Method, system, device and storage medium for monitoring environmental impact of wind farm establishment

Abstract

The application provides a wind farm environmental impact monitoring method, system, device and storage medium, wherein the method comprises the following steps: marking a risk area of an engineering site through a geographic information system to generate a baseline map; collecting sediments in the risk area when a preset condition is met, obtaining a contribution ratio of each source area as observation data through sediment fingerprint comparison, inferring a priori guess through meteorological data, and obtaining dust diffusion trajectories in the baseline map by correcting a forward diffusion model through Bayesian inversion; recording the positions of non-road mobile devices in real time, obtaining the tail gas concentration of each sub-region of the baseline map through grid monitoring, and implementing scheduling according to the contribution degree of the device when the tail gas concentration exceeds the standard; using a UAV to perform a monitoring task, carrying a sound collection device and a deep learning model for identifying flying animals on the UAV, solving an animal trajectory equation according to the relationship between the UAV trajectory, sound intensity and distance, and predicting the position of the flying animal, and adjusting the execution instruction of the UAV.

Description

Technical Field

[0001] This invention relates to the field of engineering environmental impact analysis technology, and in particular to a method, system, equipment and storage medium for monitoring the environmental impact of wind farm construction. Background Technology

[0002] With the rapid development of the wind power industry, the environmental impact of wind farm construction and operation has received increasing attention. During the construction phase, air pollution sources are mainly divided into dust pollution and exhaust emissions. Dust pollution is generated from different stages such as earthwork excavation, material transportation, and road construction. Exhaust emissions mainly originate from pollutants generated by the combustion of diesel fuel by construction machinery. During the operation phase, pollution risk reduction is mainly achieved through routine monitoring.

[0003] In related technologies, traditional methods rely on manual inspections throughout the entire process, resulting in low efficiency and poor effectiveness. Existing dust monitoring technologies mostly employ fixed-point monitoring or remote sensing inversion. Fixed-point monitoring struggles to distinguish the cumulative contributions of multiple sources and cannot quantify the actual impact of each construction area on key areas. Remote sensing inversion, on the other hand, relies on meteorological data and source strength assumptions, making accurate inversion difficult when wind direction statistics differ from actual settlement results, leading to significant deviations in dust dispersion trajectory prediction. Existing exhaust gas monitoring technologies mostly use total emission control or periodic testing, making it impossible to locate high-contribution machinery or prioritize emissions based on emission intensity and spatial distance when exceedances occur, resulting in delayed and ineffective local air quality control responses. While using drones for inspections during wind farm operation has become a trend, the high-frequency noise generated by drone rotors may disturb bird habitats or migrations. Existing technologies primarily rely on visual or radar methods to identify birds, which suffers from limitations due to lighting and weather conditions, as well as low detection rates for small birds. Furthermore, both drones and birds are moving targets, making it difficult to utilize drone microphones to collect real-time sound intensity changes and infer bird trajectories for proactive prediction and dynamic avoidance during flight.

[0004] Based on the above analysis of the development status of this technology field, the existing technologies lack intelligent monitoring methods for the environmental impact of wind farms that cover both the construction and operation phases, and lack an integrated closed-loop management solution that can realize dust diffusion trajectory inversion, coordinated control of mobile machinery exhaust, and dynamic avoidance of drones and birds. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, equipment, and storage medium for monitoring the environmental impact of wind farm establishment, in order to solve the aforementioned problems in the prior art.

[0006] According to a first aspect of the present invention, a method for monitoring the environmental impact of wind farm construction is provided, comprising: Risk areas in the engineering site selection are marked using a geographic information system, and a baseline map is generated. During the construction period of the wind farm, when the preset conditions are met, sediment is collected in the risk area. The contribution ratio of each source area is obtained by comparing the fingerprint characteristics of the sediment as observation data. Prior guesses are made based on meteorological data, and the dust diffusion trajectory in the baseline map is obtained by using Bayesian inversion to correct the forward diffusion model. During the wind farm construction period, the location of non-road mobile equipment is recorded in real time, and the exhaust gas concentration of each sub-area of ​​the baseline map is obtained by grid-based monitoring. When the exhaust gas concentration exceeds the standard, scheduling is implemented according to the contribution of the equipment. After the wind farm is established, drones are used to perform monitoring tasks. The drones are equipped with sound acquisition devices and deep learning models to identify flying animals. The animal trajectory equations are solved based on the relationship between the drone trajectory, sound intensity and distance, and the location of the flying animals is predicted. The drones are then adjusted to execute commands.

