Photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion

Abstract

The application discloses a photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, relates to the technical field of new energy, and comprises the following steps: deploying cameras and monitoring a photovoltaic power station, collecting equipment data and camera data of the photovoltaic power station in real time; performing secondary processing on the camera data to obtain processed camera data; and performing fault analysis based on the related data of the photovoltaic power station and the processed camera data, and positioning the fault of the photovoltaic power station. The application realizes quick switching between the best view angle mode and the detail view angle mode, reduces the waste of artificial resources, and improves the efficiency of fault positioning; the secondary processing of the camera video stream enables workers to quickly check the observed object; in combination with the operation data of the observed object and the topological relationship between the photovoltaic power station equipment, the causes of the fault are analyzed, and the positioning of the photovoltaic cable in the hidden engineering is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of new energy technology, in particular to a photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion. BACKGROUND

[0002] As a very popular new energy, photovoltaic power generation has more relaxed requirements for regional selection than hydropower and wind power during construction; in order to realize the monitoring of the photovoltaic power generation area, multiple cameras supporting 360-degree no-dead-angle rotation and zoom are installed in the photovoltaic power generation area; in the general working mode, the focal length of the monitoring camera in the photovoltaic power generation area is set to be relatively large, and multiple cameras can be used to see the overall situation of the photovoltaic power generation area; when it is necessary to focus on a certain key part, the focal length of the camera near the area to be viewed can be adjusted to a smaller mode, so as to facilitate the focused viewing of the concerned part; in order to realize the unattended operation of the photovoltaic power station, the photovoltaic power station generally deploys a photovoltaic power station operation state monitoring system, which can monitor the operation state of the photovoltaic power station in real time through a mobile App or Web, such as the current, voltage, power, switch breaker state and whether the temperature of the inverter is too high; however, in some large photovoltaic power stations, due to the large area occupied by the photovoltaic power station, the large number of equipment and cables in the photovoltaic power generation area, it is difficult to quickly locate the suspected fault part when a suspected fault occurs, so it is the research focus of the prior art to use the camera to quickly realize remote reconnaissance of the suspected fault part.

[0003] However, the current common solution has many shortcomings, including: large power station has a wide area and many devices, it is difficult to quickly locate the fault, and only manual on-site troubleshooting can be performed, resulting in waste of resources; in the photovoltaic power station, there are some hidden projects, such as the use of cable trench or cable bridge for laying of the photoelectric cable, which is not easy to view the direction of the photoelectric cable after being buried; in addition, although multiple photovoltaic panels are coded as a photovoltaic string in the design drawings, multiple photovoltaic strings are connected to an inverter, and multiple inverters are connected to a box transformer, after the construction is completed, due to the fact that the photoelectric cable is a hidden project, it is difficult to clearly know the boundary range of the photovoltaic panels at the string level and the inverter level in the actual site when the fault at the string level or the inverter level is monitored in the photovoltaic power station operation monitoring system, which brings inconvenience to construction and operation. SUMMARY

[0004] In view of the problems in the prior art that it is difficult to quickly locate the fault when the photovoltaic power station is fault located, only manual on-site troubleshooting can be performed, some hidden projects are not easy to view the direction of the photoelectric cable after being buried, and it is difficult to clearly know the boundary range of the photovoltaic panels at the string level and the inverter level in the actual site, the present application is proposed.

[0005] Therefore, the problem to be solved by the present application is how to provide a method for supporting an optimal view angle mode and a detailed view angle mode, realizing one-key quick switching between the two modes, avoiding the problems of time-consuming and laborious manual adjustment and easy errors, and enabling a staff to quickly view an object through the profile of an observed object marked in a video, analyze the cause of a fault, and realize positioning of a photoelectric cable route in a concealed project through secondary processing of a camera video stream.

[0006] To solve the above technical problems, the present application provides the following technical solutions.

[0007] In a first aspect, the present application provides a photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, which includes deploying cameras and monitoring a photovoltaic power station, collecting equipment data and camera data of the photovoltaic power station in real time, performing secondary processing on the camera data to obtain processed camera data, and performing fault analysis based on relevant data of the photovoltaic power station and the processed camera data to position a fault of the photovoltaic power station.

[0008] As a preferred scheme of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, the equipment data of the photovoltaic power station includes a photovoltaic power station power generation area equipment hardware installation state, a camera-to-control center machine room network state, a photovoltaic power station monitoring system state, and a device remote signaling signal; the camera data includes a camera number, a camera corner, a camera focal length, and a video picture; and the camera includes an optimal view angle mode and a detailed view angle mode.

