A distributed bird-repelling method and terminal for substations based on bird soundprint characteristics

Abstract

The present invention discloses a substation distributed bird-repelling method and terminal based on bird voiceprint characteristics. A bird database is established according to pre-collected bird information, and a deep learning network is trained based on the bird database, so that the deep learning network can separate and identify bird voiceprint information, and the bird information includes bird voiceprint information and bird species; the voiceprint information and bird position information sent by a front-end detection device are received, and the bird voiceprint information in the voiceprint information is separated and identified through the deep learning network to obtain the corresponding bird species; the bird database is queried according to the bird species, and a repelling strategy for the corresponding bird species is generated according to the query result; the repelling strategy is sent to a distributed bird repellent closest to the position among multiple distributed bird repellents distributed at different positions according to the bird position information, and the distributed bird repellent drives the bird away according to the repelling strategy; the method is more targeted, ensures the effectiveness of the bird repellent, and also improves the bird repellent effect.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment, and in particular to a distributed bird-repelling method and terminal for a substation based on bird voiceprint features. Background Art

[0002] Substations, crucial components for the safe operation of power grids, typically occupy large areas and are often located in suburban areas or high in the mountains. This is because the transmission lines leading to the substations can impact the surrounding environment. Considering the open terrain and cost savings, these substations are typically built outdoors. With the continuous advancement of substation management, which has gradually transitioned from manned to unmanned stations, coupled with the surrounding ecological environment and growing awareness of protecting nature and wildlife, the number of various birds is increasing, and their range of activities is expanding. This has made "bird pests" the greatest threat to the safe operation of substations. Therefore, bird repellent devices are needed to repel birds from the substation area and reduce the threat they pose to safe operation.

[0003] Most existing bird-repelling devices can only emit fixed raptor calls or gunshots, and the position of the bird-repelling device is fixed. After a period of time, birds are likely to adapt to the fixed sounds emitted in the fixed position, resulting in unsatisfactory bird-repelling effects. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a distributed bird-repelling method and terminal for substations based on bird voiceprint characteristics, thereby improving the bird-repelling effect.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A distributed bird-repelling method for substations based on bird soundprint characteristics, comprising the following steps:

[0007] S1. Establishing a bird database based on pre-collected bird information, and training a deep learning network based on the bird database, so that the deep learning network can separate and identify bird voiceprint information, wherein the bird information includes the bird voiceprint information and the bird species;

[0008] S2. Receive the voiceprint information and bird location information sent by the front-end detection device, separate and identify the bird voiceprint information in the voiceprint information through the deep learning network, and obtain the corresponding bird species;

[0009] S3. querying the bird database according to the bird species, and generating a repelling strategy corresponding to the bird species according to the query result;

[0010] S4. Sending the driving away strategy to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, so that the distributed bird repellent drives away the bird according to the driving away strategy.

[0011] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0012] A distributed bird-repellent terminal for substations based on bird voiceprint features includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:

[0013] S1. Establishing a bird database based on pre-collected bird information, and training a deep learning network based on the bird database, so that the deep learning network can separate and identify bird voiceprint information, wherein the bird information includes the bird voiceprint information and the bird species;

[0014] S2. Receive the voiceprint information and bird location information sent by the front-end detection device, separate and identify the bird voiceprint information in the voiceprint information through the deep learning network, and obtain the corresponding bird species;

[0015] S3. querying the bird database according to the bird species, and generating a repelling strategy corresponding to the bird species according to the query result;

[0016] S4. Sending the driving away strategy to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, so that the distributed bird repellent drives away the bird according to the driving away strategy.

[0017] The beneficial effects of the present invention are as follows: the present invention identifies different bird species according to bird soundprints, generates corresponding repelling strategies according to different species, and selects the closest one of multiple bird repellents distributed at different positions to execute the repelling strategy according to the position of the bird, thereby avoiding the adaptability of bird repelling caused by adopting the same repelling strategy for all birds and the fixed position of the repelling device. At the same time, the present invention is more targeted, ensuring the effectiveness of bird repelling while also improving the bird repelling effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a distributed bird-repelling method for substations based on bird voiceprint features according to an embodiment of the present invention;

[0019] Figure 2 This is a structural diagram of a substation distributed bird-repellent terminal based on bird voiceprint features according to an embodiment of the present invention;

[0020] Figure 3 This is a partial structural diagram of a substation distributed bird-repellent terminal based on bird voiceprint features according to an embodiment of the present invention;