[0007] According to a second aspect of the present invention, a wind farm establishment environmental impact monitoring system is provided, comprising: The map creation module is used to mark risk areas in engineering site selection using a geographic information system and generate a baseline map; The dust monitoring module is used to collect sediment in the risk area during the wind farm construction period when preset conditions are met. The contribution ratio of each source area is obtained by comparing the fingerprint characteristics of the sediment as observation data. Prior guesses are made based on meteorological data, and the dust diffusion trajectory in the baseline map is obtained by using Bayesian inversion to correct the forward diffusion model. The gas monitoring module is used to record the location of non-road mobile equipment in real time during the wind farm construction period. It uses grid-based monitoring to obtain the exhaust gas concentration of each sub-area of ​​the baseline map. When the exhaust gas concentration exceeds the standard, scheduling is implemented according to the contribution of the equipment. The ecological monitoring module is used to perform monitoring tasks using drones after the wind farm is built. The drones are equipped with sound acquisition devices and deep learning models to identify flying animals. The module solves the animal trajectory equations based on the relationship between drone trajectory, sound intensity and distance, and predicts the location of flying animals, and adjusts the drone's execution commands accordingly.

[0008] According to a third aspect of the present invention, an electronic device is provided, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the wind farm establishment environmental impact monitoring method provided in the first aspect of the present disclosure.

[0009] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which an information transmission implementation program is stored, which, when executed by a processor, implements the steps of the wind farm establishment environmental impact monitoring method provided in the first aspect of the present disclosure.

[0010] The technical solution provided by the embodiments of the present invention has the following beneficial effects: by comparing the fingerprint features of sediments, the contribution ratio of each source region can be directly obtained as observation data to avoid source strength calibration errors, and the parameters of the forward diffusion model can be automatically corrected through Bayesian inversion to improve the simulation effect of dust diffusion trajectory; during gas monitoring, the emission size and spatial distance can be quantified simultaneously by the degree of equipment contribution, and the machinery with the highest contribution can be prioritized for scheduling, without the need for manual intervention and with fast response speed; after the wind farm is established, an animal trajectory equation model can be established through polynomials, without relying on any visual equipment, to simulate the animal flight curve that conforms to the laws of physical movement, avoiding interference to organisms during the monitoring process.

[0011] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of the method for monitoring the environmental impact of wind farm establishment according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the full-cycle monitoring architecture according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the environmental impact monitoring system for wind farm establishment according to an embodiment of the present invention; Figure 4 This is a schematic diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0015] Method Implementation Examples According to embodiments of the present invention, a method for monitoring the environmental impact of wind farm construction is provided. Figure 1This is a flowchart of the environmental impact monitoring method for wind farm establishment according to an embodiment of the present invention, such as... Figure 1 As shown, the method for monitoring the environmental impact of wind farm construction according to an embodiment of the present invention specifically includes: In step S110, the risk areas of the project site selection are marked using a geographic information system, and a baseline map is generated, specifically including: The project site is defined in the geographic information system by user input. The project site is located in the user-defined space where the wind farm is located. The delineation of project sites involves natural protected areas and migratory bird migration areas as prerequisites. In other words, the project site selection further delineates areas that are easily affected by wind farm recommendations and require special attention. Risk areas are automatically generated by spreading out at a preset distance from the boundary of the area where the preconditions are located. In this embodiment of the invention, the risk area is defined as a range of 500 meters from the boundary of the key area, thus obtaining the final baseline map.

