A power transmission network bird hazard risk management method and system

Abstract

The application provides a power transmission network bird hazard risk management and control method and system, the system comprises a regional management module, a field monitoring module, a bird hazard analysis module and a regional control module; the regional management module is used for monitoring area division of a power transmission network coverage area based on a preset three-dimensional GIS map model, obtaining division information of each monitoring area; the field monitoring module is used for receiving field monitoring information transmitted by a field monitoring device, and integrating the obtained field monitoring information into the corresponding monitoring area; the bird hazard analysis module is used for identifying bird targets existing in the monitoring area according to the obtained field monitoring information, and further extracting the motion trajectory of the bird targets, obtaining bird monitoring results; the regional control module is used for sending driving instructions to the field bird repelling device of the corresponding monitoring area according to the obtained bird monitoring results. The application helps to improve the initiative and intelligent level of power transmission network bird hazard risk management and control, and improve the safety of daily operation and control of the power transmission network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transmission network bird hazard risk management and control, and particularly relates to a power transmission network bird hazard risk management and control method and system. BACKGROUND

[0002] With the continuous expansion of the power grid scale in China, the length of power transmission lines increases year by year, and the conflict between power transmission lines and wildlife habitats is becoming increasingly serious. Birds, as an important biological resource in nature, have important significance for ecological balance and human social and economic development. However, due to the construction and operation of power transmission lines, the bird habitats and activity ranges are seriously affected, which leads to an increase in conflicts between birds and power transmission lines and causes a series of safety problems.

[0003] At present, for the prevention and maintenance of bird hazards of power transmission lines, an independent processing method is usually used, that is, a maintenance personnel or a UAV is used to patrol along the power transmission line, and when a bird or a nest is found on the power transmission line, the bird hazard target is processed separately. However, the traditional bird hazard prevention and maintenance method has the problems of insufficient real-time performance and poor processing of the bird hazard target, which easily forms a "guerrilla war" situation and affects the reliability of the bird hazard risk management and control of the power transmission line. SUMMARY

[0004] In view of the above problems, the present application aims to provide a power transmission network bird hazard risk management and control method and system.

[0005] The object of the present application is achieved by using the following technical scheme:

[0006] In a first aspect, the present application shows a power transmission network bird hazard risk management and control system, comprising a regional management module, a field monitoring module, a bird hazard analysis module and a regional control module; wherein,

[0007] The regional management module is used to divide the monitoring area of the power transmission network coverage area according to the position distribution information of the key equipment of the power transmission network based on the preset three-dimensional GIS map model, and obtain the division information of each monitoring area;

[0008] The field monitoring module is used to receive the field monitoring information transmitted by the field monitoring device, and integrate the obtained field monitoring information into the corresponding monitoring area, wherein the field monitoring information includes video monitoring data;

[0009] The bird hazard analysis module is used to identify the bird target existing in the monitoring area according to the obtained field monitoring information, and further extract the motion trajectory of the bird target to obtain the bird monitoring result;

[0010] The regional management module is configured to send driving instructions to the on-site bird repelling device in the corresponding monitoring area according to the acquired bird monitoring result, so that the on-site bird repelling device is started and bird repelling is completed.

[0011] Preferably, the regional management module comprises a GIS map unit, a power grid information unit and a regional division unit; wherein,

[0012] The GIS map unit is configured to input map data corresponding to actual geographical information of the power grid region to obtain a three-dimensional GIS map model.

[0013] The power grid information unit is configured to input related power equipment data and on-site monitoring equipment data according to the setting of the actual power equipment in the power grid region, and mark the related power equipment and on-site monitoring equipment in the three-dimensional GIS map model; wherein the power equipment includes power towers, power lines, insulators, transformer boxes and the like, and the power equipment data includes the specific setting position and / or distribution position of the power equipment, and the on-site monitoring equipment includes a camera.

[0014] The regional division unit is configured to divide the power grid coverage region into monitoring areas according to the position distribution information of the key power equipment, to obtain division information of each monitoring area.

[0015] Preferably, the regional division unit further comprises:

[0016] According to the position distribution information of the key power equipment, the power grid coverage region is adaptively divided to obtain specific division information of each monitoring area, including:

[0017] In the initialization stage, the power grid coverage region is preliminarily divided into N adjacent sub-regions according to a preset division number;

[0018] In the adaptive evolution stage, the bird damage control weight factor of each sub-region is calculated respectively, wherein the bird damage control weight factor calculation function used is:

[0019] Y(i) = ω1 × Pt div (i) + ω2 × Pt Br (i)

[0020] Wherein, Y(i) represents the bird damage control weight factor of the i-th sub-region, Pt div (i) represents the power equipment weight factor, Pt Br (i) represents the bird damage degree weight factor, and ω1 and ω2 represent preset weights.

[0021] The size of each sub-region is adjusted according to the bird damage control weight factor of each sub-region, wherein when the bird damage control weight factor is greater than the average bird damage control weight factor of each sub-region, the area of the sub-region is reduced; otherwise, when the bird damage control weight factor is less than the average bird damage control weight factor of each sub-region, the area of the sub-region is expanded.

[0022] The process of the adaptive evolution stage is repeated, and the adaptive evolution stage is ended when the division of each sub-region tends to be stable or reaches the maximum number of repetitions.

[0023] According to the division of each sub-region after adaptive evolution, each sub-region is taken as a monitoring region to complete the division of the monitoring region and obtain the division information of each monitoring region.

