Analysis alarm video monitoring platform for wind power generation equipment

Abstract

The invention provides an analysis alarm video monitoring platform for wind power generation equipment, and relates to the technical field of video monitoring, the video monitoring platform comprises a real-time picture acquisition module, a historical picture acquisition module, a gridding division module, an analysis module and an alarm generation and management module, the video monitoring platform is also provided with a three-level architecture, the three-level architecture comprises a group headquarter unit, a province unit and a basic level unit. According to the method, after the alarm is triggered through grid analysis, the current / historical picture data is automatically associated, and the alarm event with the grade is generated by means of the event playback function, so that the false alarm caused by environmental interference is greatly reduced, and the alarm accuracy is improved. And a current picture transmitted by the video acquisition module is obtained through a real-time preview mode, and is compared with a specified historical moment picture called from the hierarchical storage system through the intelligent index, and the historical picture is used as a contrast to judge whether an abnormality exists or not and whether alarm processing needs to be carried out or not.

Description

Technical Field

[0001] The present invention relates to the technical field of video monitoring, and in particular to an analysis and alarm video monitoring platform for wind power generation equipment. Background Art

[0002] The analysis and alarm video monitoring platform for wind turbines monitors the operating status and external environmental changes of wind turbines (including blades, nacelles, towers, foundations, and surrounding environments) in real time, remotely, and automatically. It intelligently identifies abnormal conditions or potential fault hazards such as personnel wandering monitoring, helmet monitoring, smoking detection, remote equipment monitoring, equipment oil leakage detection, and respirator detection, and issues accurate and timely alarms. The significance of this platform lies in greatly improving the safety protection level and operation and maintenance response efficiency of wind farms, transforming traditional manual passive inspections into active predictive maintenance, effectively avoiding the risk of major safety accidents. In the prior art, publication number CN 117201737A discloses a video stream cascade transmission method and system based on upper and lower level video surveillance platforms. The technology includes: notifying the public network IP and port of the upper-level video surveillance platform to the lowest-level video surveillance platform as the source of the video stream. After the lowest-level video surveillance platform defines the required intranet IP address and port, it sends a packet to the specific port of the upper-level monitoring platform. In this way, the upper-level platform can obtain the corresponding public network IP address and port required by the lower-level monitoring platform. After that, the two modify the IP address and port contained in the application layer protocol unit to the public network, and then signaling interaction and direct transmission of video streams can be carried out through these public network IP and ports.

[0003] However, existing technologies rely solely on single-point image analysis, primarily real-time monitoring images, lacking a comparison mechanism with historical benchmark images. Relying solely on isolated real-time single-frame images for analysis, it is unable to establish a historical benchmark image library and lacks the ability to capture dynamic features of spatiotemporal continuity. For example, it is unable to verify the authenticity of an object's movement trajectory through multiple consecutive frames. As a result, in dynamic scenes such as wind farms, environmental noise such as high-frequency periodic swaying of vegetation, morphological changes in equipment projections, meteorological disturbances, and low-light noise interference at night are mistakenly identified as real threats. Furthermore, by ignoring the correlation between the equipment's operating status and the environment, for example, the morphological changes in wind turbine projections at different speeds are not modeled, resulting in a serious lack of filtering capabilities for non-threatening dynamic targets. This ultimately results in a high false alarm rate, significantly reducing monitoring effectiveness.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide an analysis and alarm video monitoring platform for wind power generation equipment to solve the problems raised in the above-mentioned background technology. After the alarm is triggered by grid analysis, the present invention automatically associates the current / historical screen data, generates graded alarm events with the help of the event playback function, and pushes them to the management interface, which greatly reduces the false alarms caused by environmental interference and improves the accuracy of the alarm. The current screen transmitted by the video acquisition module is obtained through the real-time preview mode, and is compared with the screen of the specified historical moment retrieved from the hierarchical storage system through intelligent indexing. The historical screen is used as a reference to determine whether there is an abnormality and whether alarm processing is required. Instantaneous interference can be filtered out and the accuracy is improved.