[0016] In step S120, during the wind farm construction period, when preset conditions are met, sediment is collected in the risk area. The contribution ratio of each source area is obtained by comparing the fingerprint characteristics of the sediment as observation data. Prior guesses are made based on meteorological data, and the dust diffusion trajectory in the baseline map is obtained by using a Bayesian inversion correction model. Specifically, this includes: When the average wind speed in the baseline map exceeds the wind speed threshold, dust deposition collectors deployed at different locations in the risk area are used to collect sediment to simulate the dust diffusion trajectory at each location.

[0017] During the construction of a wind farm, the corresponding production processes in different areas vary, resulting in different dust pollution levels. For example, work area 1 is for excavating wind turbine foundations, where the soil is deep and has a high iron content, while work area 2 is for building access roads, where the soil is top-layered and has a lot of organic matter. Even if the differences between the source areas are not significant, the differences can be highlighted by manual marking.

[0018] Formula 1 is used to mathematically decouple the contribution scenario using an endmember hybrid model, and the contribution ratio of each source region is obtained by using the non-negative constrained least squares method: Formula 1; in, This represents the multidimensional fingerprint vector of the sediment observation sample. Indicates the first fingerprint vector of the source region, Indicates the first The contribution ratio of the source region Indicates the number of source regions. Represents the residual term; In this embodiment of the invention, the sediment content in source region 1 is 40%, and the sediment content in source region 2 is 60%.

[0019] The contribution ratio of each source region is used as observation data, and the wind rose diagram is used as meteorological data. Prior guesses are inferred based on the wind direction ratio in the meteorological data. For example, during the sampling period, the east wind accounts for 50%, the north wind also accounts for 50%, source region 1 is in the east and source region 2 is in the north. Therefore, the prior guesses are that the contribution of the sources are 50% respectively. It should be noted that the prior guesses are the proportion emitted from the source, while the proportion reflected by the sediment is the proportion that finally affects the key area after propagation through the path. Based on prior guesses, the predicted proportion is output through a forward diffusion model. The forward diffusion model is an existing prediction tool. Input information such as source location, source strength (i.e., dust generation amount), and diffusion parameters, and it can output how much dust will be received at key locations. If the difference between the predicted proportion and the observed data exceeds the preset proportion, for example, the output source area 1 sediment ratio is 60%, which is far greater than the actual 40%, then the parameters of the forward diffusion model are updated by Bayesian inversion, and continuously adjusted until the prediction is close to the fact, and the dust diffusion trajectory is output through the final forward diffusion model.

[0020] In step S130, during the wind farm construction period, the location of non-road mobile equipment is recorded in real time, and the exhaust gas concentration of each sub-region of the baseline map is obtained using grid-based monitoring. When the exhaust gas concentration exceeds the standard, scheduling is implemented based on the contribution of the equipment, specifically including: Real-time recording of the location of off-road mobile equipment, including excavators and trucks; The concentration of nitrogen oxides or carbon monoxide in the exhaust gas of the collected sub-region is used. When the exhaust gas concentration exceeds a preset threshold, the contribution of non-road mobile equipment is calculated using Formula 2. Formula 2; in, Indicates equipment The degree of contribution, Indicates equipment Emission intensity, Indicates equipment The distance from the nearest sub-interval where the exhaust gas concentration does not exceed the preset threshold is defined as the target area. The non-road mobile equipment is dispatched to the target area in descending order of contribution. If the exhaust gas concentration in the target area is about to exceed the preset threshold, it is dispatched to the second closest sub-interval that has not exceeded the preset threshold, and so on.

[0021] Preferably, in this embodiment of the invention, the exhaust gas concentration in a sub-region can be predicted using Formula 6: Formula 6; in, express Real-time predicted exhaust gas concentration Indicates the current moment. Indicates the prediction time interval. Indicates diffusion attenuation. This refers to meteorological influences, namely the effects of temperature, humidity, and atmospheric stability on the diffusion of pollutants.

