Transformer substation intelligent bird repelling method, system and device, and electronic equipment

Abstract

The application relates to the field of substation safety protection, and provides a substation intelligent bird repelling method, a system, a device and electronic equipment, the method comprising the following steps: acquiring bird activity data collected by a bird repelling device of a substation, video monitoring data collected by an intelligent patrol system and environment monitoring data collected by an environment monitoring system; obtaining a bird motion state vector after time-space alignment of the bird activity data and the video monitoring data; performing bird behavior prediction based on the motion state vector; obtaining an environment state vector based on the environment monitoring data, and correcting output power of the bird repelling device in combination with the environment state vector; and controlling the bird repelling device to repel birds when a bird behavior prediction result meets a preset early warning standard. The application solves the defects of a bird repelling system response lag and hidden dangers in related technologies, integrates data of the bird repelling system, the intelligent patrol system and the environment monitoring system, and improves the bird repelling effect.

Description

Technical Field

[0001] This invention relates to the field of substation safety protection technology, and in particular to an intelligent bird deterrence method, system, device, and electronic equipment for substations. Background Technology

[0002] Bird activity has become the second leading external disaster factor for substations, after lightning strikes.

[0003] While some bird control methods and systems exist in related technologies, existing solutions still suffer from response lag. From the discovery of birds during patrols to the activation of bird control devices, manual intervention is required, resulting in delays of up to minutes. This fails to meet the protection requirements against instantaneous bird damage. Furthermore, most solutions in related technologies do not consider real-time environmental conditions and equipment operating status, leading to safety hazards and further impacting the effectiveness of bird control. Summary of the Invention

[0004] This invention provides a method, system, device, and electronic equipment for intelligent bird control in substations, which addresses the shortcomings of existing bird control systems, such as delayed response and safety hazards. The solution in this application integrates data from bird control systems, intelligent patrol systems, and environmental monitoring systems, thereby improving the effectiveness of bird control.

[0005] This invention provides an intelligent bird deterrence method for substations, comprising: Acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems; The bird's motion state vector is obtained by spatiotemporally aligning the bird activity data and the video surveillance data. Bird behavior prediction is performed based on the aforementioned motion state vector; An environmental state vector is obtained based on the environmental monitoring data, and the output power of the bird deterrent device is corrected based on the environmental state vector. When the bird behavior prediction results meet the preset warning standards, the bird deterrent device is controlled to deter birds.

[0006] According to the intelligent bird control method for substations provided by the present invention, the step of predicting bird behavior based on the motion state vector includes: A uniformly accelerated motion model is used to predict the state of the motion state vector, and an observation matrix is ​​constructed. The observation matrix and the predicted trajectory are matched using Mahalanobis distance; The matching results are weighted and fused using extended Kalman filtering to obtain bird behavior prediction results.

[0007] According to the intelligent bird deterrence method for substations provided by the present invention, the step of correcting the output power of the bird deterrence device by combining the environmental state vector includes: Construct a strategy vector for the output power of the bird deterrent device; A mapping function between the environment state vector and the policy vector is established using a reinforcement learning algorithm; Based on the environmental state vector and the mapping function, the strategy vector for the output power of the bird deterrent device is modified.

[0008] According to the intelligent bird deterrence method for substations provided by the present invention, the step of controlling the bird deterrence device to deter birds when the bird behavior prediction result meets the preset warning standard includes: If it is determined that birds will enter the high-risk area after the set time, the bird deterrent device will be turned in advance and the laser will be preheated. When birds enter a high-risk area, the bird deterrent device is activated to use lasers to drive them away.

[0009] According to the intelligent bird deterrence method for substations provided by the present invention, the step of predicting bird behavior based on the motion state vector further includes: Send a pre-activation command to the bird deterrent device in the direction of bird movement; After receiving the pre-activation command, the bird deterrent device will turn in advance and perform laser preheating.

