information, target position information, and target distance information.
[0096] Specifically, the above step S302 includes:
[0097] Step S3021, perform feature extraction on the image data to obtain the target biological feature information of the target bird.
[0098] Step S3022, calculate the target position information and target distance information of the target bird by combining the image data and the point cloud data.
[0099] Obtain the basic biological feature information of the bird from the bird database, perform feature matching on multiple groups of target biological feature information and the basic bird feature information to obtain the species of the target bird; based on the species of the target bird, obtain the behavioral feature information of the target bird from the bird database, and the behavioral feature information includes flight features and living features.
[0100] Step S303: Predict the target flight trajectory of the target bird based on multiple groups of target feature information.
[0101] Specifically, the above step S303 includes:
[0102] Step S3031: Calculate the historical flight trajectory of the target bird at historical moments based on multiple groups of target position information and target distance information.
[0103] Step S3032: Predict the target flight trajectory of the target bird at future moments based on the historical flight trajectory and flight characteristics.
[0104] Step S304: Set the rotational speed of the wind farm turbines based on the target flight trajectory and the blade position information of the wind farm turbines, so that the wind farm turbines rotate according to the rotational speed.
[0105] Specifically, the above step S204 includes:
[0106] Step S3041: When the flight trajectory passes through the target area, reduce the rotational speed of the corresponding wind farm turbine.
[0107] Obtain the blade position information of each wind farm turbine;
[0108] Determine the target area based on the blade position information. The target area is the area affected by the operating airflow when the wind farm turbines are operating;
[0109] When the flight trajectory passes through the target area, reduce the rotational speed of the corresponding wind farm turbine.
[0110] Step S3042: When the distance between the target bird and the wind farm turbine is less than the distance threshold, turn off the corresponding wind farm turbine.
[0111] Based on the position information of the target bird and the target area, determine whether the target bird enters the target area; when the target bird enters the target area, calculate the distance between the target bird and the wind farm turbine based on the position information of the target bird and the blade position information of the wind farm turbine; when the distance between the target bird and the wind farm turbine is less than the distance threshold, turn off the corresponding wind farm turbine.
[0112] In this implementation, multiple characteristic information of birds is collected and calculated. Based on the comprehensive calculation method of images and radar, the position and distance information of the target birds is extracted, which can improve the detection accuracy. Taking the bird database as a comparison benchmark to judge the characteristic information of the target birds can improve the accuracy of the characteristic information. At the same time, using the neural network deep learning algorithm to identify the species, flight characteristics, and living characteristics of the target birds can analyze the bird habits in the wind farm from multiple dimensions, facilitating the prediction of the bird flight trajectory. Meanwhile, predicting the future flight trajectory based on the historical flight trajectory can further improve the prediction accuracy.
[0113] Different response methods are adopted according to whether the birds are close to or enter the dangerous area. When it is predicted that the birds are about to enter the dangerous area, the rotational speed of the wind farm turbines is reduced to reduce the impact of the turbine airflow on the birds. When the birds are within the dangerous area and close to the turbine blades, the wind farm turbines are shut down, which can prevent the wind farm turbines from damaging the birds, avoid the birds hitting the blades, and ensure the flight safety of the birds and the safety of the blades. This method does not drive away the birds and does not affect the normal flight trajectory of the birds. Instead, it adjusts the wind farm turbines based on the normal flight of the birds to avoid the birds hitting the turbine blades and causing blade damage, and can protect the safety of the birds in the wind farm while protecting the turbine blades.
[0114] The wind farm bird identification and protection device in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.
[0115] The embodiment of the present invention also provides a computer device having the above Figure 1 shown wind farm bird identification and protection device.
[0116] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of a computer device provided by an alternative embodiment of the present invention. As Figure 4As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories. Similarly, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a set of blade servers, or a multi-processor system). Figure 4 Taking one processor 10 as an example in
[0117] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a generic array logic, or any combination thereof.
[0118] Among them, the memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiments.
[0119] The memory 20 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device. In addition, the memory 20 can include high-speed random access memory and can also include non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 can optionally include a memory remotely set relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.
[0120] The memory 20 can include volatile memory, such as random access memory; the memory can also include non-volatile memory, such as flash memory, a hard disk, or a solid-state drive; the memory 20 can also include a combination of the above types of memory.
[0121] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 may be connected by a bus or other means. Figure 4 Taking the connection by bus as an example.