[0024] Preferably, the on-site monitoring module comprises a receiving unit and an association unit; wherein,

[0025] The receiving unit is configured to establish a communication connection with the on-site monitoring device arranged in each monitoring region, and to receive the on-site monitoring information returned by the on-site monitoring device in real time;

[0026] The association unit is configured to associate the obtained on-site monitoring information with the three-dimensional GIS map model, associate the obtained on-site monitoring information with the corresponding monitoring region, and integrate the on-site monitoring information into the three-dimensional GIS map model.

[0027] Preferably, the bird damage analysis module comprises a video analysis unit; wherein,

[0028] The video analysis unit is configured to perform bird target identification according to the obtained video monitoring data to obtain a bird target identification result; and further perform trajectory analysis on the bird target according to the video monitoring data after identifying the bird target, to predict the flight trajectory of the bird target and obtain a bird target flight trajectory prediction result;

[0029] The video analysis unit is further configured to, according to the bird target identification result, when the bird target is identified, obtain the associated monitoring region as a control region according to the source video monitoring data; and further obtain the monitoring region in the direction corresponding to the control region as an early warning region according to the bird target flight trajectory prediction result.

[0030] Preferably, the video analysis unit comprises an extraction unit, a preprocessing unit, an identification unit, a trajectory unit and a result output unit; wherein,

[0031] The extraction unit is configured to extract the video monitoring data to be processed;

[0032] The preprocessing unit is configured to pre-process the video monitoring data to be processed, including picture extraction and picture enhancement processing, to obtain pre-processed video monitoring data;

[0033] The recognition unit is configured to perform image recognition processing on the obtained video monitoring data, recognize the bird target in the video monitoring picture, and obtain a bird target recognition result;

[0034] The trajectory unit is configured to track the movement trajectory of the bird target according to the continuous video monitoring pictures after recognizing the bird target, further predict the movement route according to the movement trajectory, and obtain a bird target flight trajectory prediction result;

[0035] The result output unit is configured to, according to the bird target recognition result, when the bird target is recognized, acquire the associated monitoring area as a management and control area according to the source video monitoring data; and further according to the bird target flight trajectory prediction result, acquire the monitoring area in the corresponding direction of the management and control area as a warning area.

[0036] Preferably, the area management module comprises a control unit and a warning unit.

[0037] The control unit is configured to, according to the obtained management and control area information, send a driving instruction to the on-site bird repelling device of the management and control area, so that the on-site bird repelling device is started and bird chasing is completed.

[0038] The warning unit is configured to, according to the obtained warning area information, send a driving instruction to the on-site bird repelling device of the warning area, so that the on-site bird repelling device is started and bird directional prevention is completed.

[0039] Preferably, the system further comprises a log module.

[0040] The log module is configured to generate a corresponding bird damage supervision log according to the bird monitoring result obtained by each monitoring area and the corresponding driving instruction.

[0041] Preferably, the system further comprises an information database module.

[0042] The information database module is configured to extract bird characteristic information according to the obtained bird monitoring result, and construct a power grid bird damage information database, wherein the power grid bird damage information database records the bird species, the bird quantity, the appearing time period, the frequency, the time characteristic and the like data appearing in the power grid area.

[0043] In a second aspect, the present application shows a power grid bird damage risk management and control method of any one of the above-mentioned first aspect, comprising:

[0044] The area management module divides the monitoring area of the power grid coverage area according to the position distribution information of the key equipment of the power grid based on the preset three-dimensional GIS map model, and obtains the division information of each monitoring area.

[0045] The field monitoring module receives field monitoring information transmitted by the field monitoring device, and integrates the obtained field monitoring information into a corresponding monitoring area, wherein the field monitoring information includes video monitoring data;

[0046] The bird damage analysis module identifies bird targets existing in the monitoring area according to the obtained field monitoring information, and further extracts the motion trajectory of the bird targets to obtain a bird monitoring result.

[0047] The regional control module sends a driving instruction to the field bird repelling device of the corresponding monitoring area according to the obtained bird monitoring result, so that the field bird repelling device is started and bird chasing is completed.

[0048] The beneficial effects of the present application are: a power grid bird damage risk control method and system are proposed, first, based on a preset three-dimensional GIS map model, the area where the power grid is located is divided into regions to obtain corresponding monitoring area division information; real-time receiving of field monitoring information returned by each region, and bird target identification and bird target motion trajectory prediction according to the field monitoring information, obtaining a bird monitoring result, according to the bird monitoring result, further controlling the bird repelling device of the corresponding region, so as to realize bird chasing work in the control area, which helps to improve the real-time level of power grid bird damage risk detection, and helps to improve the initiative and intelligent level of bird damage risk control, and improves the safety of daily operation and control of the power grid. BRIEF DESCRIPTION OF DRAWINGS

[0049] The present application is further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present application. For ordinary skilled in the art, other drawings can be obtained without creative labor according to the following drawings.

[0050] Figure 1 The framework structure diagram of a power grid bird damage risk control system shown in an embodiment of the present application;

[0051] Figure 2 For Figure 1 The module setting schematic diagram of a power grid bird damage risk control system shown in an embodiment of the present application;

[0052] Figure 3 The flowchart of a power grid bird damage risk control method shown in an embodiment of the present application. DETAILED DESCRIPTION

[0053] The present application is further described in combination with the following application scenarios.

[0054] Referring to Figure 1 , which shows a power grid bird damage risk control system, including a regional management module, a field monitoring module, a bird damage analysis module and a regional control module; wherein,

[0055] The area management module is used to divide the monitoring area of ​​the power transmission network coverage area based on a preset 3D GIS map model and the location distribution information of key equipment in the power transmission network, and obtain the division information of each monitoring area.