[0010] According to the intelligent bird deterrence method for substations provided by the present invention, the method further includes controlling the bird deterrence device to deter birds, and then further comprising: The model applied to spatiotemporal alignment is iteratively optimized based on the bird deterrence results.

[0011] This invention also provides a substation intelligent bird deterrence system, which applies a substation intelligent bird deterrence method, including: The data acquisition module is used to acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems. The spatiotemporal alignment module is used to spatiotemporally align the bird activity data and the video surveillance data to obtain the bird's motion state vector. The behavior prediction module is used to predict bird behavior based on the motion state vector; The output power correction module is used to obtain an environmental state vector based on the environmental monitoring data, and to correct the output power of the bird deterrent device in combination with the environmental state vector. The bird deterrence module is used to control the bird deterrence device to deter birds when the bird behavior prediction results meet the preset warning standards.

[0012] The present invention also provides an intelligent bird deterrent device for substations, comprising: The integrated management platform is used to implement intelligent bird control methods in substations; Bird deterrence device, used to deter birds under the control of the linkage management platform; The intelligent patrol system is used to automatically collect video surveillance data of birds in the substation and send it to the linkage management platform. The environmental monitoring system is used to collect environmental parameters and operating parameters of bird deterrent devices within the substation and send them to the linkage management platform.

[0013] According to the intelligent bird deterrent device for substations provided by the present invention, the number of bird deterrent devices is several; The linkage management platform is also used to send a pre-activation command to the adjacent bird deterrent device when birds move toward it.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement any of the above-described intelligent bird deterrence methods for substations.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described intelligent bird-repelling methods for substations.

[0016] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described intelligent bird deterrence methods for substations.

[0017] The intelligent bird deterrence method for substations provided by this invention can simultaneously acquire bird activity data collected by bird deterrence devices, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems. By integrating data from multiple systems, multi-system linkage can be achieved. Specifically, bird motion state vectors can be obtained by integrating video surveillance data and bird activity data. Combining bird activity data can provide a reference for video surveillance data, thereby improving the accuracy of bird behavior prediction. In addition, environmental monitoring data can also correct the output power of bird deterrence devices to avoid environmental factors affecting the normal operation of bird deterrence devices. This allows for dynamic optimization of bird deterrence strategies, improving bird deterrence effectiveness and efficiency. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the intelligent bird deterrence method for substations provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of the intelligent bird deterrence system for substations provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the intelligent bird deterrent device for substations provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the physical structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] Figure 1 This is a flowchart illustrating the intelligent bird-repelling method for substations provided in an embodiment of the present invention.

[0022] like Figure 1 As shown, this embodiment provides an intelligent bird deterrence method for substations, including: Step 101: Obtain bird activity data collected by the bird deterrent device of the substation, video surveillance data collected by the intelligent patrol system, and environmental monitoring data collected by the environmental monitoring system. In practical applications, video streams from bird deterrent devices can be obtained via the GB28181 protocol; structured data stored in the bird deterrent devices, such as bird detection results, deterrent action records, and device operating status, can be obtained via the MQTT protocol.

[0023] The video stream from the inspection camera can be obtained via the RTSP protocol; the equipment inspection results can be obtained via the IEC 61850 MMS protocol.

[0024] Sensor data collected for environmental monitoring, such as oil temperature, humidity, SF6 concentration, and illuminance, can be obtained via the Modbus-TCP protocol; rapid change signals such as equipment switch status can be obtained via the IEC 61850 GOOSE protocol.

[0025] Step 102: After spatiotemporally aligning the bird activity data and the video surveillance data, the bird's motion state vector is obtained; In practice, the constructed motion state vector can conform to the following formula X = [x, y, z, vx, vy, vz, ax, ay, az]ᵀ, where x, y, and z are the three-dimensional positions of the identified birds, vx, vy, and vz are the velocities of the identified birds, and ax, ay, and az are the acceleration information of the identified birds.