[0021] Description of labels:

[0022] 1. A distributed bird-repellent terminal for substations based on bird voiceprint characteristics; 2. A processor; 3. A memory. DETAILED DESCRIPTION

[0023] To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0024] Please refer to Figure 1 A distributed bird-repelling method for substations based on bird soundprint features comprises the following steps:

[0025] S1. Establishing a bird database based on pre-collected bird information, and training a deep learning network based on the bird database, so that the deep learning network can separate and identify bird voiceprint information, wherein the bird information includes the bird voiceprint information and the bird species;

[0026] S2. Receive the voiceprint information and bird location information sent by the front-end detection device, separate and identify the bird voiceprint information in the voiceprint information through the deep learning network, and obtain the corresponding bird species;

[0027] S3. querying the bird database according to the bird species, and generating a repelling strategy corresponding to the bird species according to the query result;

[0028] S4. Sending the driving away strategy to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, so that the distributed bird repellent drives away the bird according to the driving away strategy.

[0029] From the above description, it can be seen that the beneficial effects of the present invention are: the present invention identifies different bird species based on bird soundprints, and generates corresponding repelling strategies according to different species. At the same time, according to the position of the bird, the closest one of multiple bird repellents distributed at different positions is selected to execute the repelling strategy, avoiding the adaptability of bird repelling caused by adopting one repelling strategy for all birds and the fixed position of the repelling device. At the same time, it is more targeted, ensuring the effectiveness of bird repelling while also improving the bird repelling effect.

[0030] Furthermore, the bird information also includes food chain data;

[0031] The step S3 specifically includes:

[0032] S31, querying the bird database according to the bird species to obtain corresponding food chain data, and obtaining natural enemy information of the bird species according to the food chain data;

[0033] S32: acquiring corresponding expulsion audio according to the natural enemy information, and generating an expulsion strategy for playing the expulsion audio;

[0034] The step S4 is specifically as follows:

[0035] The expulsion strategy is sent to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, and the distributed bird repellent plays the expulsion audio in the expulsion strategy.

[0036] From the above description, it can be seen that the present invention searches for information about natural enemies of birds based on the food chain of the identified birds, selects the audio of their natural enemies' calls to scare and drive them away, and calls the nearest distributed bird scarer to play according to the location of the birds each time they are driven away, so as to avoid the same audio each time and avoid the sound source being in the same position each time, thereby reducing the possibility of birds adapting.

[0037] Furthermore, the step S31 is specifically as follows:

[0038] querying the bird database according to the bird species to obtain corresponding food chain data, and judging whether the bird species has natural enemies based on the food chain data; if the bird species has no natural enemies, the natural enemy information is set to none; otherwise, obtaining natural enemy information of the bird species; if the bird species has more than one natural enemy, obtaining information of one natural enemy at random;

[0039] The step S32 is specifically as follows:

[0040] Determine whether the natural enemy information is none. If so, obtain the sound of gunfire as the expulsion audio. Otherwise, obtain the corresponding species call as the expulsion audio according to the natural enemy information, and generate an expulsion strategy for playing the expulsion audio.

[0041] As can be seen from the above description, after querying the bird's food chain data in the database, a judgment is made. If the bird species is a bird of prey and has no natural enemies, the sound of gunfire is used as the driving-away audio. If the bird has many natural enemies, it is best not to choose one as the driving-away audio, so that each bird can find the corresponding driving-away audio, and random natural enemy audio is used for the same bird to reduce the possibility of bird adaptation.

[0042] Furthermore, the step S3 further includes the steps of:

[0043] S33, obtaining real-time sunlight intensity information from a light sensing device, obtaining sunlight direction information if the sunlight intensity is greater than a preset threshold, and adding a light drive-away strategy to the drive-away strategy, wherein the light drive-away strategy includes the bird position information and sunlight direction information;

[0044] The step S4 is specifically as follows:

[0045] According to the bird position information, the repelling strategy is sent to the distributed bird repellent that is closest to the position among multiple distributed bird repellents distributed at different positions. The distributed bird repellent plays the repelling audio in the repelling strategy and adjusts the angle of the reflective fan according to the bird position information and the sunlight direction information in the light repelling strategy, so that the reflective fan reflects sunlight to the bird position to repel the bird by light.

[0046] From the above description, it can be seen that in the case of sufficient sunlight, while using the calls of natural enemies to drive them away, a light driving away strategy is also used to scare the birds away by reflecting light, thereby further improving the bird driving effect.