[0056] The on-site monitoring module is used to receive on-site monitoring information transmitted by on-site monitoring equipment and integrate the acquired on-site monitoring information into the corresponding monitoring area, wherein the on-site monitoring information includes video monitoring data;

[0057] The bird damage analysis module is used to identify bird targets in the monitoring area based on the acquired on-site monitoring information, and further extract the movement trajectory of the bird targets to obtain bird monitoring results;

[0058] The area control module is used to send drive commands to the on-site bird deterrent devices in the corresponding monitoring area based on the obtained bird monitoring results, so that the on-site bird deterrent devices can be activated and complete the bird driving away.

[0059] The above-described embodiments of the present invention propose a power transmission network bird damage risk management system. First, the area where the power transmission network is located is divided into regions based on a preset 3D GIS map model, obtaining corresponding monitoring area division information. Real-time reception of on-site monitoring information from each region is performed, and bird target identification and trajectory prediction are conducted based on the on-site monitoring information to obtain bird monitoring results. Based on the bird monitoring results, bird deterrence devices in the corresponding areas are further controlled, thereby achieving bird deterrence within the managed area. This helps improve the real-time level of bird damage risk detection in the power transmission network, while also enhancing the initiative and intelligence level of bird damage risk management, and improving the safety of daily operation and management of the power transmission network.

[0060] Preferred, see Figure 2 The regional management module includes GIS map units, power transmission network information units, and regional division units; among them,

[0061] The GIS map unit is used to input map data corresponding to the actual geographic information of the power transmission network area to obtain a three-dimensional GIS map model.

[0062] The power transmission network information unit is used to input relevant power transmission equipment data and field monitoring equipment data according to the actual setting of power transmission equipment in the power transmission network area, and to mark the relevant power transmission equipment and field monitoring equipment in the three-dimensional GIS map model; wherein the power transmission equipment includes power transmission towers, power transmission lines, insulators, transformer boxes, etc., the power transmission equipment data includes the specific setting location and / or distribution location of the power transmission equipment, and the field monitoring equipment includes cameras;

[0063] The region division unit is configured to divide the monitoring region of the power transmission network according to the location distribution information of the key equipment of the power transmission network, and obtain the division information of each monitoring region.

[0064] Based on the GIS map building technology, the three-dimensional GIS map model is built according to the actual geographic information and the actual data of the corresponding equipment distribution information of the region where the power transmission network is located, which can help to reflect the actual situation of the monitoring region with the real GIS map model, and further associate the corresponding information based on the obtained three-dimensional model, which can help to truly feedback the feature data of different dimensions in the power transmission network region through the GIS map model, and improve the data management level. At the same time, the division of the control region is further completed based on the three-dimensional GIS map model, which can help to realize accurate control region division according to the real and detailed map model data, and improve the fine and intelligent level of regional control.

[0065] According to the region division method, the bird damage control of the power transmission network is divided into regions, which can help to form targeted directional driving of the bird damage target in the control region, and also can form effective protection for the key region, and improve the effect of bird damage control of the power transmission network. By combining the region division method, the corresponding early warning sensitivity can be formulated for different regions, which can help to improve the overall planning level of bird damage protection.

[0066] Preferably, the region division unit further comprises:

[0067] According to the location distribution information of the key equipment of the power transmission network, the power transmission network coverage region is adaptively divided, and the specific division information of each monitoring region is obtained, including:

[0068] In the initialization stage, the power transmission network coverage region is preliminarily divided into N adjacent sub-regions according to the preset division number;

[0069] In the adaptive evolution stage, the bird damage control weight factor of each sub-region is calculated respectively, wherein the bird damage control weight factor calculation function adopted is:

[0070] Y(i) = ω1 × Pt div (i) + ω2 × Pt Br (i)

[0071] Wherein, Y(i) represents the bird damage control weight factor of the i-th sub-region, Pt div (i) represents the power transmission equipment weight factor, Pt Br (i) represents the bird damage degree weight factor, and ω1 and ω2 represent the preset weight.

[0072] According to the bird damage control weight factor size of each sub-region, the size of the sub-region is adjusted, wherein when the bird damage control weight factor is greater than the average bird damage control weight factor of each sub-region, the area of the sub-region is reduced; otherwise, when the bird damage control weight factor is less than the average bird damage control weight factor of each sub-region, the area of the sub-region is expanded.

[0073] The process of the adaptive evolution stage is repeated, and when the division of each sub-region tends to be stable or reaches the maximum number of repetitions, the adaptive evolution stage is ended.

[0074] According to the division of each sub-region after adaptive evolution, each sub-region is taken as a monitoring region, the division of the monitoring region is completed, and the division information of each monitoring region is obtained.

[0075] In the above embodiments, a technical scheme of adaptively dividing the control region according to the coverage region of the power transmission network is proposed, which can comprehensively consider the importance of the power transmission equipment and the severity of the bird damage and realize adaptive monitoring region division, thereby helping to improve the pertinence and adaptability of bird damage risk control.

[0076] Among them, the more critical the region in the power transmission network is, the higher the degree of refinement of the region division is, that is, the higher the degree of refinement of the bird damage monitoring effect of the critical region is. Through the region division, the bird damage control refinement degree of different regions in the power transmission network can be distinguished, thereby improving the bird damage control precision of the critical region, and for the general less important region, the monitoring precision of the region can be reduced, thereby effectively optimizing the effect of the monitoring equipment and data processing resource investment, and improving the overall effect of the regional bird damage control of the power transmission network.

[0077] Further, the region division unit further comprises:

[0078] After completing the division of each monitoring region, the region where the key equipment of the power transmission network (for example, the power transmission equipment whose weight factor is greater than a preset standard value) is located is marked as a key region, so as to further control the bird damage of the key region subsequently.