[0026] Step 103: Predict bird behavior based on the motion state vector; In practical applications, a uniformly accelerated motion model can be used to predict the state of the motion state vector and construct an observation matrix; The observation matrix and the predicted trajectory are matched using Mahalanobis distance; The matching results are weighted and fused using extended Kalman filtering to obtain bird behavior prediction results.

[0027] In this step, the bird target detection rate can be improved from 76%~82% with a single sensor to over 98.5% by using extended Kalman filtering, and the three-dimensional positioning accuracy can be improved from ±2.5m to ±0.3m.

[0028] Bird behavior prediction can be performed based on a temporal convolutional network (TCN). The temporal convolutional network used in this embodiment can contain 4 residual modules, with dilation factors d=1, 2, 4, 8, convolutional kernel k=3, and number of channels C=64.

[0029] The bird behavior prediction results output in this embodiment may include hovering, circling, approaching, flocking, and nesting. The specific behavioral patterns and threat levels are shown in the table below: Table 1. Schematic diagram of bird behavior patterns

[0030] Step 104: Obtain an environmental state vector based on the environmental monitoring data, and adjust the output power of the bird deterrent device in conjunction with the environmental state vector; In practical applications, a strategy vector for the output power of a bird deterrent device can be constructed. A mapping function between the environment state vector and the policy vector is established using a reinforcement learning algorithm; Based on the environmental state vector and the mapping function, the strategy vector for the output power of the bird deterrent device is modified.

[0031] The environmental state vector can include parameters such as illuminance, visibility, temperature, humidity, and wind speed, while the strategy vector for the output power of the bird deterrent device can include laser power, scanning frequency, pulse duty cycle, working duration, and sound wave mode.

[0032] During implementation, the PPO reinforcement learning algorithm can be used to optimize the mapping function online, ultimately increasing the expulsion success rate by 18.5 percentage points, reducing power consumption by 37.3%, and extending the bird adaptation period from 7 days to more than 30 days.

[0033] In practical applications, the protection level of the substation can be dynamically adjusted based on the equipment operating status collected by the environmental monitoring system to avoid safety accidents caused by excessive output power. For example, please refer to the table below: Table 2 Safety Constraint Mechanism Table

[0034] The execution priority of the above safety constraints is higher than that of all linkage commands. That is to say, the linkage process will be interrupted immediately and a protective action will be performed when any constraint is triggered. In addition, the system can automatically adjust its working mode based on environmental data collected by the environmental monitoring system to improve bird deterrence efficiency. For example, it can switch to daytime mode when the light intensity is >10000 lux, switch to cloudy mode when the light intensity is 500 < 10000 lux, switch to dusk mode when the light intensity is 50 < 500 lux, and switch to nighttime mode when the light intensity is ≤50 lux. It can also increase power and perform sound wave compensation in windy conditions, and reduce power and enhance sound waves in dense fog conditions.

[0035] Step 105: When the bird behavior prediction results meet the preset warning standards, control the bird deterrent device to deter birds.

[0036] In practical applications, if it is determined that birds will enter a high-risk area after a set time, the bird deterrent device will be turned in advance and the laser will be preheated. When birds enter a high-risk area, the bird deterrent device is activated to use lasers to drive them away.

[0037] In practice, after predicting bird behavior based on motion state vectors, the direction of bird movement can also be determined. If it is determined that the bird is moving towards an adjacent bird deterrent device, a pre-activation command can be sent to the bird deterrent device in the direction of the bird's movement. After receiving the pre-activation command, the adjacent bird deterrent device will turn around and perform laser preheating in advance, thus forming a seamless relay defense line.

[0038] In an exemplary embodiment, after bird deterrence is completed, the model applied to the spatiotemporal alignment can be iteratively optimized based on the bird deterrence results.