[0006] To achieve the above object, the present invention provides the following technical solutions: The video monitoring platform for wind power generation equipment analysis and alarm includes the following functional modules: Real-time image acquisition module: This module acquires the current image data captured by the camera in real time. The system then encodes and transmits the video stream to the storage and analysis unit based on the real-time preview mode. The historical image acquisition module retrieves and quickly locates historical video files and obtains image data from a specified time point before the current time point. Through indexing and association, combined with an event-triggered hierarchical storage system, this data is used for subsequent grid-based comparative analysis. Gridding module: This module is used to generate grids, dividing the image into multiple independent units through the grid. It performs pixel-level matching of grids at the same location in the current image and historical images, eliminating coordinate offsets caused by camera jitter or lighting changes. It binds a corresponding detection algorithm to each grid and supports user-adjustable sensitivity. By isolating non-critical areas, it focuses only on specific grids for change comparison. Users can manually select parts of the gridded area through the front-end interactive interface. Analysis module: used to detect anomalies in each grid area, including human behavior, equipment status, and environmental risks. The operation process is to load the grid data of the current and historical images into the analysis engine, compare the image changes at two time points, calculate the absolute difference between the current frame grid and the historical reference frame grid, and determine whether there is an anomaly in the current monitoring image based on the absolute difference. If an anomaly is detected, an alarm signal is triggered; Alarm generation and management module: This module includes an alarm server and a front-end management interface. It is used to automatically generate alarm information when an anomaly is detected and issue an alarm through the alarm management function. The operating process is that after the abnormal signal triggers the alarm, the system combines traceable playback technology to store the video before and after the alarm to achieve rapid response and processing.

[0007] Furthermore, the real-time image acquisition module captures the original data stream in real time, uses the H.264 / H.265 encoding standard for frame-level compression and encapsulation, establishes a multi-channel transmission channel for video stream distribution, and synchronously embeds timestamps, device IDs, and GPS positioning metadata; for the historical image acquisition module, the video surveillance platform constructs a dual index system of time and address to generate key frame fingerprints; The real-time image acquisition module also has a dynamic bit rate adjustment mechanism for adaptively adjusting the resolution and frame rate according to network bandwidth fluctuations. In abnormal situations, the frame loss compensation algorithm is triggered and data verification and retransmission are started to ensure image integrity.

[0008] Furthermore, the gridding module obtains the resolution and object scale characteristics of the real-time monitoring screen and the historical screen, and uses an adaptive grid algorithm to divide the real-time video frame into several independent unit grids. Each grid has a unique coordinate identifier, and the grayscale value / feature point comparison is performed on the grids at the same position in the current frame and the historical frame. The offset vector is calculated through the feature matching algorithm. If the overall displacement of the grid exceeds the threshold, the homography transformation matrix is ​​triggered to automatically correct the camera jitter. In response to lighting changes, histogram equalization is used to eliminate interference. The threshold of the overall displacement is 5 pixels. The gridding module also provides sensitivity adjustment and regional focus functions. Users can set the change detection sensitivity by dragging a slider. The sensitivity adjustment range is 1%-100%. Low sensitivity only recognizes the movement of large objects, while high sensitivity captures subtle changes. In the regional focusing function, the key grid is manually selected in the front-end interface, and the system automatically isolates the non-selected area and only performs the inter-frame difference method or background subtraction algorithm on the target grid to reduce computing power consumption.