[0079] Preferably, the field monitoring module comprises a receiving unit and an association unit; wherein,

[0080] The receiving unit is configured to establish a communication connection with the field monitoring device arranged in each monitoring region, and to receive the field monitoring information returned by the field monitoring device in real time;

[0081] The association unit is configured to associate the obtained field monitoring information with the three-dimensional GIS map model, associate the obtained field monitoring information with the corresponding monitoring region, and integrate the field monitoring information into the three-dimensional GIS map model.

[0082] The system is divided into monitoring zones, with at least one on-site monitoring device installed in each zone. By establishing a connection with these devices and acquiring their data in real time, bird damage analysis can be performed on the monitored areas. The on-site monitoring zones include cameras, lidar, and other monitoring equipment. The acquired video monitoring data consists of video images collected from power grid equipment within the monitored area. These video images include views of the power grid equipment or the sky over the monitored area, allowing for the identification of bird-related targets within the monitored area.

[0083] Based on the 3D GIS map model that has completed the division of control areas, the equipment within the control areas is marked according to the division information. For field monitoring equipment, the corresponding field monitoring data is associated and marked according to the control area information where the field monitoring equipment is located, laying the foundation for subsequent bird damage identification and bird drive-away in the sub-control areas.

[0084] Preferably, the bird damage analysis module includes a video analysis unit; wherein,

[0085] The video analysis unit is used to identify bird targets based on the acquired video monitoring data and obtain bird target identification results; after identifying bird targets, it further performs trajectory analysis on bird targets based on video monitoring data to predict the flight trajectory of bird targets and obtain bird target flight trajectory prediction results.

[0086] The video analysis unit is also used to obtain the associated monitoring area as the control area based on the bird target recognition results when a bird target is identified; and further, based on the bird target flight trajectory prediction results, the monitoring area in the corresponding direction of the control area is used as the warning area.

[0087] The video analysis unit analyzes the obtained video monitoring data using video analysis technology. The bird recognition model based on image analysis can accurately identify bird targets appearing in the video monitoring data. At the same time, the identified bird targets are further tracked using trajectory tracking technology to predict their trajectories. Based on the bird recognition results and trajectory prediction results, control areas where bird damage occurs and warning areas requiring targeted bird damage warnings are determined, laying the foundation for subsequent targeted regional bird control based on the obtained control area information and warning area information.

[0088] Preferably, the video analytics module further includes:

[0089] According to the obtained management area information, when the monitoring area adjacent to the management area contains a key area, the key area adjacent to the management area is also taken as a pre-warning area.

[0090] Preferably, the video analysis unit comprises an extraction unit, a preprocessing unit, an identification unit, a trajectory unit and a result output unit; wherein,

[0091] The extraction unit is configured to extract the video monitoring data to be processed.

[0092] The preprocessing unit is configured to perform preprocessing on the video monitoring data to be processed, including picture extraction and picture enhancement processing, to obtain preprocessed video monitoring data.

[0093] The identification unit is configured to perform image recognition processing on the obtained video monitoring data, to identify the bird target in the video monitoring picture, and to obtain a bird target identification result.

[0094] The trajectory unit is configured to track the movement trajectory of the bird target according to continuous video monitoring pictures after the bird target is identified, and to further predict the movement route according to the movement trajectory, to obtain a bird target flight trajectory prediction result.

[0095] The result output unit is configured to, according to the bird target identification result, when the bird target is identified, acquire the associated monitoring area as a management area according to the source video monitoring data; and further according to the bird target flight trajectory prediction result, take the monitoring area in the corresponding direction of the management area as a pre-warning area.

[0096] In the above embodiment, when the video analysis unit performs bird damage analysis on the obtained video monitoring data, the video analysis data is first extracted for the monitoring area, and the obtained video analysis data is subjected to picture extraction and preprocessing, the video monitoring picture needing to be analyzed is extracted, and the extracted video monitoring picture is preprocessed to improve the quality of the video monitoring picture. According to the preprocessed video monitoring picture, the bird identification processing is performed based on the image recognition model to obtain the bird target recognition result. Further, according to the bird target recognition result, the recognized bird target is further tracked to obtain the motion trajectory of the bird target. And according to a period of motion trajectory as a basis, the motion trajectory of the bird target is further predicted to obtain the flight trajectory prediction result of the bird target. According to the recognition result and the flight trajectory prediction result of the bird target, the current monitoring area and the adjacent management and control area obtained based on the current monitoring area and combined with the flight trajectory prediction result of the bird target are taken as the management and control area and the warning area, and further driving measures are taken for the analyzed management and control area and warning area to solve the bird damage risk. The above bird damage target recognition and warning based on video monitoring data can quickly and accurately identify and locate the bird target in the area, which helps to improve the accuracy and real-time level of bird damage target monitoring.

[0097] The model for bird target recognition is completed by using a trained bird intelligent recognition model based on YOLOv8 deep learning, or by using other image recognition models that can also realize bird image recognition. The present application does not make specific limitations here.

[0098] Considering that the obtained video monitoring data inevitably contains a large area of sky part in the picture, when the picture contains a large amount of sky part, the image picture is easily affected by the change of strong light or dullness of the sky part, resulting in unclear situation in the region, which affects the accuracy of bird target recognition in the sky area. Therefore, based on the preprocessing unit, a technical scheme for picture enhancement processing of video monitoring data is particularly proposed to improve the clarity of the picture and the bird target recognition effect.

[0099] Preferably, the picture extraction and picture enhancement processing of the obtained video monitoring data in the preprocessing unit comprises:

[0100] The picture extraction is performed according to the obtained video monitoring data, and the video monitoring picture Pic(t) at each time is extracted according to a preset time interval, wherein Pic(t) represents the video monitoring picture at time t;

[0101] For the video monitoring image Pic(t0) at the current moment, the video monitoring image is converted to the Lab three-channel color space to obtain the brightness channel sub-image Lpic(t0), the first color channel sub-image apic(t0), and the second color channel sub-image bpic(t0) of the video monitoring image.