[0022] In step S140, after the wind farm is established, a drone is used to perform a monitoring task. The drone is equipped with sound acquisition equipment and a deep learning model to identify flying animals. Based on the drone's trajectory, the relationship between sound intensity and distance, the animal trajectory equation is solved to predict the animal's position. The drone's execution commands are then adjusted, specifically including: Once the wind farm is completed, its environmental impact is relatively small, so routine monitoring with drones is sufficient. Drones can detect the impact of the wind farm on the atmospheric environment, water environment, sound environment, and solid waste during the operation period. In addition to monitoring, drones can also carry other tasks.

[0023] During the flight of the drone, high-frequency noise is the main component. Although the penetration is weak, it may disturb sound-sensitive organisms such as birds. High-frequency noise may disturb birds’ habitats or migrations, especially in ecological areas such as nature reserves and wetlands. Furthermore, birds are uncontrollable and cannot actively avoid drones. However, drones can form a unified scheduling through communication protocols. Therefore, the focus of this embodiment of the invention is on aerial biological avoidance. The original sound and actual sound intensity are collected using sound acquisition equipment; The original sound is input into a deep learning model, and sound features are extracted through the convolutional layers of the deep learning model. The sound features are then input into a channel consisting of a classification head and a voiceprint matching module. The classification head is a lightweight classifier that determines whether the current scene is mechanical noise or animal noise. If the result is animal noise, the corresponding type of flying animal is obtained by comparing with a module library in the voiceprint matching module. The module library stores the corresponding sound feature standards for each type, which is a serial dual-channel system.

[0024] Based on acoustic feature analysis, bird calls are easily confused with mechanical noises such as wind turbine blade rotation, drone rotor noise, mechanical friction noise, and wind turbine vortex noise. Therefore, this invention first uses a classification head to make a rough classification judgment, and then uses a module library to further compare which type of bird it is. Both modules obtain the judgment result based on sound features.

[0025] Obtain the drone trajectory, where the three-dimensional coordinates of the drone trajectory points are: Multiple time points correspond to trajectory points to form a sequence trajectory; Assuming the trajectory is smooth, the animal trajectory equation can be expressed as a polynomial using Formula 3: Formula 3; in, Represents the three-dimensional coordinates of the animal's trajectory point. Indicates time, and , This indicates the total number of data collection points. Indicates the coefficients to be solved; Formula 4 represents the relationship between sound intensity and distance, and Formula 5 represents the distance between the drone and the flying animal at the same moment: Formula 4; Formula 5; in, Indicates the predicted sound intensity. This indicates the original intensity of the sound emitted by a bird before propagation, representing the raw intensity of the sound produced by that type of flying animal. express The distance between the drone and the flying animal at any given time.

[0026] Formulas 3 to 5 describe the model building process. Given the drone trajectory and the corresponding actual sound intensity at each moment, the animal trajectory equation is solved. The specific solution process is as follows: Initialize the coefficients to be solved to calculate the predicted sound intensity Calculate the sum of the errors between the predicted sound intensity and the actual sound intensity at each time point. , This indicates the actual sound intensity that the drone can receive; Since the model involves a large number of variables to be solved, the coefficients to be solved are adjusted by gradient descent to obtain the final animal trajectory equation. Using animal trajectory equations to predict the three-dimensional coordinates of animal trajectory points, and if the distance between the three-dimensional coordinates of the animal trajectory points and the planned trajectory points of the UAV at the same time is within the risk range, it is determined that the interference with the biological environment will affect the execution of UAV commands. Preferably, the actual trajectory points and the predicted trajectory points can be marked on the baseline map for visualization.

[0027] Preferably, a solid waste traceability management function is established throughout the entire life cycle. Smart hazardous waste temporary storage bins with Internet of Things (IoT) functions are installed in the booster station. Through weight sensors and RFID tags, the source of hazardous waste generation, generation time, transfer manifest, and disposal destination are recorded.

[0028] The above technical solutions of the embodiments of the present invention will be illustrated with reference to the following accompanying drawings.