[0009] Furthermore, the analysis module is equipped with a time and address comparison engine, and the anomaly detection mechanism operation process of the time and address comparison engine includes the following: The analysis module calls the data pre-processed by the gridding module: the current frame grid and the historical reference frame grid are strictly matched in terms of coordinates, and the grid position deviation caused by lens offset is automatically corrected through affine transformation to ensure the consistency of the comparison space; First, pixel-level spatiotemporal alignment is performed, using the coordinate offset correction results of the gridding module. Then, data within each grid cell is collected. The collection range includes the mean absolute difference in the grayscale values ​​of all pixels in the current frame grid and the historical reference frame grid. The absolute difference between the current frame grid and the historical reference frame grid is calculated using the following formula; , in: C(i,j) is the grayscale value of the pixel in the i-th row and j-th column of the current frame image grid; S(i,j) is the grayscale value of the pixel in the i-th row and j-th column of the current frame image grid; is the absolute grayscale difference between the current frame C and the historical frame S; W is the image grid width; H is the image grid height; Compare the calculated results Δ1 with Δ2, where Δ2 is the abnormality judgment threshold for the change area. If: Δ1<Δ2 It is determined that no abnormality has occurred and no alarm signal is triggered if: Δ1≥Δ2 It is determined that an abnormality has occurred and an alarm signal is sent from the station to the provincial unit.

[0010] Furthermore, the alarm generation and management module implements full-link alarm handling through an intelligent pipeline: when the grid analysis module detects an abnormal signal, the alarm server immediately triggers a five-step closed-loop process: Automatically associate abnormal grid coordinates, equipment types, and abnormal types to generate structured alarms, call traceable playback technology, store video clips 60 seconds before and 60 seconds after the alarm is triggered, mark the abnormal grid location, and distribute alarms according to preset rules; operation and maintenance personnel view alarm videos and mark processing status through the interactive interface, and automatically escalate and supervise when the processing timeouts; after processing is completed, the system automatically archives the alarm data to the database, associates it with equipment maintenance records, and provides samples for algorithm optimization.

[0011] Furthermore, the distributed alarms are accurately routed and distributed according to the alarm type, risk level and management responsibility: when the alarm server generates an event, the impact range is divided and the push level is determined according to the impact range.

[0012] Furthermore, the video surveillance platform is also built with a three-level architecture, which includes group headquarters units, provincial units, and grassroots units. The group headquarters unit is the top-level decision-making center of the monitoring platform; the provincial units serve as regional hubs to collect video streams and equipment data from grassroots units in real time, perform preliminary screening and compressed storage, and support real-time monitoring, historical playback and event alarms, so as to facilitate rapid response and coordination of provincial management personnel. The grassroots units are directly deployed in photovoltaic power stations and onshore wind farms to provide video surveillance capabilities.

[0013] Furthermore, in the three-level architecture of the video surveillance platform, alarm information follows a bottom-up transmission process: after the front-end device detects an anomaly, it triggers a local alarm and captures the real-time image, which is initially filtered and formatted by the station-side system and uploaded to the provincial unit's centralized control-level master station; the centralized control-level master station integrates multi-site alarm information, conducts risk assessment and linkage analysis, and further reports to the group headquarters unit master station if cross-domain coordination or major incident handling is required; the group headquarters unit master station activates the emergency plan based on global data, centrally displays the alarm situation through the large-screen visualization system, and coordinates resource scheduling. Conversely, control instructions are transmitted from top to bottom: the group headquarters unit master station issues emergency command strategies or system parameter configurations to the centralized control-level master station through the management server, the centralized control-level master station executes the instructions and distributes them to the station-side systems to which it belongs, and ultimately the site-side system controls the front-end equipment.

[0014] Furthermore, in the three-level architecture of the video surveillance platform, control instructions are transmitted from top to bottom: the group headquarters unit master station sends emergency command strategies or system parameter configurations to the provincial unit's centralized control-level master station through the management server, the centralized control-level master station executes the instructions and distributes them to the station-end system, and finally the station-end system controls the front-end equipment.