[0047] Furthermore, the method further comprises the steps of:

[0048] S5. Obtain and display the device status information of each distributed bird-repellent device, accept the user's operation request, and send a mode adjustment request to the distributed bird-repellent device specified by the operation request. The distributed bird-repellent device switches between day and night modes according to the mode adjustment request. In the night mode, the distributed bird-repellent device will ignore the light-repelling strategy.

[0049] From the above description, it can be seen that the staff can switch the distributed bird-repellent equipment between day and night modes according to the equipment status and usage requirements, thereby enabling or disabling the light bird-repellent function, which is more suitable for user needs.

[0050] Please refer to Figure 2 A substation distributed bird-repellent terminal 1 based on bird voiceprint characteristics includes a processor 2, a memory 3, and a computer program stored in the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, the steps in the above embodiment 1 are implemented.

[0051] Please refer to Figure 1 , embodiment 1 of the present invention is:

[0052] A distributed bird-repelling method for substations based on bird soundprint characteristics, comprising the following steps:

[0053] S1. Establish a bird database based on pre-collected bird information, and train a deep learning network based on the bird database so that the deep learning network can separate and identify bird voiceprint information, wherein the bird information includes the bird voiceprint information, bird species, and food chain data.

[0054] In this example, we conducted a preliminary field survey of the substation's vicinity, recording visible bird species, numbers, morphological characteristics, bird voice patterns, habits, breeding cycles, nesting habits, population relationships, and food chain data. Based on this data, we established a bird database. Using this database, we used the bird voice patterns and corresponding bird species information to train a deep learning network, enabling rapid identification and separation of bird calls. Based on the survey results, we established distributed bird repellent devices within the substation, determining their specific installation locations and quantities to maximize their utilization, reduce costs, and achieve energy savings and efficiency gains.

[0055] In this embodiment, the bird voiceprint information of our province is collected. Our province has a wide variety of birds, with 557 species recorded. That is to say, among the birds naturally distributed in my country, more than 40% of the bird species can be seen in the wild in our province.

[0056] This project focused on annotating the sounds of 557 common bird species in the province, focusing on the birds that often nest at substations. The sounds were collected online, at completed substation projects, and at various bird and flower markets. At least 500 voice clips were collected and annotated based on the specific bird species to increase the success rate of bird voiceprint identification. The goal was to achieve a recognition rate of at least 95% within a 15-meter range in a field environment.

[0057] S2. Receive the voiceprint information and bird location information sent by the front-end detection device, separate and identify the bird voiceprint information in the voiceprint information through the deep learning network, and obtain the corresponding bird species.

[0058] In this embodiment, we use a highly sensitive sound recognition element to capture the sounds in the installation environment, use a deep learning algorithm to identify and separate the bird calls from the background noise, identify the bird species to which the bird calls belong, and obtain the source of the bird calls through radar.

[0059] S3. querying the bird database according to the bird species, and generating a repelling strategy corresponding to the bird species according to the query result;

[0060] The step S3 specifically includes:

[0061] S31, querying the bird database according to the bird species to obtain corresponding food chain data, and obtaining natural enemy information of the bird species according to the food chain data;

[0062] The step S31 is specifically as follows:

[0063] According to the bird species, the bird database is queried to obtain corresponding food chain data, and whether the bird species has natural enemies is determined based on the food chain data. If the bird species has no natural enemies, the natural enemy information is none; otherwise, the natural enemy information of the bird species is obtained. If the bird species has more than one natural enemy, the information of one of the natural enemies is randomly obtained.

[0064] In this example, we query bird species to obtain information about their natural enemies, allowing us to implement targeted removal strategies. For example, if we identify a sparrow as a bird, we might query the database to find information about its natural enemies, such as sparrowhawks and shrikes. However, for birds of prey like falcons, which may have no natural enemies, the natural enemy information would be "None."

[0065] S32: acquiring corresponding expulsion audio according to the natural enemy information, and generating an expulsion strategy for playing the expulsion audio;

[0066] The step S32 is specifically as follows:

[0067] Determine whether the natural enemy information is none. If so, obtain the sound of gunfire as the expulsion audio. Otherwise, obtain the corresponding species call as the expulsion audio according to the natural enemy information, and generate an expulsion strategy for playing the expulsion audio.