[0029] Figure 2 This is a schematic diagram of the full-cycle monitoring architecture of an embodiment of the present invention, as shown below. Figure 2As shown, the framework for monitoring the environmental impact of wind farms is established. The overall framework is divided into a basic layer, a construction phase layer, an operation phase layer, and an output layer, forming a closed-loop management system throughout the entire lifecycle.

[0030] In summary, addressing the existing problems, this invention presents a method for monitoring the environmental impact of wind farm construction. It directly obtains the contribution ratio of each source region through sediment fingerprint feature comparison as observation data, avoiding source strength calibration errors. Furthermore, it automatically corrects the parameters of the forward diffusion model through Bayesian inversion, improving the simulation effect of dust diffusion trajectories. During gas monitoring, it quantifies emission magnitude and spatial distance simultaneously based on the contribution of each device, prioritizing the scheduling of the highest-contributing machinery without manual intervention and with rapid response. After the wind farm is established, it uses a polynomial model to establish animal trajectory equations. Gradient descent fitting avoids extensive computation and effectively filters noise, simulating animal flight curves that conform to the laws of physical motion without relying on any visual equipment, thus avoiding interference with organisms during monitoring. Overall, this method covers intelligent environmental impact monitoring throughout the entire lifecycle of wind farm construction and operation.

[0031] System Implementation Examples According to embodiments of the present invention, a system for monitoring the environmental impact of wind farm construction is provided. Figure 3 This is a schematic diagram of the environmental impact monitoring system for wind farms according to an embodiment of the present invention, as shown below. Figure 3 As shown, the environmental impact monitoring system for wind farm construction according to an embodiment of the present invention specifically includes: Map creation module 30 is used to mark risk areas in engineering site selection using a geographic information system and generate baseline maps. Specifically, it is used for: In a geographic information system, the engineering site is delineated by user input. The delineation of the engineering site includes nature reserves and migratory bird migration areas as prerequisites. Risk areas are automatically generated by spreading out at a preset distance along the boundaries of the areas where the prerequisites are located, resulting in the final baseline map.

[0032] Dust monitoring module 32 is used during the wind farm construction period to collect sediment samples in risk areas when preset conditions are met. The contribution ratio of each source area is obtained by comparing sediment fingerprint characteristics as observation data. Prior guesses are made based on meteorological data, and a Bayesian inversion correction model is used to obtain the dust diffusion trajectory in the baseline map. Specifically, it is used for: When the average wind speed in the baseline map exceeds the wind speed threshold, dust deposition collectors deployed at different locations in the risk area are used to collect sediment to simulate the dust diffusion trajectory at each location.

[0033] Formula 1 is used to mathematically decouple the contribution scenario using an endmember hybrid model, and the contribution ratio of each source region is obtained by using the non-negative constrained least squares method: Formula 1; in, This represents the multidimensional fingerprint vector of the sediment observation sample. Indicates the first fingerprint vector of the source region, Indicates the first The contribution ratio of the source region Indicates the number of source regions. Represents the residual term; The contribution ratio of each source region is used as observation data, and the wind rose diagram is used as meteorological data. Prior guesses are inferred based on the wind direction ratio in the meteorological data. Based on prior guesses, the forward diffusion model outputs the predicted proportion. If the difference between the predicted proportion and the observed data exceeds the preset proportion, the parameters of the forward diffusion model are updated through Bayesian inversion. Finally, the dust diffusion trajectory is output through the forward diffusion model.

[0034] Gas monitoring module 34 is used to record the location of non-road mobile equipment in real time during the wind farm construction period. It uses grid-based monitoring to obtain the exhaust gas concentration of each sub-area of ​​the baseline map. When the exhaust gas concentration exceeds the standard, it implements scheduling based on the contribution of the equipment. Specifically, it is used for: Real-time recording of the location of off-road mobile equipment, including excavators and trucks; The concentration of nitrogen oxides or carbon monoxide in the exhaust gas of the collected sub-region is used. When the exhaust gas concentration exceeds a preset threshold, the contribution of non-road mobile equipment is calculated using Formula 2. Formula 2; in, Indicates equipment The degree of contribution, Indicates equipment Emission intensity, Indicates equipment The distance from the nearest sub-interval where the exhaust gas concentration does not exceed the preset threshold is defined as the target area. The non-road mobile devices are dispatched to the target area in descending order of their contribution level.