[0039] The intelligent bird deterrence method for substations provided in this embodiment can simultaneously acquire bird activity data collected by bird deterrence devices, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems. By integrating data from multiple systems, multi-system linkage can be achieved. Specifically, the bird motion state vector can be obtained by integrating video surveillance data and bird activity data. Combining bird activity data can provide a reference for video surveillance data, thereby improving the accuracy of bird behavior prediction. In addition, environmental monitoring data can also correct the output power of bird deterrence devices to avoid environmental factors affecting the normal operation of bird deterrence devices. This allows for dynamic optimization of bird deterrence strategies, improving bird deterrence effectiveness and efficiency.

[0040] The intelligent bird deterrence system for substations provided by the present invention is described below. The intelligent bird deterrence system for substations described below can be referred to in correspondence with the intelligent bird deterrence method for substations described above.

[0041] like Figure 2 As shown, the intelligent bird control system for substations provided in this embodiment includes: The data acquisition module 201 is used to acquire bird activity data collected by the bird deterrent device of the substation, video surveillance data collected by the intelligent patrol system, and environmental monitoring data collected by the environmental monitoring system. Spatiotemporal alignment module 202 is used to spatiotemporally align the bird activity data and the video surveillance data to obtain the bird's motion state vector; Behavior prediction module 203 is used to predict bird behavior based on the motion state vector; The output power correction module 204 is used to obtain an environmental state vector based on the environmental monitoring data, and to correct the output power of the bird deterrent device in combination with the environmental state vector. The bird deterrence module 205 is used to control the bird deterrence device to deter birds when the bird behavior prediction results meet the preset warning standards.

[0042] The specific implementation method of the intelligent bird deterrence system for substations provided in this embodiment can be implemented with reference to the above embodiments, and will not be repeated here.

[0043] The intelligent bird deterrent device for substations provided in this application is described below, such as... Figure 3 As shown, it includes: External Security Interaction Layer (Top Layer): This layer forms the sole security boundary between the system and the external network. A TLS / SSL encrypted tunnel is established between the bastion host / firewall and the provincial federated learning server, performing two functions: Uplink, the federated learning client uploads the local model gradient parameters from each site to the provincial federated learning server via the bastion host after encryption, to participate in cross-site aggregation; downlink, the aggregated model issued by the provincial federated learning server is transmitted back to the federated learning client through the encrypted tunnel to complete the model update.

[0044] Core Layer of the Linkage Control Platform (Third Layer, enclosed in a dashed box): This layer is the decision-making center of the system, containing 7 sub-modules interconnected via a PCIe 4.0 x16 high-speed internal bus. The data access module receives real-time video streams and image data collected by the patrol front-end equipment through the process layer TSN switch, completing GB28181 / MQTT / IEC 61850 protocol access. The EKF fusion module → TCN prediction module → PPO decision module constitute a three-level pipeline of perception → prediction → decision, performing 9-dimensional state vector extended Kalman fusion on multi-source data to output bird 3D trajectory and behavior prediction. The data fusion / PPO decision module receives the decision output, completes the encapsulation of the bird-repelling command protocol, and sends it to the bird-repelling device through the TSN switch. The safety constraint verification module is independent of the main data stream and acts as a bypass arbitrator to perform safety reviews of each PPO decision. The federated learning client periodically interacts with the provincial federated learning server to synchronize model parameters.

[0045] Process layer switching network (Layer 2): Centered on a TSN switch, it connects upwards to data access modules and data fusion / PPO decision modules, and downwards to patrol front-end equipment and bird deterrence devices. The communication protocol adopts IEEE 802.1Qbv time-aware scheduling + IEEE 802.1AS clock synchronization, achieving six-level traffic classification, with end-to-end latency of linkage commands ≤5ms and GOOSE message latency ≤10ms.

[0046] The on-site perception and execution layer (bottom layer): The left-side patrol front-end equipment includes an intelligent patrol system and an environmental monitoring system, responsible for collecting environmental and bird activity information; the right-side bird deterrent devices (including AI laser bird deterrents, ultrasonic bird deterrents, and voice bird deterrents) are responsible for deterring birds. The laser power supply control terminal is marked with a dashed sub-module and receives direct physical control from the safety constraint verification module of the core layer of the linkage control platform.