[0068] In this embodiment, for birds with many natural enemies such as sparrows, we can randomly select one of them. For example, when the bird species is a sparrow, we may randomly select a sparrowhawk and obtain the call of the sparrowhawk as the bird-repelling audio; and for birds with no natural enemy information, the sound of gunfire is selected as the driving-away audio.

[0069] The step S3 further comprises the steps of:

[0070] S33, obtaining real-time sunlight intensity information from a light sensing device, obtaining sunlight direction information if the sunlight intensity is greater than a preset threshold, and adding a light drive-away strategy to the drive-away strategy, wherein the light drive-away strategy includes the bird position information and sunlight direction information;

[0071] In this embodiment, we determine whether to formulate a light repelling strategy based on the light intensity. A light repelling strategy is automatically generated when the sunlight is strong. At the same time, we can also use a preset flashing device to repel birds at night according to the mode adjustment request described later.

[0072] The principle of the light bird repellent method is to use optical principles to achieve the purpose of bird repelling based on the habit of birds being afraid of light and color. By adding a flash device and a reflector, a variable light source is generated, which makes it impossible for birds to move normally nearby, thereby enhancing the bird repelling effect.

[0073] The principle of onomatopoeia bird-repelling is to produce different bird-repelling sounds for different types of birds, making it difficult for birds to adapt and further improving the bird-repelling effect.

[0074] In this embodiment, an ultrasonic bird-repelling pulse bird-repelling strategy is added. The ultrasonic bird-repelling pulse strategy stimulates the nervous system and physiological system of birds through ultrasonic pulse interference, causing physiological disorders without harming the birds.

[0075] The present invention combines multiple bird-repelling principles to achieve the effect of superimposed effects and complementary defects, thereby enhancing the bird-repelling effect.

[0076] S4. Sending the driving away strategy to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, so that the distributed bird repellent drives away the bird according to the driving away strategy.

[0077] The step S4 is specifically as follows:

[0078] According to the bird position information, the repelling strategy is sent to the distributed bird repellent that is closest to the position among multiple distributed bird repellents distributed at different positions. The distributed bird repellent plays the repelling audio in the repelling strategy and adjusts the angle of the reflective fan according to the bird position information and the sunlight direction information in the light repelling strategy, so that the reflective fan reflects sunlight to the bird position to repel the bird by light.

[0079] In this example, we employ a proximity-based, randomized strategy to activate distributed bird-repelling devices, ensuring a long-lasting and effective bird-repelling effect. Birds are repelled by playing repelling audio. The devices are equipped with rotatable reflective blades that scare birds away during periods of strong sunlight and shut down during periods of weak sunlight, when the solar cell output is weak, to conserve energy.

[0080] S5. Obtain and display the device status information of each distributed bird-repellent device, accept the user's operation request, and send a mode adjustment request to the distributed bird-repellent device specified by the operation request. The distributed bird-repellent device switches between day and night modes according to the mode adjustment request. In the night mode, the distributed bird-repellent device will ignore the light-repelling strategy.

[0081] In this embodiment, we collect status information of each distributed bird-repellent device, such as battery power, module self-test, working mode, etc., and can manually set the working mode of the new bird-repellent device, including manual adjustment of the working mode and switching between day and night modes to ensure the maximum bird-repellent effect during the high-incidence season of bird damage.

[0082] When the daytime mode is turned on, if the light is weak, the flash device is turned on to drive the bird away. In addition, an ultrasonic pulse device for driving away birds is added to the distributed bird-repelling device, which can be turned on and off through the mode adjustment request.

[0083] In addition, the following is some code to implement the onomatopoeia to drive away birds:

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093]

[0094] Please refer to Figure 2 and Figure 3 , the second embodiment of the present invention is:

[0095] A substation distributed bird-repellent terminal 1 based on bird voiceprint features includes a processor 2, a memory 3, and a computer program stored in the memory 3 and executable on the processor 2. When the processor executes the computer program, the steps in the above embodiment 1 are implemented.

[0096] In addition, in this embodiment, the substation distributed bird-repelling terminal based on bird voiceprint features also includes a power module and several interfaces, such as Figure 3 As shown, the interfaces include an HDMI interface, a VGA interface, 6 USB interfaces, 2 RJ45 interfaces and 3 COM interfaces.