[0009] As a preferred scheme of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, wherein: the monitoring of the photovoltaic power station includes devices with display area less than or equal to Q, devices with display area greater than Q and devices with display area as lines in the power generation area; the monitoring of the device with display area less than or equal to Q includes the following steps: in the best view angle mode, the device with display area less than or equal to Q in the camera in the power generation area is monitored by using the rectangular contour method, and the best view angle mode parameters set N1 is constructed; in the detail view angle mode, the focal length of the camera is adjusted by the program, so that the inverter is in the most clear state of details under the maximum focal length mode supported by the camera, and the detail view angle mode parameters set N2 is constructed.

[0010] As a preferred scheme of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, the equipment with a display area greater than Q is monitored in the following steps: in the best view angle mode, the equipment with a display area greater than Q in the power generation area is monitored by using the contour sketch method, for example, a photovoltaic string is taken as an example, a rectangular contour sketch method or a polygon contour sketch method is used, a camera with the best view angle is selected near the photovoltaic string, the view angle of the camera is adjusted by using a program, if the photovoltaic string is in the best position in the video provided by the camera, the angle and the focal length of the camera are recorded when the camera rotates from the default starting position after being powered on to the current position, and a photo in the best view angle mode is saved; based on the photo in the best view angle mode, a minimum contour rectangle or a minimum contour polygon of the contour of the photovoltaic string is obtained by using pixel analysis, the starting point coordinates and the length information of each line segment of the rectangle or the polygon are recorded, and a photovoltaic string best view angle mode parameter set Z1 is constructed; in the detail view angle mode, the focal length of the camera is adjusted by using a program, the photovoltaic string is in the most clear state in the maximum focal length mode supported by the camera, a photo in the detail view angle mode is saved; based on the photo in the detail view angle mode, a minimum contour rectangle or a minimum contour polygon of the contour of the photovoltaic string is obtained by using pixel analysis, the starting point coordinates and the length information of each line segment of the rectangle or the polygon are recorded, and a photovoltaic string detail mode parameter set Z2 is constructed.

[0011] As a preferred scheme of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, the equipment with a display area greater than Q is monitored in the following steps: in the best view angle mode, the equipment with a display area greater than Q in the power generation area is monitored by using the contour sketch method, for example, a photovoltaic string is taken as an example, a rectangular contour sketch method or a polygon contour sketch method is used, a camera with the best view angle is selected near the photovoltaic string, the view angle of the camera is adjusted by using a program, if the photovoltaic string is in the best position in the video provided by the camera, the angle and the focal length of the camera are recorded when the camera rotates from the default starting position after being powered on to the current position, and a photo in the best view angle mode is saved; based on the photo in the best view angle mode, a minimum contour rectangle or a minimum contour polygon of the contour of the photovoltaic string is obtained by using pixel analysis, the starting point coordinates and the length information of each line segment of the rectangle or the polygon are recorded, and a photovoltaic string best view angle mode parameter set Z1 is constructed; in the detail view angle mode, the focal length of the camera is adjusted by using a program, the photovoltaic string is in the most clear state in the maximum focal length mode supported by the camera, a photo in the detail view angle mode is saved; based on the photo in the detail view angle mode, a minimum contour rectangle or a minimum contour polygon of the contour of the photovoltaic string is obtained by using pixel analysis, the starting point coordinates and the length information of each line segment of the rectangle or the polygon are recorded, and a photovoltaic string detail mode parameter set Z2 is constructed.

[0012] As a preferred scheme of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, wherein: the fault analysis includes device abnormalities in a display area less than or equal to Q, device abnormalities in a display area greater than Q, device abnormalities in a display area of a line, and data uploading; the device abnormalities in the display area less than or equal to Q include the following steps: analyzing the devices in the display area less than or equal to Q using multi-source data, taking the inverter as an example, if the inverter internal temperature obtained through the remote signaling signal is greater than the first threshold value, the system issues an alarm, and according to the inverter best view mode parameter set N1, it is connected to the corresponding camera and the camera is adjusted; pictures are taken from the camera video picture at a frame rate not less than M, a rectangle is drawn on the taken pictures according to the inverter best view mode parameter set N1, and a video with inverter contour is obtained; if it is necessary to view the detailed information of the inverter through the camera, according to the inverter detail mode parameter set N2, it is connected to the corresponding camera and the camera is adjusted; pictures are taken from the camera video picture at a frame rate not less than M, a rectangle is drawn on the taken pictures according to the inverter detail mode parameter set N2, and a video with inverter contour is obtained; the device abnormalities in the display area greater than Q include the following steps: analyzing the devices in the display area greater than Q using multi-source data, taking the photovoltaic string as an example, if the voltage or current of the photovoltaic string is less than the second threshold value or the third threshold value, it is determined that the photovoltaic string is faulty, and according to the photovoltaic string best view mode parameter set Z1, it is connected to the corresponding camera and the camera is adjusted; pictures are taken from the camera video picture at a frame rate not less than M, a line segment is drawn in the taken pictures according to the photovoltaic string best view mode parameter set Z1, and the line segment is displayed in the program video display area after drawing is completed; if it is necessary to remotely survey the photovoltaic string, according to the photovoltaic string detail mode parameter set Z2, it is connected to the corresponding camera and the camera is adjusted; pictures are taken from the camera video picture at a frame rate not less than M, a line segment is drawn in the taken pictures according to the photovoltaic string detail mode parameter set Z2, and the line segment is displayed in the program video display area after drawing is completed.