[0122] The input device 30 can receive input digital or character information and generate key signal inputs related to the user settings and function controls of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED), and a haptic feedback device (e.g., a vibration motor), etc. The above display device includes, but is not limited to, a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device may be a touch screen.
[0123] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented by downloading over a network from an original storage in a remote storage medium or a non-transitory machine-readable storage medium and to be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiment is implemented.
[0124] A part of the present invention can be applied as a computer program product, such as computer program instructions, which when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should be able to understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible by the computer.
[0125] Although embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims
1. A method for identifying and protecting birds in a wind farm, characterized in that: The method comprises: Use acquisition sensors to continuously obtain multiple sets of detection data of target birds; Performing feature extraction on the multiple groups of detection data to obtain multiple groups of target feature information of the target birds, wherein the feature information includes target biological feature information, target position information and target distance information; Predicting a target flight trajectory of the target bird based on multiple sets of target feature information; The rotation speed of the wind farm wind turbine is set based on the target flight trajectory and the blade position information of the wind farm wind turbine, so that the wind farm wind turbine rotates according to the rotation speed.
2. The method for identifying and protecting birds in a wind farm according to claim 1, characterized in that: The acquisition sensors include cameras, infrared thermal imagers and Doppler radar equipment; The method of continuously collecting multiple groups of detection data of target birds by using a collection sensor includes: Collecting image data based on the camera and the infrared thermal imager, and collecting point cloud data based on the Doppler radar device; The extracting features from the plurality of detection data to obtain the plurality of target feature information of the target birds comprises: Performing feature extraction on the image data to obtain the target biological feature information of the target bird; The target position information and the target distance information of the target bird are calculated in combination with the image data and the point cloud data.
3. The method for identifying and protecting birds in a wind farm according to claim 2, characterized in that: After extracting features from the plurality of detection data to obtain the plurality of target feature information of the target birds, the following further comprises: Obtain basic characteristics information of birds from the bird database; Based on a deep learning algorithm, feature matching is performed on multiple sets of target biological feature information and basic feature information of the bird to obtain the species of the target bird; Based on the type of the target bird, behavioral characteristic information of the target bird is obtained from the bird database, where the behavioral characteristic information includes flight characteristic information and life characteristic information.
4. The method for identifying and protecting birds in a wind farm according to claim 3, characterized in that: The predicting of the target flight trajectory of the target bird based on multiple groups of target feature information includes: Calculate the historical flight trajectory of the target bird at a historical moment based on multiple sets of the target position information and the target distance information; The target flight trajectory of the target bird at a future moment is predicted based on the historical flight trajectory and the flight characteristics.
5. The method for identifying and protecting birds in a wind farm according to claim 1, characterized in that: The step of setting the rotation speed of the wind farm wind turbine based on the flight trajectory and the blade position information of the wind farm wind turbine comprises: Acquiring blade information of each wind turbine in the wind farm; Determine a target area based on the blade information, the target area being an area affected by the operating airflow when the wind turbines in the wind farm are operating; When the flight trajectory passes through the target area, the rotation speed of the corresponding wind farm wind turbine is reduced.
6. The method for identifying and protecting birds in a wind farm according to claim 5, characterized in that: The step of setting the rotation speed of the wind farm wind turbine based on the flight trajectory and the blade position information of the wind farm wind turbine further includes: Determining whether the target bird has entered the target area based on the location information of the target bird and the target area; When the target bird enters the target area, the distance between the target bird and the wind farm wind turbine is calculated based on the position information of the target bird and the blade position information of the wind farm wind turbine; When the distance between the target bird and the wind farm wind turbine is less than a distance threshold, the corresponding wind farm wind turbine is turned off.
7. A bird identification and protection device for a wind farm, characterized in that: The device comprises: A collection sensor is used to continuously collect multiple sets of detection data of target birds; A processing device connected to the acquisition sensor, used to extract features from the multiple sets of detection data, obtain multiple sets of target feature information of the target bird, the feature information includes target biological feature information, target position information and target distance information; predict the target flight trajectory of the target bird based on the multiple sets of target feature information; set the speed of the wind farm wind turbine based on the target flight trajectory and the blade position information of the wind farm wind turbine; A wind turbine device in a wind farm, wherein at least one blade includes a preset color area and is connected to the processing device to rotate according to the rotation speed.
8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the wind farm bird identification and protection method according to any one of claims 1 to 5 by executing the computer instructions.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the wind farm bird identification and protection method according to any one of claims 1 to 5.
10. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions are used to enable a computer to execute the wind farm bird identification and protection method according to any one of claims 1 to 5.