[0102] Based on the obtained luminance channel sub-image Lpic(t0), the image is divided into multiple sub-regions, including:

[0103] An inspection window is used to sequentially traverse various positions in the image. During the traversal, the brightness feature value of the current inspection window is calculated based on the brightness channel values ​​of each pixel within the inspection window's range. The brightness feature calculation function used is as follows:

[0104]

[0105] Where DK(x,y) represents the brightness characteristic value obtained when the center of the inspection window is aligned with the (x,y) position; σL DK MeanL represents the standard deviation of the luminance channel values ​​of each pixel within the coverage area of ​​the inspection window. DK L represents the average value of the luminance channel of each pixel within the coverage area of ​​the inspection window. bc Represents the characteristic correction value, where L bc ∈[1,4];

[0106] Based on the brightness characteristic value of the current inspection window and the preset characteristic threshold Th tL Comparison, where Th tL ∈[8,20], when the brightness feature value is less than the feature threshold DK(x,y) <Th tL If the current inspection window covers a certain number of pixels, then the pixels in that window are marked as a class of pixels; otherwise, if the brightness feature value is greater than or equal to the feature threshold DK(x,y)≥Th, then the pixel value is marked as a class of pixels. tL When the current inspection window covers the pixel, mark it as a second-class pixel. If a pixel is marked as both a first-class pixel and a second-class pixel, then record the pixel as a second-class pixel.

[0107] The region enclosed by connected pixels of one class is marked as the feature region PA, and the region enclosed by connected pixels of two classes is marked as the transition region PB.

[0108] For the obtained feature region PA, the pixels in the feature region are subjected to a first brightness adjustment process, wherein the first brightness adjustment function is:

[0109]

[0110] Among them, L′ t0(x,y) represents the luminance channel value of the first luminance adjustment pixel point (x,y), L t0 (x,y) represents the luminance channel value of the current pixel point (x,y), L cc represents the preset feature region luminance standard value, wherein L cc ∈[40,60], L tc and L gc respectively represent the preset feature region luminance adjustment value and the luminance span value, wherein L tc ∈[60,65], L gc ∈[20,25], and αc represents the preset luminance feature retention factor, wherein αc∈[8,12];

[0111] According to the luminance channel values of the pixels after the first luminance adjustment processing, a feature region after the first luminance adjustment processing is obtained, and the luminance channel subgraph Lpic'(t0) of the current video monitoring picture is updated;

[0112] According to the current luminance channel subgraph Lpic'(t0), the transition region PB is further subjected to second luminance adjustment processing, including:

[0113] The following judgment function is used to judge the pixels in the transition region:

[0114] (1)

[0115] (2) IFture2(x,y): D tA (x,y)>D AB

[0116] (3)

[0117] IFture1(x,y) represents the first judgment function, represents that the pixel point (x,y) in the video monitoring picture at the previous time t0-1 is marked as a transition region, and respectively represent the luminance channel values of the pixel point (x,y) in the video monitoring picture at the current time and the previous time, L AB represents the preset luminance change value, L AB ∈[15,30]; when D and are simultaneously satisfied, IFture1(x,y) = ω1, otherwise IFture1(x,y) = 0, wherein ω1 represents the first weight factor, ω1∈[0.9,1.1]; IFture2(x,y) represents the second judgment function, D tA (x,y) represents the distance from the pixel point (x,y) to the nearest feature region boundary, DAB represents a preset boundary change value, D AB ∈[3,5]; when D tA (x,y)>D AB IFture2(x,y)=0, wherein ω2 represents a second weight factor, ω2∈[0.9,1.1]; IFture3(x,y) represents a third judgment function, L ABmax represents a preset maximum change value, wherein L ABmax ∈[90,95], when IFture3(x,y)=0, wherein ω3 represents a third weight factor, ω3∈[0.9,1.1];

[0118] According to the judgment results of each pixel point, the pixel points in the transition region are subjected to a second brightness adjustment processing, wherein a second brightness adjustment processing function is adopted:

[0119]

[0120] wherein, represents a brightness channel value of the pixel point (x,y) after the second adjustment processing, represents a brightness channel value of the current pixel point (x,y); L cd represents a preset brightness channel adjustment value, L cd ∈[60,70];

[0121] According to the brightness channel values of each pixel point after the second brightness adjustment processing, a second brightness adjustment processing transition region is obtained, and the brightness channel subgraph Lpic''(t0) of the current video monitoring picture is updated.

[0122] According to the updated brightness channel subgraph Lpic''(t0), the first color channel subgraph apic(t0) and the second color channel subgraph bpic(t0), channel reconstruction is performed to obtain a preprocessed video monitoring picture.

[0123] The recognition unit further performs image recognition processing according to the obtained preprocessed video monitoring picture to recognize a bird target in the video monitoring picture, and obtains a bird target recognition result.

[0124] In the method, the inspection window is used to sequentially traverse each position of the picture, and specifically, a rectangular window with a size of N*N is used to move the inspection window in a 50% overlapping manner to traverse each position in the video monitoring picture to obtain a corresponding region division result; N can be set as N=3, 5, 7, 9… according to actual conditions, or N=10%L according to the total size of the video monitoring picture, where L represents the length and width size of the video monitoring picture.