[0015] Compared with the prior art, the present invention has the following beneficial effects: When grid analysis triggers an alarm, it automatically correlates current and historical footage data. Using the event playback function, a graded alarm event is generated and pushed to the management interface. Users can quickly process the event through multi-dimensional filtering (time, status, and type), ensuring full process traceability. The system segments the footage into customizable detection units and combines multiple detection algorithms (such as loitering detection, hard hat detection, smoking detection, remote equipment monitoring, oil leak detection, respirator detection, open flame detection, surface water detection, door detection, fence climbing detection, human boundary detection, warehouse exit notification, smoke detection, tooling detection, oil level detection, and fire escape blockage) to focus on critical areas for image comparison. This design significantly reduces false alarms caused by environmental interference and improves alarm accuracy. The current footage transmitted by the video capture module is captured in real-time preview mode and compared with footage at specified historical moments retrieved from the tiered storage system using intelligent indexing. Using historical footage as a reference, the system determines whether anomalies exist and whether alarm processing is necessary. This eliminates transient interference and improves accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is an operation flow chart of the analysis and alarm video monitoring platform for wind power generation equipment of the present invention; Figure 2 3. It is a three-level architecture diagram of the video surveillance platform in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0018] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly. Example

[0019] See also Figure 1-Figure 2 , the present invention provides a technical solution: The video monitoring platform for wind power generation equipment analysis and alarm includes the following functional modules: Real-time image acquisition module: This module acquires the current image data captured by the camera in real time. The system then encodes and transmits the video stream to the storage and analysis unit based on the real-time preview mode. The historical image acquisition module retrieves and quickly locates historical video files and obtains image data from a specified time point before the current time point. Through indexing and association, combined with an event-triggered hierarchical storage system, this data is used for subsequent grid-based comparative analysis. Gridding module: This module is used to generate grids, dividing the image into multiple independent units through the grid. It performs pixel-level matching of grids at the same location in the current image and historical images, eliminating coordinate offsets caused by camera jitter or lighting changes. It binds a corresponding detection algorithm to each grid and supports user-adjustable sensitivity. By isolating non-critical areas, it focuses only on specific grids for change comparison. Users can manually select parts of the gridded area through the front-end interactive interface. Analysis module: used to detect anomalies in each grid area, including human behavior, equipment status, and environmental risks. The operation process is to load the grid data of the current and historical images into the analysis engine, compare the image changes at two time points, calculate the absolute difference between the current frame grid and the historical reference frame grid, and determine whether there is an anomaly in the current monitoring image based on the absolute difference. If an anomaly is detected, an alarm signal is triggered; Alarm generation and management module: This module includes an alarm server and a front-end management interface. It is used to automatically generate alarm information when an anomaly is detected and issue an alarm through the alarm management function. The operating process is that after the abnormal signal triggers the alarm, the system combines traceable playback technology to store the video before and after the alarm to achieve rapid response and processing.

[0020] In this embodiment, a video surveillance platform is also provided to access a variety of monitoring terminals, including box cameras, dome cameras, and drones. The box cameras are fixed monitoring devices: deployed at key locations, they use high-precision target detection algorithms to identify intrusions by people, vehicles, or abnormal objects in real time. Combined with behavioral analysis algorithms, they automatically trigger regional intrusion alarms for behaviors such as photovoltaic panel theft and wind power equipment damage, and link with dome cameras for tracking. In this embodiment, the gun camera serves as a fixed monitoring terminal. It uses a high-resolution CMOS / CCD sensor to capture continuous images of key points, uses H.265 encoding to compress the video stream in real time, and transmits it to the platform's central server via Ethernet or fiber optic network. In the pre-processing stage, it automatically performs grayscale correction and Gaussian filtering to remove noise, and embeds timestamps and geographic location metadata. It identifies intrusions by people, vehicles, or abnormal objects in real time, and performs pattern matching on specific actions (such as photovoltaic panel theft and wind power equipment damage). Once a threat is detected, the regional intrusion alarm mechanism is automatically triggered, and the dome camera is linked to perform dynamic tracking to ensure that the entire process delay is less than 300ms. For behaviors such as photovoltaic panel dismantling, wind power equipment damage, and illegal detention, the regional intrusion alarm is automatically triggered and captured for evidence. The dome camera relies on a panoramic stitching algorithm to achieve large-scale dynamic monitoring. It uses deep learning behavior recognition to intelligently analyze fire risks around wind turbine tower bases and photovoltaic areas. It uses a heat map algorithm to calculate personnel density, assisting grassroots units in dispatching security resources, and sends instant pop-up alerts to provincial platforms for abnormal behavior. The dome camera in this implementation provides 360° horizontal and 90° vertical pan / tilt rotation, eliminates parallax ghosting through local homography grid transformation (APAP algorithm), generates seamless panoramas, and adapts to lighting changes, capable of automatically adjusting the camera's viewing angle in any panoramic area. The drone provides mobile inspections and is equipped with a sensor fusion algorithm. It uses a path planning algorithm to autonomously navigate photovoltaic arrays and wind turbine clusters. It utilizes an infrared hotspot detection algorithm to locate photovoltaic panel faults and a blade deformation analysis algorithm to identify cracks or ice on wind turbine blades. It also incorporates SLAM technology to generate a three-dimensional real-world model. Abnormal data is transmitted back to the group headquarters algorithm platform in real time to generate equipment health statistics. The drone's core sensor configuration in this embodiment includes a hyperspectral camera, which analyzes abnormalities such as surface contamination of photovoltaic panels (such as dust accumulation and bird droppings corrosion) and the aging of blade coatings.