[0013] As a preferred scheme of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, wherein: the device anomaly of the display area being a line includes the following steps: analyzing the device with the display area being a line using multi-source data, taking the optical fiber as an example, if the link of the optical fiber fails, connecting to the corresponding camera according to the optical cable optimal view mode parameter set G1 and adjusting the camera; taking pictures from the camera video picture at a frame rate not less than M, drawing a line segment in the taken picture according to the optical cable optimal view mode parameter set G1, and displaying in the program video display area after drawing is completed; if it is necessary to remotely view the wiring details of the optical fiber, connecting to the corresponding camera according to the optical cable detail view mode parameter set G2 and adjusting the camera; taking pictures from the camera video picture at a frame rate not less than M, drawing a line segment in the taken picture according to the optical cable detail view mode parameter set G2, and displaying in the program video display area after drawing is completed, and observing whether the cable slot box through by the optical fiber or the wire is damaged in the camera, wherein if the optical fiber is buried underground or put into the cable slot box after executing the step of monitoring the device with the display area being a line, the trend of the optical fiber in the concealment project is marked in the video; the data uploading is analyzing the uploaded data using multi-source data, if there is no data uploading of the monitoring device in the continuous T acquisition time periods, it is determined that the corresponding device or the link between the corresponding device and the upper device fails; the no data uploading in the continuous T acquisition time periods includes: when there is no current or voltage data uploading of a certain photovoltaic string in the continuous T acquisition time periods, it is determined whether the data of other photovoltaic strings under the same inverter as the corresponding photovoltaic string can be correctly uploaded, if the data can be correctly uploaded, it is determined that the fault range is the photovoltaic panel under the corresponding photovoltaic string or the link between the corresponding photovoltaic string and the inverter, the device anomaly step of the display area being greater than Q is executed, the outline of the corresponding photovoltaic string is positioned, and the device anomaly step of the display area being a line is executed, and the link trend of the photovoltaic string to the inverter is positioned in the video; when there is no running state data uploading of a certain inverter in the continuous T acquisition time periods, it is determined whether the other inverters connected to the same transformer of the corresponding inverter can upload the running state data in real time, if the inverters can upload the running state data in real time, it is determined that the fault range is the corresponding inverter or the link between the corresponding inverter and the transformer, the device anomaly step of the display area being less than or equal to Q is executed, the outline of the corresponding inverter is positioned, and the device anomaly step of the display area being a line is executed, and the link trend of the inverter to the transformer is positioned in the video.

[0014] In a second aspect, to further solve the safety problem in the fault positioning of the photovoltaic power station, the embodiment provides a photovoltaic power station fault positioning system based on multi-camera and multi-source data fusion, which comprises: a data acquisition module, which is used to acquire equipment data of the photovoltaic power station and make a judgment, deploy cameras and acquire video pictures by using an optimal view angle mode and a detail view angle mode, and construct a parameter set; a secondary processing module, which is used to intercept pictures from the video pictures of the cameras according to a frame rate and draw in a specified area at the same frequency to obtain a video with an outline of an observed object; and a fault positioning module, which is used to analyze a fault according to running data and state information of the observed object and a topological relationship between the photovoltaic power station equipment, locate a fault position by using the video with the outline of the observed object, and mark a cable direction related to the fault in the video.

[0015] In a third aspect, the embodiment of the present application provides a computer device, which comprises a memory and a processor, and the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion is realized.

[0016] In a fourth aspect, the embodiment of the present application provides a computer readable storage medium, which stores a computer program, wherein: when the computer program is executed by the processor, any step of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion is realized.