[0125] In a more extreme case, the traversal manner of the inspection window can also be sequentially using the traversal window to align each pixel point in the image, that is, each pixel point obtains a corresponding inspection result, but this manner needs to consume a large amount of computing resources, and in the actual process, a suitable window overlap amount can be selected according to actual conditions.

[0126] In an optimal case, according to the performance of the image recognition engine, the size of the inspection window can be set as the minimum image size required to recognize the birds, for example, the minimum image block that the image recognition engine can accurately recognize the birds is 5*5, and the size of the inspection window can be set as 5*5 or 7*7 to avoid invalid processing caused by too small window size or over-processing caused by too large window size.

[0127] The above embodiment of the present application is directed to the problem that the sky part of a large range area usually exists in the obtained video monitoring data, which is easy to cause the image to be affected by the change of sky brightness, resulting in unclear image. A technical solution based on a preprocessing unit for picture extraction and picture enhancement of the obtained video monitoring picture is particularly proposed. For the obtained video monitoring data, the video monitoring data is first sampled based on a video picture frame extraction technology to obtain a video monitoring picture. Further, color space conversion is performed according to the obtained video monitoring picture, and the brightness channel subgraph in the Lab color space is taken as the basis to enhance the video monitoring picture. According to the obtained brightness channel subgraph, a sub-region division technology is first proposed, and the brightness characteristic value of each region in the image is calculated by the proposed inspection window. In the process of calculating the brightness characteristic value, the performance characteristic that the brightness change gradient of the sky area part in the monitoring picture is not large is considered, and then the brightness characteristic function is constructed by parameters such as standard value and average value to numerically represent the brightness characteristic of the region, and further the feature region (such as a large range of sky part) and the transition region (such as a bird target region with large brightness change characteristic, a background region, a noise region, etc.) in the image are extracted by the brightness characteristic value. For the feature region, a technical solution of overall brightness adjustment is proposed to adaptively adjust the brightness level in the feature region (which can be adjusted for both day and night), which can effectively adapt to the change of sky brightness in various situations, adjust the overall brightness level of the image, and improve the overall clarity of the video monitoring picture. Further, for the transition region, a judgment scheme is further proposed to judge the characteristics of each pixel point, which can accurately analyze the situation of the pixel point by the judgment function. In particular, a first judgment function is proposed to adapt to the characteristics of the bird target monitoring video picture (in the actual monitoring process, the video monitoring picture usually monitors a large range of area, but in some special cases, such as a position very close to the lens, if other flying targets such as insects appear, the target will disturb the picture in a high-speed moving way, affecting the picture clarity, or the shaking leaves will cause the sun to penetrate and appear a flickering light, which will also cause unclear areas in the picture), and a second judgment function is combined to adapt to the blurred area of the day object backlight or the night flash point, and a second brightness adjustment processing function is combined to intelligently identify and adaptively adjust the situation of the bird target in the actual light environment, adaptively weaken the interference target, improve the clarity of the bird target, and maximize the representation level of the bird target feature details in the video monitoring picture.Finally, the video monitoring picture is reconstructed based on the brightness channel subgraphs which complete the brightness adjustment of the feature region and the transition region respectively, the definition of the bird target in the picture is improved, and the foundation is laid for further bird target recognition according to the video monitoring picture.

[0128] Preferably, the regional control module comprises a control unit and a warning unit.

[0129] The control unit is configured to send a driving instruction to the on-site bird repelling device in the control region according to the obtained control region information, so that the on-site bird repelling device is started and the bird is driven.

[0130] The warning unit is configured to send a driving instruction to the on-site bird repelling device in the warning region according to the obtained warning region information, so that the on-site bird repelling device is started and the bird is prevented.

[0131] The bird repelling device comprises an ultrasonic bird repelling device, a flashing light bird repelling device, a laser bird repelling device, etc.

[0132] Based on the obtained bird monitoring result, for the control region, the control region where the bird target currently appears is controlled, and the bird repelling device in the control region is started to drive the bird target in the control region. Meanwhile, based on the obtained warning region information, the next region where the bird target may move to or the key region where the key equipment exists is jointly driven, so as to avoid the bird moving to other regions which may cause adverse effects under the influence of the bird repelling device in the control region. Through the above directional bird repelling mode, the effect of controlling the bird damage can be improved, and the influence of the traditional single-point bird repelling mode is reduced, so as to improve the intelligent level and reliability of the bird damage control of the power transmission network.

[0133] When the bird target intrudes into the power transmission network monitoring region, the outermost control target can identify the existence of the bird, the region where the bird target currently exists is marked as a control region, the bird repelling device in the control region is started to drive the bird target, and in order to further intervene in the driving route during the driving of the bird target, the adjacent monitoring region in the corresponding direction (the adjacent monitoring region extending along the predicted flight trajectory direction) is marked as a warning region according to the prediction result of the bird flight trajectory, and when the adjacent monitoring region of the control region exists a key region, the key region is also regarded as a warning region. When the bird repelling device in the control region is started, the bird repelling device in the adjacent warning region is also started, so that the bird repelling devices in the warning region form a cooperation, avoid the bird target flying further in the warning region, change the flight trajectory of the bird target, form the directional bird repelling of the bird target, and realize the effective protection of the power transmission network monitoring region.

[0134] Preferably, the system further comprises a log module;

[0135] The log module is used to generate a corresponding bird damage supervision log according to the bird monitoring results obtained by each monitoring area and the corresponding driving instructions.

[0136] Through the log module, the bird monitoring results and the starting data of the bird repelling device are effectively recorded, which can help the manager to further maintain and update the management and control system.

[0137] Preferably, the system further comprises an information database module;

[0138] The information database module is used to extract bird characteristic information according to the obtained bird monitoring results, and to construct a power grid bird damage information database, wherein the power grid bird damage information database records the bird species, quantity, occurrence time period, frequency, time characteristics and other data in the power grid area.