[0035] Ecological monitoring module 36 is used to perform monitoring tasks using drones after the wind farm is established. The drones are equipped with sound acquisition devices and deep learning models to identify flying animals. Based on the drone's trajectory, sound intensity, and distance relationship, the module solves the animal trajectory equation and predicts the animal's location, adjusting the drone's execution commands accordingly. Specifically, it is used for: The original sound and actual sound intensity are collected using sound acquisition equipment; The original sound is input into a deep learning model, and sound features are extracted through the convolutional layer of the deep learning model. The sound features are then input into a channel consisting of a classification head and a voiceprint matching module. The classification head determines whether the current scene is mechanical noise or animal noise. If the result is animal noise, the voiceprint matching module compares the corresponding type of flying animal with the module library.

[0036] Obtain the drone trajectory, where the three-dimensional coordinates of the drone trajectory points are: ; The equation for the animal's trajectory is expressed using a polynomial from Formula 3: Formula 3; in, Represents the three-dimensional coordinates of the animal's trajectory point. Indicates time, and , This indicates the total number of data collection points. Indicates the coefficients to be solved; Formula 4 represents the relationship between sound intensity and distance, and Formula 5 represents the distance between the drone and the flying animal at the same moment: Formula 4; Formula 5; in, Indicates the predicted sound intensity. This indicates the original intensity of the sound of the corresponding type of flying animal. express Real-time distance between drones and flying animals; The predicted sound intensity is calculated by initializing the coefficients to be solved. The sum of the error values ​​between the predicted sound intensity and the actual sound intensity at each time step is calculated. The coefficients to be solved are adjusted by gradient descent to obtain the final animal trajectory equation. The animal trajectory equation is used to predict the three-dimensional coordinates of the animal trajectory point. If the distance between the three-dimensional coordinates of the animal trajectory point and the planned trajectory point of the drone at the same time is within the risk range, it is determined that the biological environment is being disturbed and the drone's execution commands are being adjusted.

[0037] In summary, addressing the existing problems, this invention establishes an environmental impact monitoring system for wind farms. It directly obtains the contribution ratio of each source region through sediment fingerprint feature comparison as observation data, avoiding source strength calibration errors. Furthermore, it automatically corrects the parameters of the forward diffusion model through Bayesian inversion, improving the simulation effect of dust diffusion trajectories. During gas monitoring, it quantifies emission magnitude and spatial distance simultaneously based on the contribution of each device, prioritizing the scheduling of the highest-contributing machinery without manual intervention and with rapid response. After the wind farm is established, it uses a polynomial model to establish animal trajectory equations. Gradient descent fitting avoids extensive computation and effectively filters noise, simulating animal flight curves that conform to the laws of physical motion without relying on any visual equipment, thus avoiding interference with organisms during monitoring. Overall, it covers intelligent environmental impact monitoring throughout the entire lifecycle of wind farm construction and operation.

[0038] Electronic device examples Figure 4 This is a schematic diagram of an electronic device according to an embodiment of the present invention. The electronic device 400 may include at least one processor 410 and a memory 420. The processor 410 can execute instructions stored in the memory 420. The processor 410 is communicatively connected to the memory 420 via a data bus. In addition to the memory 420, the processor 410 can also be communicatively connected to an input device 430, an output device 440, and a communication device 450 via the data bus.

[0039] Processor 410 can be any conventional processor, such as a commercially available CPU. Processors may also include graphics processing units (GPUs), field-programmable gate arrays (FPGAs), systems-on-chips (SoCs), application-specific integrated circuits (ASICs), or combinations thereof.