[0047] In the above-mentioned intelligent bird deterrence devices for substations, the inspection front-end equipment and the data access module communicate bidirectionally via the TSN protocol with a latency of ≤10ms; the bird deterrence device and the data fusion / PPO decision module communicate bidirectionally via the TSN protocol with a latency of ≤10ms; the data access module and the EKF fusion module communicate unidirectionally; the EKF fusion module and the TCN prediction module communicate unidirectionally; the TCN prediction module and the PPO decision module communicate unidirectionally; the PPO decision module and the data fusion / PPO decision module communicate unidirectionally; the safety constraint verification module and the laser power supply control terminal communicate unidirectionally; the federated learning client and the bastion host / firewall communicate bidirectionally via a TLS or SSL encrypted tunnel; and the bastion host / firewall and the federated learning server communicate bidirectionally via a TLS or SSL encrypted tunnel.

[0048] In addition, such as Figure 3 As shown, the safety constraint verification module and the bird deterrent device are connected by a physical electrical hardwire. This connection bypasses the process layer TSN switching network and PPO decision module, allowing the safety constraint verification module to directly control the laser power supply of the bird deterrent device. When safety events such as intrusion into the equipment protection zone, abnormal SF6 density, or partial discharge alarms are detected, the safety constraint verification module immediately cuts off the laser power supply via a hardwire signal, with a response delay of ≤1ms. This ensures intrinsic safety without relying on any intermediate links or communication protocols.

[0049] The following are some specific application scenarios to illustrate the solution in this application.

[0050] In the first application scenario, a bird deterrence process was carried out, from patrol discovery and prediction to precise bird removal, and the time process is shown below; 14:32:15.000: The patrol system camera CAM-07 detected a magpie, and the YOLOv8n-Bird output a confidence level of 0.92; 14:32:15.050: The patrol system sends a bird intrusion alarm to the linkage management platform via MQTT; 14:32:15.060: The EKF fusion engine associates and matches pixel coordinates with LiDAR point clouds, outputting 3D coordinates [15.2, 8.7, 4.3] m, with a speed of [-1.2, -0.8, -0.3] m / s; 14:32:15.075: TCN behavior prediction determines "approaching" mode (probability 0.78), predicting that it will enter the area above the high voltage bushing of the No. 1 main transformer (A+ level area) in 3 seconds. 14:32:15.090: The linkage control engine queries the protection level table, confirms the bird deterrent device BD-03, and generates L2 level linkage command (pan-tilt preset position [127°, 35°], laser 450mW / 2Hz / zigzag scan, sonic mode #3); 14:32:15.100: The instruction was issued via the TSN network (priority 7), and the transmission time was 4.2ms; 14:32:15.200: BD-03 completes gimbal pre-steering and laser preheating; 14:32:17.850: EKF confirms birds have entered the BD-03 coverage area and sends a command to initiate a removal operation; 14:32:17.920: BD-03 initiates laser scanning and sonic deterrence; 14:32:20.150: The magpie was driven away, and BD-03 successfully reported the result (2.25 seconds).

[0051] The entire process took 5.2 seconds, with the decision-making process taking less than 100ms, network transmission taking less than 5ms, and the removal execution taking 2.25 seconds. Compared to the manual response mode (average 5 minutes), this represents a 57-fold increase in efficiency.

[0052] In the second application scenario, an example is shown where the surrounding environment can cause changes in bird deterrence strategies, and the timeline is as follows: 17:15:00: Illuminance dropped sharply from 8500 lux to 320 lux (dusk), and wind speed increased from 3.2 m / s to 7.8 m / s; 17:15:02: The PPO decision engine outputs the optimal strategy: 200mW laser, 1.5Hz scanning, and 50% duty cycle; 17:15:04: Safety constraint verification passed (power limit of 250mW in Dusk Mode), all 23 devices in the station switched to "Dusk Mode"; 17:15:05: The auxiliary monitoring system automatically turns on the area lighting and adjusts the gain parameters of the patrol cameras.