[0139] Based on the recorded data, the information database module is further constructed to collect and manage the data, and to construct a bird damage monitoring and control large database, which helps to further analyze and call the data based on the constructed information database in the subsequent, and improves the data management level.

[0140] Referring to Figure 3 , the present application shows a power grid bird damage risk control method, wherein the method is realized based on the power grid bird damage risk control system as shown in Figure 1 , and the method comprises:

[0141] The regional management module divides the monitoring areas of the power grid coverage area according to the position distribution information of the key equipment of the power grid based on the preset three-dimensional GIS map model, and obtains the division information of each monitoring area;

[0142] The field monitoring module receives the field monitoring information transmitted by the field monitoring equipment, and integrates the obtained field monitoring information into the corresponding monitoring area, wherein the field monitoring information comprises video monitoring data;

[0143] The bird damage analysis module identifies the bird targets existing in the monitoring area according to the obtained field monitoring information, and further extracts the motion trajectory of the bird targets to obtain the bird monitoring results;

[0144] The regional control module sends driving instructions to the field bird repelling device of the corresponding monitoring area according to the obtained bird monitoring results, so that the field bird repelling device starts and completes the bird repelling.

[0145] It should be noted that the above method can also be based on the power grid bird damage risk control system as shown in Figure 1The power transmission network bird hazard risk management and control system shown in the embodiments corresponds to the implementation of the management and control system shown in the embodiments, that is, the above method also corresponds to containing Figure 1 The data processing method corresponding to each functional module in the system shown in the embodiments is not repeated in the present application.

[0146] Through the above-mentioned power transmission network bird hazard risk management and control method, first, based on the preset three-dimensional GIS map model, the area of the power transmission network is divided into regions, and the corresponding monitoring region division information is obtained; real-time receiving of the field monitoring information returned by each region, and according to the field monitoring information, bird target identification and bird target motion trajectory prediction are carried out, and the bird monitoring result is obtained, according to the bird monitoring result, further control of the bird repelling device in the corresponding region is carried out, so as to realize the bird driving work in the management and control region, which helps to improve the real-time level of the power transmission network bird hazard risk detection, and helps to improve the initiative and intelligent level of the bird hazard risk management and control, and improves the safety of the daily operation and control of the power transmission network.

[0147] It should be noted that each functional unit / module in each embodiment of the present application can be integrated in one processing unit / module, or each unit / module can be physically present alone, or two or more units / modules can be integrated in one unit / module. The above integrated unit / module can be realized in the form of hardware or in the form of software functional unit / module.

[0148] Those skilled in the art can clearly understand from the description of the above embodiments that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be instructed by a computer program to relevant hardware. When implemented, the above program can be stored in a computer readable medium or transmitted as one or more instructions or codes on a computer readable medium. The computer readable medium includes computer storage medium and communication medium, wherein the communication medium includes any medium facilitating the transmission of computer programs from one place to another. The storage medium can be any available medium accessible by a computer. The computer readable medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not limited to the scope of the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should analyze that the technical solutions of the present application can be modified or replaced by equivalents without departing from the essence and scope of the present application.

Claims

1. A power transmission network bird damage risk management system, characterized in that, It includes a regional management module, a field monitoring module, a bird damage analysis module, and a regional control module; among which, The area management module is used to divide the monitoring area of ​​the power transmission network coverage area based on a preset 3D GIS map model and the location distribution information of key equipment in the power transmission network, and obtain the division information of each monitoring area. The on-site monitoring module is used to receive on-site monitoring information transmitted by on-site monitoring equipment and integrate the acquired on-site monitoring information into the corresponding monitoring area, wherein the on-site monitoring information includes video monitoring data; The bird damage analysis module is used to identify bird targets in the monitoring area based on the acquired on-site monitoring information, and further extract the movement trajectory of the bird targets to obtain bird monitoring results; The area control module is used to send drive commands to the on-site bird deterrence devices in the corresponding monitoring area based on the acquired bird monitoring results, so that the on-site bird deterrence devices can be activated and the birds can be driven away. The bird damage analysis module includes a video analysis unit; The video analysis unit includes an extraction unit, a preprocessing unit, a recognition unit, a trajectory unit, and a result output unit. The preprocessing unit performs image extraction and image enhancement processing on the obtained video monitoring data, including: Based on the acquired video monitoring data, images are extracted, and video monitoring images at each time point are extracted according to a preset time interval. ,in express Real-time video monitoring footage; Based on the video monitoring footage at the current moment The video monitoring image was converted to the Lab three-channel color space, and the luminance channel sub-images of the video monitoring image were obtained. First color channel sub-image Second color channel subgraph ; Based on the obtained brightness channel subgraph The screen is divided into multiple sub-regions, including: The inspection window is used to traverse each position in the image sequentially; during the traversal, the brightness feature value of the current inspection window is calculated based on the brightness channel value of each pixel within the inspection window range. Based on the brightness characteristic value of the current inspection window With preset feature threshold Comparison, among which When the brightness feature value is less than the feature threshold If the brightness feature value is greater than or equal to the feature threshold, then the pixels covered by the current inspection window are marked as a class of pixels; otherwise, if the brightness feature value is greater than or equal to the feature threshold, then the pixels are marked as a class of pixels. When the current inspection window covers the pixel, mark it as a second-class pixel. If a pixel is marked as both a first-class pixel and a second-class pixel, then record the pixel as a second-class pixel. The region bounded by a connected class of pixels is marked as the feature region. The area enclosed by the connected two types of pixels is marked as the transition region. ; Regarding the obtained Then for The pixels in the image undergo initial brightness adjustment. Based on the brightness channel values ​​of each pixel after the first brightness adjustment process, the brightness adjustment result is obtained. And update the brightness channel sub-image of the current video monitoring screen. ; Based on the current brightness channel subgraph Further Perform a second brightness adjustment process, including: right Each pixel is judged; Based on the judgment results of each pixel, The pixels within the range undergo a second brightness adjustment process, where the second brightness adjustment function is: in, Indicates the pixel after the second adjustment process. The brightness channel value, Indicates the current pixel. The brightness channel value; This indicates the preset brightness channel adjustment value. ; This is the first judgment function. This is the second judgment function. This is the third judgment function; As the first weighting factor, As the second weighting factor, It is the third weighting factor; Based on the luminance channel values ​​of each pixel after the second luminance adjustment process, the PB after the second luminance adjustment process is obtained, and the luminance channel sub-image of the current video monitoring frame is updated. ; According to the updated , and Channel reconstruction is performed to obtain pre-processed video monitoring images.