[0021] In this embodiment, the real-time image acquisition module captures the original data stream in real time, uses the H.264 / H.265 encoding standard for frame-level compression and encapsulation, establishes multiple transmission channels for video stream distribution, and synchronously embeds timestamps, device IDs, and GPS positioning metadata. For the historical image acquisition module, the video surveillance platform constructs a dual index system of time and address to generate key frame fingerprints. The real-time image acquisition module also has a dynamic bit rate adjustment mechanism for adaptively adjusting the resolution and frame rate according to network bandwidth fluctuations. In abnormal situations, the frame loss compensation algorithm is triggered and data verification and retransmission are started to ensure image integrity.

[0022] The gridding module also provides sensitivity adjustment and regional focus functions. Users can set the change detection sensitivity by dragging a slider. The sensitivity adjustment range is 1%-100%. Low sensitivity only recognizes the movement of large objects, while high sensitivity captures subtle changes. In the regional focusing function, the key grid is manually selected in the front-end interface, and the system automatically isolates the non-selected area and only performs the inter-frame difference method or background subtraction algorithm on the target grid to reduce computing power consumption.

[0023] In this embodiment, the analysis module is used to detect anomalies in each grid area in the real-time monitoring screen and the comparison historical screen. The operation process is that the analysis engine loads the grid data of the current and historical screens, and by comparing the screen changes at two time points, calculates the absolute difference between the current frame grid and the historical reference frame grid. The absolute difference is used to determine whether there is an anomaly in the current monitoring screen. If an anomaly is detected, an alarm signal is triggered.

[0024] In this embodiment, the analysis module is equipped with a time and address comparison engine, and the anomaly detection mechanism operation process of the time and address comparison engine includes the following: The analysis module calls the data pre-processed by the gridding module: the current frame grid and the historical reference frame grid are strictly matched in terms of coordinates, and the grid position deviation caused by lens offset is automatically corrected through affine transformation to ensure the consistency of the comparison space; First, pixel-level spatiotemporal alignment is performed, using the coordinate offset correction results of the gridding module. Then, data within each grid cell is collected. The collection range includes the mean absolute difference in the grayscale values ​​of all pixels in the current frame grid and the historical reference frame grid. The absolute difference between the current frame grid and the historical reference frame grid is calculated using the following formula; , in: C(i,j) is the grayscale value of the pixel in the i-th row and j-th column of the current frame image grid, and the theoretical range of the grayscale value is 0~255; S(i,j) is the grayscale value of the pixel in the i-th row and j-th column of the current frame image grid, and the theoretical range of the grayscale value is 0~255; is the absolute grayscale difference between the current frame C and the historical frame S. Each group (i, j) represents a fixed pixel position. In this formula, each pixel in the image collected by the unified monitoring device is compared with the pixel at the same position in the historical image. W is the image grid width, which is the number of pixels of width in the grid within the demarcated range; H is the image grid height, which is the number of height pixels in the grid within the defined range; The larger the value of Δ1 in the calculation result is, the greater the degree of image change between the current frame grid and the historical reference frame grid is.