[0040] The memory 420 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0041] In this embodiment of the present disclosure, the memory 420 stores executable instructions, and the processor 410 can read the executable instructions from the memory 420 and execute the instructions to implement all or part of the steps of the wind farm establishment environmental impact monitoring method in any of the exemplary embodiments described above.

[0042] Computer-readable storage medium embodiments In addition to the methods and apparatus described above, exemplary embodiments of this disclosure may also be a computer program product or a computer-readable storage medium storing the computer program product, the computer product including computer program instructions that can be executed by a processor to implement all or part of the steps described in the wind farm establishment environmental impact monitoring method of any of the exemplary embodiments described above.

[0043] Computer program products can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this application. Programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages, and scripting languages ​​(e.g., Python). The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0044] Computer-readable storage media may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media include: static random access memory (SRAM) having one or more electrically connected wires; electrically erasable programmable read-only memory (EEPROM); erasable programmable read-only memory (EPROM); programmable read-only memory (PROM); read-only memory (ROM); magnetic storage; flash memory; magnetic disk or optical disk; or any suitable combination thereof.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring the environmental impact of wind farm construction, characterized in that, include: Risk areas in the engineering site selection are marked using a geographic information system, and a baseline map is generated. During the construction period of the wind farm, when the preset conditions are met, sediment is collected in the risk area. The contribution ratio of each source area is obtained by comparing the fingerprint characteristics of the sediment as observation data. Prior guesses are made based on meteorological data, and the dust diffusion trajectory in the baseline map is obtained by using Bayesian inversion to correct the forward diffusion model. During the wind farm construction period, the location of non-road mobile equipment is recorded in real time, and the exhaust gas concentration of each sub-region of the baseline map is obtained by grid-based monitoring. When the exhaust gas concentration exceeds the standard, scheduling is implemented according to the contribution of the equipment. After the wind farm is established, drones are used to perform monitoring tasks. The drones are equipped with sound acquisition devices and deep learning models to identify flying animals. The animal trajectory equations are solved based on the relationship between the drone trajectory, sound intensity and distance, and the position of the flying animals is predicted. The drones are then adjusted to execute commands.

2. The method according to claim 1, characterized in that, The step of marking risk areas for engineering site selection and generating baseline maps using a geographic information system specifically includes: In a geographic information system, the engineering site is delineated by user input. The delineation of the engineering site includes natural protected areas and migratory bird migration areas as prerequisites. Risk areas are automatically generated by spreading out at a preset distance along the boundaries of the areas where the prerequisites are located, resulting in the final baseline map.

3. The method according to claim 1, characterized in that, The specific steps of collecting sediment in the risk area when preset conditions are met include: When the average wind speed in the baseline map exceeds the wind speed threshold, dust deposition collectors deployed at different locations in the risk area are used to collect sediment to simulate the dust diffusion trajectory corresponding to each location.

4. The method according to claim 1, characterized in that, The process of obtaining the contribution ratio of each source region through comparison of sediment fingerprint features as observation data, using meteorological data to infer prior guesses, and employing Bayesian inversion to correct the forward diffusion model to obtain the dust diffusion trajectory in the baseline map specifically includes: Formula 1 is used to mathematically decouple the contribution scenario using an endmember hybrid model, and the contribution ratio of each source region is obtained by using the non-negative constrained least squares method: Official 1; in, This represents the multidimensional fingerprint vector of the sediment observation sample. Indicates the first fingerprint vector of the source region, Indicates the first The contribution ratio of the source region Indicates the number of source regions. Represents the residual term; The contribution ratio of each source region is used as the observation data, and the wind rose diagram is used as the meteorological data. Prior guesses are inferred based on the wind direction ratio in the meteorological data. Based on the prior guess, the forward diffusion model outputs a predicted proportion. If the difference between the predicted proportion and the observed data exceeds a preset proportion, the parameters of the forward diffusion model are updated through Bayesian inversion, and the dust diffusion trajectory is output through the final forward diffusion model.