2. The power transmission network bird damage risk management system according to claim 1, characterized in that, The regional management module includes GIS map units, power transmission network information units, and regional division units; among them, The GIS map unit is used to input map data corresponding to the actual geographic information of the power transmission network area to obtain a three-dimensional GIS map model. The power transmission network information unit is used to input relevant power transmission equipment data and field monitoring equipment data according to the actual setting of power transmission equipment in the power transmission network area, and to mark the relevant power transmission equipment and field monitoring equipment in the three-dimensional GIS map model; wherein the power transmission equipment includes power transmission towers, power transmission lines, insulators, and transformer boxes, the power transmission equipment data includes the specific setting location and / or distribution location of the power transmission equipment, and the field monitoring equipment includes cameras; The area division unit is used to divide the monitoring area of ​​the power transmission network coverage area according to the location distribution information of key equipment in the power transmission network, and obtain the division information of each monitoring area.

3. The power transmission network bird damage risk management system according to claim 2, characterized in that, The regional division units further include: Based on the location distribution information of key equipment in the power transmission network, the coverage area of ​​the power transmission network is adaptively divided to obtain specific division information for each monitoring area, including: During the initialization phase, the power grid coverage area is initially divided into N adjacent sub-regions according to the preset number of divisions. During the adaptive evolution phase, the bird control weight factor for each sub-region is calculated, and the calculation function for the bird control weight factor is as follows: in, This represents the bird control weight factor for the i-th sub-region. This represents the weighting factor for power transmission equipment. Indicates the weighting factor for the severity of bird damage. and Indicates the preset weight; The size of each sub-region is adjusted according to the bird control weight factor of each sub-region. When the bird control weight factor is greater than the average bird control weight factor of each sub-region, the area of ​​the sub-region is reduced; otherwise, when the bird control weight factor is less than the average bird control weight factor of each sub-region, the area of ​​the sub-region is expanded. Repeat the above adaptive evolution phase process, and end the adaptive evolution phase when the division of each sub-region tends to be stable or the maximum number of repetitions is reached; Based on the division of each sub-region after adaptive evolution, each sub-region is treated as a monitoring area, thus completing the division of monitoring areas and obtaining the division information of each monitoring area.

4. The power transmission network bird damage risk management system according to claim 2, characterized in that, The on-site monitoring module includes a receiving unit and a correlation unit; among which, The receiving unit is used to establish a communication connection with the field monitoring equipment set up in each monitoring area and to receive the field monitoring information transmitted back by the field monitoring equipment in real time. The association unit is used to associate the acquired on-site monitoring information with the 3D GIS map model, associate the acquired on-site monitoring information with the corresponding monitoring area, and integrate the on-site monitoring information into the 3D GIS map model.

5. A power transmission network bird damage risk management system according to claim 1, characterized in that, The regional control module includes a control unit and an early warning unit; The control unit is used to send drive commands to the on-site bird deterrence device in the controlled area based on the obtained control area information, so that the on-site bird deterrence device can be activated and complete the bird driving away; The early warning unit is used to send drive commands to the on-site bird deterrence devices in the early warning area based on the obtained early warning area information, so that the on-site bird deterrence devices can be activated and complete the directional prevention of birds.

6. The power transmission network bird damage risk management system according to claim 1, characterized in that, It also includes a logging module; The log module is used to generate corresponding bird damage monitoring logs based on the bird monitoring results obtained from each monitoring area and the corresponding drive instructions.

7. A power transmission network bird damage risk management system according to claim 1, characterized in that, It also includes an information database module; The information database module is used to extract bird characteristic information based on the obtained bird monitoring results and construct a bird damage information database for the power transmission network. The bird damage information database records data on the types, numbers, time periods, frequencies, and time characteristics of birds appearing in the power transmission network area.

8. A method for controlling bird damage risks in a power transmission network based on any one of the power transmission network bird damage risk control systems described in claims 1-7, characterized in that, include: The regional management module is based on a preset 3D GIS map model. According to the location distribution information of key equipment in the power transmission network, it divides the coverage area of ​​the power transmission network into monitoring areas and obtains the division information of each monitoring area. The on-site monitoring module receives on-site monitoring information transmitted by the on-site monitoring equipment and integrates the acquired on-site monitoring information into the corresponding monitoring area, including video monitoring data. The bird damage analysis module identifies bird targets in the monitoring area based on the acquired on-site monitoring information, and further extracts the movement trajectories of the bird targets to obtain bird monitoring results; Based on the acquired bird monitoring results, the area control module sends a drive command to the on-site bird deterrent device in the corresponding monitoring area, so that the on-site bird deterrent device can be activated and complete the bird driving away.

Images