[0097] In summary, the present invention provides a distributed bird repellent method and terminal for substations based on bird voiceprint characteristics. Different bird species are identified based on their voiceprints, and corresponding repellent strategies are generated based on the different species, including playing the calls of natural enemies and using light to repel birds. The repellent strategy is executed by selecting the closest bird repellent from multiple bird repellent devices distributed at different locations based on the bird's location. Different natural enemy calls are played for different birds, and light repellent is used to repel birds. This avoids the adaptability of using a single repellent strategy for all birds and the fixed location of the repellent device, which leads to the repelling of birds. It is more targeted, ensuring the effectiveness of the bird repellent while also improving the effect.

[0098] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A distributed bird-repelling method for substations based on bird soundprint features, characterized in that: Including steps: S1. Establishing a bird database based on pre-collected bird information, and training a deep learning network based on the bird database, so that the deep learning network can separate and identify bird voiceprint information, wherein the bird information includes the bird voiceprint information and the bird species; The training process of the deep learning network includes preprocessing the voiceprint information and inputting the preprocessed voiceprint information into the deep learning network; The pretreatment includes: Reversing the voiceprint information and then splicing it with the original voiceprint information to obtain spliced voiceprint information; Using the spliced voiceprint information to proceed to the next step with a preset probability; Performing a short-time Fourier transform on the voiceprint information to convert the time-domain audio signal into a frequency-domain spectrogram, and cropping the spectrogram to a preset fixed length to obtain preprocessed voiceprint information; The deep learning network is trained based on a learning rate scheduling strategy of piecewise constant decay and momentum SGD; S2. Receive the voiceprint information and bird location information sent by the front-end detection device, separate and identify the bird voiceprint information in the voiceprint information through the deep learning network, and obtain the corresponding bird species; S3. querying the bird database according to the bird species, and generating a repelling strategy corresponding to the bird species according to the query result; The selection of the driving away strategy includes audio driving away strategy, light driving away strategy and ultrasonic pulse bird driving away strategy; The light repelling strategy is activated at night according to the mode adjustment request, and generates a changing light source through a flash device and a reflector to repel birds; The ultrasonic pulse bird repellent strategy is activated according to the mode adjustment request, and emits non-damaging ultrasonic pulses to stimulate the birds and drive them away; The step S3 further comprises the steps of: Acquire real-time sunlight intensity information from a light sensing device, acquire sunlight direction information if the sunlight intensity is greater than a preset threshold, and add a light drive-away strategy to the drive-away strategy, wherein the light drive-away strategy includes the bird position information and sunlight direction information; S4, sending the driving away strategy to a distributed bird repellent closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, so that the distributed bird repellent drives away the bird according to the driving away strategy; The step S4 is specifically as follows: According to the bird position information, the repelling strategy is sent to the distributed bird repellent that is closest to the position among multiple distributed bird repellents distributed at different positions. The distributed bird repellent plays the repelling audio in the repelling strategy and adjusts the angle of the reflective fan according to the bird position information and the sunlight direction information in the light repelling strategy, so that the reflective fan reflects sunlight to the bird position to repel the bird by light.

2. The distributed bird-repelling method for substations based on bird soundprint characteristics according to claim 1 is characterized in that: The bird information also includes food chain data; The step S3 specifically includes: S31, querying the bird database according to the bird species to obtain corresponding food chain data, and obtaining natural enemy information of the bird species according to the food chain data; S32: acquiring corresponding expulsion audio according to the natural enemy information, and generating an expulsion strategy for playing the expulsion audio; The step S4 is specifically as follows: The expulsion strategy is sent to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, and the distributed bird repellent plays the expulsion audio in the expulsion strategy.

3. The distributed bird-repelling method for substations based on bird soundprint characteristics according to claim 2 is characterized in that: The step S31 is specifically as follows: querying the bird database according to the bird species to obtain corresponding food chain data, and judging whether the bird species has natural enemies based on the food chain data; if the bird species has no natural enemies, the natural enemy information is set to none; otherwise, obtaining natural enemy information of the bird species; if the bird species has more than one natural enemy, obtaining information of one natural enemy at random; The step S32 is specifically as follows: Determine whether the natural enemy information is none. If so, obtain the sound of gunfire as the expulsion audio. Otherwise, obtain the corresponding species call as the expulsion audio according to the natural enemy information, and generate an expulsion strategy for playing the expulsion audio.

4. The distributed bird-repelling method for substations based on bird soundprint characteristics according to claim 1 is characterized in that: Also includes the steps: S5. Obtain and display the device status information of each distributed bird-repellent device, accept the user's operation request, and send a mode adjustment request to the distributed bird-repellent device specified by the operation request. The distributed bird-repellent device switches between day and night modes according to the mode adjustment request. In the night mode, the distributed bird-repellent device will ignore the light-repelling strategy.