[0025] Compare the calculated results Δ1 with Δ2, where Δ2 is the abnormality judgment threshold for the change area. If: Δ1<Δ2 It is determined that no abnormality has occurred and no alarm signal is triggered if: Δ1≥Δ2 It is determined that an abnormality has occurred and an alarm signal is sent from the station to the provincial unit.

[0026] In this embodiment, a mean absolute difference calculation scheme is also provided, namely: , where Δ m is the mean absolute difference, and its theoretical range is also 0~255. m When Δ<5, it indicates a slight change, such as light fluctuations. m ≤30, it is considered abnormal. m A value of ≥30 indicates a significant change, such as that caused by object movement or scene changes. This formula is insensitive to abnormal pixels (such as transient noise), further improving the reliability of the final detection results.

[0027] In this embodiment, the alarm generation and management module implements full-link alarm handling through an intelligent pipeline: when the grid analysis module detects an abnormal signal, the alarm server immediately triggers a five-step closed-loop process: Automatically associate abnormal grid coordinates, equipment types, and abnormal types to generate structured alarms, call traceable playback technology, store video clips 60 seconds before and 60 seconds after the alarm is triggered, mark the abnormal grid location, and distribute alarms according to preset rules; operation and maintenance personnel view alarm videos and mark processing status through the interactive interface, and automatically escalate and supervise when the processing timeouts; after processing is completed, the system automatically archives the alarm data to the database, associates it with equipment maintenance records, and provides samples for algorithm optimization.

[0028] In this embodiment, the alarm distribution is based on the alarm type, risk level and management responsibility to achieve precise routing distribution: when the alarm server generates an event, the impact range is divided and the push level is determined according to the impact range: Alarm information is pushed to grassroots units: triggering pop-up windows and sound and light alarms on the monitoring screen, synchronously linking the dome camera to automatically turn and lock onto the target, and push handling instructions; Alarm information is pushed to provincial units: a structured operation and maintenance work order is generated, including video clips, grid coordinates, and device ID, and assigned to the nearest maintenance team's mobile terminal. If there is no response for more than 10 minutes, it will be automatically escalated to the on-duty supervisor; Alarm information is pushed to the group headquarters: Alarms are aggregated by device type / geographic region to generate health heat maps and forecast reports, driving algorithm model iteration.

[0029] In this embodiment, the video surveillance platform is also built with a three-level architecture, which includes group headquarters units, provincial units, and grassroots units. The group headquarters unit is the top-level decision-making center of the monitoring platform; the provincial units serve as regional hubs for collecting video streams and equipment data from grassroots units in real time, performing preliminary screening and compressed storage, and supporting real-time monitoring, historical playback, and event alarms, so as to facilitate rapid response and coordination of provincial management personnel. The grassroots units are directly deployed in photovoltaic power stations and onshore wind farms to provide video surveillance capabilities.

[0030] In this embodiment, in the three-level architecture of the video surveillance platform, alarm information follows a bottom-up transmission process: after the front-end device detects an anomaly, it triggers a local alarm and captures the real-time image, which is initially filtered and formatted by the station-side system and uploaded to the provincial unit's centralized control-level master station; the centralized control-level master station integrates multi-site alarm information, conducts risk assessment and linkage analysis, and further reports to the group headquarters unit master station if cross-domain coordination or major incident handling is required; the group headquarters unit master station activates the emergency plan based on global data, centrally displays the alarm situation through the large-screen visualization system, and coordinates resource scheduling. Conversely, control instructions are transmitted from top to bottom: the group headquarters unit master station issues emergency command strategies or system parameter configurations to the centralized control-level master station through the management server, the centralized control-level master station executes the instructions and distributes them to the station-side system to which it belongs, and ultimately the site-side system controls the front-end equipment.