[0053] The third application scenario illustrates the process of sending pre-activation commands to adjacent bird-repelling devices to achieve cross-regional relay deterrence. The timeline is as follows: 06:45:30: BD-01 detected a flock of sparrows (8 birds, in cluster mode), speed [2.1, -1.5, 0.2] m / s; 06:45:31: TCN predicts entering the BD-02 area in 2.8 seconds (v_r=2.1>1.5, d=22m<30m, heading deviation 18°<45°); 06:45:33: The linkage platform sends a pre-activation command to BD-02: gimbal pre-rotation [85°, 28°], laser preheating, and acoustic preloading; 06:45:35: Bird deterrent device BD-01 is activated to drive away sparrows (spiral scan), and the flock of sparrows moves towards BD-02; 06:45:35.6: The bird deterrent device BD-02 initiates laser scanning (zigzag scanning) within 60ms, forming a seamless relay; 06:45:38: The sparrows were completely driven away.

[0054] The fourth application scenario illustrates the process of security protection when a device malfunctions, and its timeline is as follows: 10:20:00: The oil temperature of the No. 2 main transformer rose from 68℃ to 76℃ (exceeding the 75℃ alarm threshold); 10:20:03: Safety constraint module execution: Oil temperature > 65℃, BD-04 / BD-05 derating to 70% power (350mW upper limit); 10:20:05: Automatically generate maintenance work orders and push them to maintenance personnel; 10:25:00: If the oil temperature continues to rise to 85℃, completely stop the laser and switch to pure acoustic mode.

[0055] The fifth application scenario illustrates the process of establishing a coordinated defensive line when large-scale bird migration is detected, and its timeline is as follows: 06:00: 187 bird activity events were recorded at the station in one hour (3.02 times the daily average σ), indicating the station has entered "migration mode"; 06:00:05: The 23 devices are organized into three layers of defense: the forward early warning layer (BD-01~08, 200mW), the core deterrence layer (BD-09~16, 450~500mW), and the rear interception layer (BD-17~23, 300mW); the spacing between the layers is 15~20m, the depth is about 60m, and the bird tracks are shared in real time between the layers through TSN; The effectiveness of the defense line is assessed every 10 minutes, and the strategy is dynamically adjusted. The migration mode is automatically exited when the activity frequency drops below 2σ.

[0056] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a substation intelligent bird deterrence method, which includes: Acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems; The bird's motion state vector is obtained by spatiotemporally aligning the bird activity data and the video surveillance data. Bird behavior prediction is performed based on the aforementioned motion state vector; An environmental state vector is obtained based on the environmental monitoring data, and the output power of the bird deterrent device is corrected based on the environmental state vector. When the bird behavior prediction results meet the preset warning standards, the bird deterrent device is controlled to deter birds.

[0057] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0058] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the substation intelligent bird deterrence method provided by the above methods, the method including: Acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems; The bird's motion state vector is obtained by spatiotemporally aligning the bird activity data and the video surveillance data. Bird behavior prediction is performed based on the aforementioned motion state vector; An environmental state vector is obtained based on the environmental monitoring data, and the output power of the bird deterrent device is corrected based on the environmental state vector. When the bird behavior prediction results meet the preset warning standards, the bird deterrent device is controlled to deter birds.

[0059] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the substation intelligent bird deterrence method provided by the methods described above, the method comprising: Acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems; The bird's motion state vector is obtained by spatiotemporally aligning the bird activity data and the video surveillance data. Bird behavior prediction is performed based on the aforementioned motion state vector; An environmental state vector is obtained based on the environmental monitoring data, and the output power of the bird deterrent device is corrected based on the environmental state vector. When the bird behavior prediction results meet the preset warning standards, the bird deterrent device is controlled to deter birds.