5. A distributed bird-repelling terminal for substations based on bird voiceprint features, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: S1. Establishing a bird database based on pre-collected bird information, and training a deep learning network based on the bird database, so that the deep learning network can separate and identify bird voiceprint information, wherein the bird information includes the bird voiceprint information and the bird species; The training process of the deep learning network includes preprocessing the voiceprint information and inputting the preprocessed voiceprint information into the deep learning network; The pretreatment includes: Reversing the voiceprint information and then splicing it with the original voiceprint information to obtain spliced voiceprint information; Using the spliced voiceprint information to proceed to the next step with a preset probability; Performing a short-time Fourier transform on the voiceprint information to convert the time-domain audio signal into a frequency-domain spectrogram, and cropping the spectrogram to a preset fixed length to obtain preprocessed voiceprint information; The deep learning network is trained based on a learning rate scheduling strategy of piecewise constant decay and momentum SGD; S2. Receive the voiceprint information and bird location information sent by the front-end detection device, separate and identify the bird voiceprint information in the voiceprint information through the deep learning network, and obtain the corresponding bird species; S3. querying the bird database according to the bird species, and generating a repelling strategy corresponding to the bird species according to the query result; The selection of the driving away strategy includes audio driving away strategy, light driving away strategy and ultrasonic pulse bird driving away strategy; The light repelling strategy is activated at night according to the mode adjustment request, and generates a changing light source through a flash device and a reflector to repel birds; The ultrasonic pulse bird repellent strategy is activated according to the mode adjustment request, and emits non-damaging ultrasonic pulses to stimulate the birds and drive them away; The step S3 further comprises the steps of: Acquire real-time sunlight intensity information from a light sensing device, acquire sunlight direction information if the sunlight intensity is greater than a preset threshold, and add a light drive-away strategy to the drive-away strategy, wherein the light drive-away strategy includes the bird position information and sunlight direction information; S4, sending the driving away strategy to a distributed bird repellent closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, so that the distributed bird repellent drives away the bird according to the driving away strategy; The step S4 is specifically as follows: According to the bird position information, the repelling strategy is sent to the distributed bird repellent that is closest to the position among multiple distributed bird repellents distributed at different positions. The distributed bird repellent plays the repelling audio in the repelling strategy and adjusts the angle of the reflective fan according to the bird position information and the sunlight direction information in the light repelling strategy, so that the reflective fan reflects sunlight to the bird position to repel the bird by light.

6. The substation distributed bird-repellent terminal based on bird voiceprint characteristics according to claim 5 is characterized in that: The bird information also includes food chain data; The step S3 specifically includes: S31, querying the bird database according to the bird species to obtain corresponding food chain data, and obtaining natural enemy information of the bird species according to the food chain data; S32: acquiring corresponding expulsion audio according to the natural enemy information, and generating an expulsion strategy for playing the expulsion audio; The step S4 is specifically as follows: The expulsion strategy is sent to a distributed bird repellent that is closest to the location among a plurality of distributed bird repellents distributed at different locations according to the bird location information, and the distributed bird repellent plays the expulsion audio in the expulsion strategy.

7. The substation distributed bird-repellent terminal based on bird voiceprint characteristics according to claim 6, characterized in that: The step S31 is specifically as follows: querying the bird database according to the bird species to obtain corresponding food chain data, and judging whether the bird species has natural enemies based on the food chain data; if the bird species has no natural enemies, the natural enemy information is set to none; otherwise, obtaining natural enemy information of the bird species; if the bird species has more than one natural enemy, obtaining information of one natural enemy at random; The step S32 is specifically as follows: Determine whether the natural enemy information is none. If so, obtain the sound of gunfire as the expulsion audio. Otherwise, obtain the corresponding species call as the expulsion audio according to the natural enemy information, and generate an expulsion strategy for playing the expulsion audio.

8. The substation distributed bird-repellent terminal based on bird voiceprint characteristics according to claim 5, characterized in that: Also includes the steps: S5. Obtain and display the device status information of each distributed bird-repellent device, accept the user's operation request, and send a mode adjustment request to the distributed bird-repellent device specified by the operation request. The distributed bird-repellent device switches between day and night modes according to the mode adjustment request. In the night mode, the distributed bird-repellent device will ignore the light-repelling strategy.

Images