[0031] In this embodiment, in the three-level architecture of the video surveillance platform, control instructions are transmitted from top to bottom: the group headquarters unit master station sends emergency command strategies or system parameter configurations to the provincial unit's centralized control-level master station through the management server, the centralized control-level master station executes the instructions and distributes them to the station-end system, and finally the station-end system controls the front-end equipment.

[0032] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0033] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0034] 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, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0035] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. Analysis and alarm video monitoring platform for wind power generation equipment, characterized by: The video surveillance platform includes the following functional modules: Real-time image acquisition module: This module acquires the current image data captured by the camera in real time. The system then encodes and transmits the video stream to the storage and analysis unit based on the real-time preview mode. The historical image acquisition module retrieves and quickly locates historical video files and obtains image data from a specified time point before the current time point. Through indexing and association, combined with an event-triggered hierarchical storage system, this data is used for subsequent grid-based comparative analysis. Gridding module: This module is used to generate grids, dividing the image into multiple independent units through the grid. It performs pixel-level matching of grids at the same location in the current image and historical images, eliminating coordinate offsets caused by camera jitter or lighting changes. It binds a corresponding detection algorithm to each grid and supports user-adjustable sensitivity. By isolating non-critical areas, it focuses only on specific grids for change comparison. Users can manually select parts of the gridded area through the front-end interactive interface. Analysis module: used to detect anomalies in each grid area, including human behavior, equipment status, and environmental risks. The operation process is to load the grid data of the current and historical images into the analysis engine, compare the image changes at two time points, calculate the absolute difference between the current frame grid and the historical reference frame grid, and determine whether there is an anomaly in the current monitoring image based on the absolute difference. If an anomaly is detected, an alarm signal is triggered; Alarm generation and management module: This module includes an alarm server and a front-end management interface. It is used to automatically generate alarm information when an anomaly is detected and issue an alarm through the alarm management function. The operating process is that after the abnormal signal triggers the alarm, the system combines traceable playback technology to store the video before and after the alarm to achieve rapid response and processing.

2. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 1 is characterized by: The real-time image acquisition module captures the original data stream in real time, uses the H.264 / H.265 encoding standard for frame-level compression and encapsulation, establishes multiple transmission channels for video stream distribution, and synchronously embeds timestamps, device IDs, and GPS positioning metadata. For the historical image acquisition module, the video surveillance platform constructs a dual index system of time and address to generate key frame fingerprints. The real-time image acquisition module also has a dynamic bit rate adjustment mechanism for adaptively adjusting the resolution and frame rate according to network bandwidth fluctuations. In abnormal situations, the frame loss compensation algorithm is triggered and data verification and retransmission are started to ensure image integrity.

3. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 2 is characterized by: The gridding module obtains the resolution and object scale characteristics of real-time monitoring images and historical images, and uses an adaptive grid algorithm to divide the real-time video frame into several independent unit grids. Each grid has a unique coordinate identifier. The grayscale value / feature point comparison is performed on the grids at the same position in the current frame and the historical frame, and the offset vector is calculated using a feature matching algorithm. If the overall displacement of the grid exceeds a threshold, the homography transformation matrix is ​​triggered to automatically correct camera jitter. In response to lighting changes, histogram equalization is used to eliminate interference. The threshold for the overall displacement is 5 pixels.

4. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 3 is characterized by: The gridding module also provides sensitivity adjustment and regional focus functions. Users can set the change detection sensitivity by dragging a slider. The sensitivity adjustment range is 1%-100%. Low sensitivity only recognizes the movement of large objects, while high sensitivity captures subtle changes. In the regional focusing function, the key grid is manually selected in the front-end interface, and the system automatically isolates the non-selected area and only performs the inter-frame difference method or background subtraction algorithm on the target grid to reduce computing power consumption.

5. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 4 is characterized in that: The analysis module is equipped with a time and address comparison engine, and the anomaly detection mechanism operation process of the time and address comparison engine includes the following: The analysis module calls the data pre-processed by the gridding module: the current frame grid and the historical reference frame grid are strictly matched in terms of coordinates, and the grid position deviation caused by lens offset is automatically corrected through affine transformation to ensure the consistency of the comparison space; First, pixel-level spatiotemporal alignment is performed, using the coordinate offset correction results of the gridding module. Then, data within each grid cell is collected. The collection range includes the mean absolute difference in the grayscale values ​​of all pixels in the current frame grid and the historical reference frame grid. The absolute difference between the current frame grid and the historical reference frame grid is calculated using the following formula; ， in: C(i,j) is the grayscale value of the pixel in the i-th row and j-th column of the current frame image grid; S(i,j) is the grayscale value of the pixel in the i-th row and j-th column of the current frame image grid; is the absolute grayscale difference between the current frame C and the historical frame S; W is the image grid width; H is the image grid height; Compare the calculated results Δ1 with Δ2, where Δ2 is the abnormality judgment threshold for the change area. If: Δ1<Δ2 It is determined that no abnormality has occurred and no alarm signal is triggered if: Δ1≥Δ2 It is determined that an abnormality has occurred and an alarm signal is sent from the station to the provincial unit.

6. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 1 is characterized by: The alarm generation and management module implements full-link alarm handling through an intelligent pipeline: when the grid analysis module detects an abnormal signal, the alarm server immediately triggers a five-step closed-loop process: Automatically associate abnormal grid coordinates, equipment types, and abnormal types to generate structured alarms, call traceable playback technology, store video clips 60 seconds before and 60 seconds after the alarm is triggered, mark the abnormal grid location, and distribute alarms according to preset rules; operation and maintenance personnel view alarm videos and mark processing status through the interactive interface, and automatically escalate and supervise when the processing timeouts; after processing is completed, the system automatically archives the alarm data to the database, associates it with equipment maintenance records, and provides samples for algorithm optimization.

7. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 6 is characterized in that: The distributed alarms are accurately routed and distributed according to the alarm type, risk level and management responsibility: when an alarm server generates an event, the impact range is divided and the push level is determined according to the impact range.

8. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 1 is characterized by: The video surveillance platform also has a three-level architecture, which includes group headquarters units, provincial units, and grassroots units. The group headquarters unit is the top-level decision-making center of the monitoring platform; the provincial units serve as regional hubs to collect video streams and equipment data from grassroots units in real time, conduct preliminary screening and compressed storage, and support real-time monitoring, historical playback and event alarms, so as to facilitate rapid response and coordination of provincial management personnel. The grassroots units are directly deployed in photovoltaic power stations and onshore wind farms to provide video surveillance capabilities.

9. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 8, characterized in that: In the three-level architecture of the video surveillance platform, alarm information follows a bottom-up transmission process: after the front-end device detects an anomaly, it triggers a local alarm and captures the real-time image, which is then initially filtered and formatted by the station-side system and uploaded to the provincial unit's centralized control-level master station; The centralized control master station integrates multi-site alarm information, conducts risk assessment and linkage analysis, and reports to the group headquarters unit master station if cross-domain coordination or major incident handling is required; The main station of the group headquarters unit initiates the emergency plan based on global data, centrally displays the alarm situation through the large-screen visualization system, and coordinates resource scheduling; conversely, the control instructions are transmitted from top to bottom: the main station of the group headquarters unit sends the emergency command strategy or system parameter configuration to the centralized control-level master station through the management server, the centralized control-level master station executes the instructions and distributes them to the station-end system, and finally the station-end system controls the front-end equipment.

10. The analysis and alarm video monitoring platform for wind power generation equipment according to claim 9, characterized in that: In the three-level architecture of the video surveillance platform, control instructions are transmitted from top to bottom: the group headquarters unit master station sends emergency command strategies or system parameter configurations to the provincial unit's centralized control-level master station through the management server, the centralized control-level master station executes the instructions and distributes them to the station-end system, and finally the station-end system controls the front-end equipment.

Images