5. The method according to claim 1, characterized in that, The real-time recording of the location of non-road mobile equipment, the use of grid-based monitoring to obtain the exhaust gas concentration of each sub-region of the baseline map, and the implementation of scheduling based on the equipment's contribution when the exhaust gas concentration exceeds the standard specifically include: Real-time recording of the location of off-road mobile equipment, including excavators and trucks; The concentration of nitrogen oxides or carbon monoxide in the exhaust gas of the sub-region is collected. When the exhaust gas concentration exceeds a preset threshold, the contribution of non-road mobile equipment is calculated using Formula 2: Official 2; in, Indicates equipment The degree of contribution, Indicates equipment emission intensity, Indicates equipment The distance from the nearest sub-interval where the exhaust gas concentration does not exceed the preset threshold is defined as the target area. The non-road mobile devices are scheduled to the target area in descending order of their contribution level.

6. The method according to claim 1, characterized in that, The specific steps of equipping the drone with sound acquisition equipment and a deep learning model to identify flying animals include: The original sound and actual sound intensity are collected using sound acquisition equipment; The original sound is input into a deep learning model, and sound features are extracted through the convolutional layer of the deep learning model. The sound features are then input into a channel consisting of a classification head and a voiceprint matching module. The classification head determines whether the current scene is mechanical noise or animal noise. If the determination result is animal noise, the voiceprint matching module compares the corresponding type of flying animal with the module library.

7. The method according to claim 6, characterized in that, The process of solving the animal trajectory equation and predicting the flying animal's position based on the relationship between the drone trajectory, sound intensity, and distance, and then adjusting the drone's execution commands, specifically includes: Obtain the drone trajectory, where the three-dimensional coordinates of the drone trajectory points are: ; The equation for the animal's trajectory is expressed using a polynomial from Formula 3: Official 3; in, Represents the three-dimensional coordinates of the animal's trajectory point. Indicates time, and , This indicates the total number of data collection points. Indicates the coefficients to be solved; Formula 4 represents the relationship between sound intensity and distance, and Formula 5 represents the distance between the drone and the flying animal at the same moment: Official 4; Official 5; in, Indicates the predicted sound intensity. This indicates the original intensity of the sound of the corresponding type of flying animal. express Real-time distance between drones and flying animals; The predicted sound intensity is calculated by initializing the coefficients to be solved. The sum of the error values ​​between the predicted sound intensity and the actual sound intensity at each time point is calculated. The coefficients to be solved are adjusted by gradient descent to obtain the final animal trajectory equation. Using the animal trajectory equation to predict the three-dimensional coordinates of the animal trajectory point, if the distance between the animal trajectory point and the planned trajectory point of the UAV at the same time is within the risk range, it is determined that the biological environment is being disturbed and the UAV's execution command is being adjusted.

8. A wind farm environmental impact monitoring system, characterized in that, include: The map creation module is used to mark risk areas in engineering site selection using a geographic information system and generate a baseline map; The dust monitoring module is used to collect sediment in the risk area during the wind farm construction period when preset conditions are met. The contribution ratio of each source area is obtained by comparing the fingerprint characteristics of sediment as observation data. Prior guesses are made based on meteorological data, and the dust diffusion trajectory in the baseline map is obtained by using Bayesian inversion to correct the forward diffusion model. The gas monitoring module is used to record the location of non-road mobile equipment in real time during the wind farm construction period. It uses grid-based monitoring to obtain the exhaust gas concentration of each sub-area of ​​the baseline map. When the exhaust gas concentration exceeds the standard, scheduling is implemented according to the contribution of the equipment. The ecological monitoring module is used to perform monitoring tasks using drones after the wind farm is built. The drones are equipped with sound acquisition devices and deep learning models to identify flying animals. The module solves the animal trajectory equations based on the relationship between the drone trajectory, sound intensity and distance, and predicts the location of the flying animals, and adjusts the drone's execution commands accordingly.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method for monitoring the environmental impact of a wind farm as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an information transmission implementation program, which, when executed by a processor, implements the steps of the wind farm establishment environmental impact monitoring method as described in any one of claims 1 to 7.

Images