[0060] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0061] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0062] 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 of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A smart bird deterrence method for substations, characterized in that, include: Acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems; The bird's motion state vector is obtained by spatiotemporally aligning the bird activity data and the video surveillance data. Bird behavior prediction is performed based on the aforementioned motion state vector; An environmental state vector is obtained based on the environmental monitoring data, and the output power of the bird deterrent device is corrected based on the environmental state vector. When the bird behavior prediction results meet the preset warning standards, the bird deterrent device is controlled to deter birds.

2. The intelligent bird-repelling method for substations according to claim 1, characterized in that, The bird behavior prediction based on the motion state vector includes: A uniformly accelerated motion model is used to predict the state of the motion state vector, and an observation matrix is ​​constructed. The observation matrix and the predicted trajectory are matched using Mahalanobis distance; The matching results are weighted and fused using extended Kalman filtering to obtain bird behavior prediction results.

3. The intelligent bird-repelling method for substations according to claim 1, characterized in that, The step of correcting the output power of the bird deterrent device by incorporating the environmental state vector includes: Construct a strategy vector for the output power of the bird deterrent device; A mapping function between the environment state vector and the policy vector is established using a reinforcement learning algorithm; Based on the environmental state vector and the mapping function, the strategy vector for the output power of the bird deterrent device is modified.

4. The intelligent bird-repelling method for substations according to claim 1, characterized in that, When the bird behavior prediction results meet the preset warning criteria, controlling the bird deterrent device to deter birds includes: If it is determined that birds will enter the high-risk area after the set time, the bird deterrent device will be turned in advance and the laser will be preheated. When birds enter a high-risk area, the bird deterrent device is activated to use lasers to drive them away.

5. The intelligent bird-repelling method for substations according to claim 1, characterized in that, The process of predicting bird behavior based on the motion state vector further includes: Send a pre-activation command to the bird deterrent device in the direction of bird movement; After receiving the pre-activation command, the bird deterrent device will turn in advance and perform laser preheating.

6. The intelligent bird-repelling method for substations according to claim 1, characterized in that, The bird-repelling device is used to repel birds, and then the following is also included: The model applied to spatiotemporal alignment is iteratively optimized based on the bird deterrence results.

7. A substation intelligent bird deterrence system, employing the substation intelligent bird deterrence method according to any one of claims 1-6, characterized in that, include: The data acquisition module is used to acquire bird activity data collected by bird deterrent devices in substations, video surveillance data collected by intelligent patrol systems, and environmental monitoring data collected by environmental monitoring systems. The spatiotemporal alignment module is used to spatiotemporally align the bird activity data and the video surveillance data to obtain the bird's motion state vector. The behavior prediction module is used to predict bird behavior based on the motion state vector; The output power correction module is used to obtain an environmental state vector based on the environmental monitoring data, and to correct the output power of the bird deterrent device in combination with the environmental state vector. The bird deterrence module is used to control the bird deterrence device to deter birds when the bird behavior prediction results meet the preset warning standards.

8. A smart bird deterrent device for substations, characterized in that, include: A linkage management platform is used to execute the intelligent bird deterrence method for substations as described in any one of claims 1-6; Bird deterrence device, used to deter birds under the control of the linkage management platform; The intelligent patrol system is used to automatically collect video surveillance data of birds in the substation and send it to the linkage management platform. The environmental monitoring system is used to collect environmental parameters and operating parameters of bird deterrent devices within the substation and send them to the linkage management platform.

9. The intelligent bird deterrent device for substations according to claim 8, characterized in that, The number of bird-repelling devices is several; The linkage management platform is also used to send a pre-activation command to the adjacent bird deterrent device when birds move toward it.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the intelligent bird-repelling method for substations as described in any one of claims 1-6.

Images