[0017] The present application has the following beneficial effects: the present application provides two view angle modes, in the optimal view angle mode, the observed object and the surrounding environment are quickly located by the video, in the detail view angle mode, the details of the observed object are maximally understood by the video, and the efficiency of fault positioning is improved; by performing secondary processing on the camera video stream, pictures are intercepted from the camera video according to a frequency, the observed object is marked in the intercepted pictures, then the pictures are drawn in a specified area at the same frequency, according to the principle of visual persistence, a continuous video is obtained, a worker can quickly view the object by the outline of the observed object marked in the video, and analyze the cause of the fault; by recording the working parameters of the camera in the optimal view angle mode and the detail view angle mode by the program, one-key fast switching between the two modes is realized, the problem that manual adjustment is time-consuming and laborious and is prone to errors is avoided; in combination with the running data and state information of the observed object acquired by the photovoltaic power station monitoring system and the topological relationship between the photovoltaic power station equipment, the cause of the fault is analyzed, the photovoltaic cable direction related to the fault is marked in the video, and the photovoltaic cable direction in the concealed engineering is positioned. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings. Among them:

[0019] Figure 1 The flow chart for collecting the equipment data and the camera data of the photovoltaic power station in Embodiment 1.

[0020] Figure 2 The flow chart for locating the fault of the photovoltaic power station in Embodiment 1. DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings.

[0022] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the concept of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0023] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0024] Embodiment 1

[0025] Reference Figure 1 and Figure 2 , the first embodiment of the present application provides a photovoltaic power station fault locating method based on multi-camera and multi-source data fusion, including the following steps:

[0026] S1: deploying a camera and monitoring the photovoltaic power station to collect the equipment data and the camera data of the photovoltaic power station in real time.

[0027] Preferably, the equipment data of the photovoltaic power station includes the installation state of the photovoltaic power station power generation area equipment hardware, the network state of the camera to the control center machine room, the photovoltaic power station monitoring system state and the equipment remote signaling signal.

[0028] Specifically, the equipment remote signaling signal includes current, voltage, circuit breaker state, power and equipment temperature.

[0029] Preferably, the camera data includes camera number, camera rotation angle, camera focal length and video picture.

[0030] Preferably, as Figure 1 The collection process of the equipment data and the camera data of the photovoltaic power station is shown in the figure, and the deployment of the camera includes the following steps: judging the equipment hardware installation state of the photovoltaic power station power generation area, the network state of the camera to the control center machine room and the photovoltaic power station monitoring system state, if the equipment hardware installation state of the photovoltaic power station power generation area is normal, the network state of the camera to the control center machine room is smooth and the photovoltaic power station monitoring system state is normal, then the camera in the photovoltaic power station power generation area is numbered.

[0031] The equipment or component that needs to be analyzed and positioned is uniformly identified, and the communication link topology relationship of the string to the center machine room is recorded.

[0032] Specifically, the photovoltaic power station monitoring system state is that the photovoltaic power station monitoring system can monitor the real-time running data of the string, the inverter and the transformer equipment.

[0033] Further, the real-time running data includes current, voltage, power, transformer oil temperature and circuit breaker on-off state.

[0034] Specifically, by deploying multiple cameras supporting rotation and zoom, panoramic monitoring of the photovoltaic power station power generation area is realized, and then the real-time state data and video data of the equipment are quickly obtained, providing a data basis for fault analysis and positioning.

[0035] Further, the equipment or component that needs to be analyzed and positioned includes photovoltaic string, inverter, transformer, communication controller and communication cable.

[0036] Specifically, the topology relationship of the communication link includes: multiple photovoltaic panels form a photovoltaic string.

[0037] Multiple photovoltaic strings are connected to an inverter through a cable.

[0038] Multiple inverters are connected to a transformer.

[0039] One to multiple transformers are divided into a square matrix.

[0040] One to multiple square matrices form a photovoltaic power station.

[0041] Specifically, by uniformly identifying the equipment that needs to be monitored, recording the communication link topology relationship between components, it is helpful to quickly lock the fault range when a fault occurs, and improve the efficiency of fault positioning.

[0042] Preferably, the camera includes the best view mode and the detail view mode.

[0043] Specifically, in the best view mode, the observed object and the surrounding environment are observed by video positioning, and then the on-site personnel are guided to the designated position when a fault occurs.

[0044] Specifically, in the detail view mode, the details of the observed object are maximally understood by video, and the number of on-site personnel going to the site is reduced.

[0045] Preferably, the monitoring of the photovoltaic power station includes equipment with a display area less than or equal to Q in the power generation area, equipment with a display area greater than Q, and equipment with a display area as a line.

[0046] Specifically, monitoring the equipment with a display area less than or equal to Q includes the following steps: in the best view mode, the equipment with a display area less than or equal to Q in the power generation area is monitored by using a rectangular contour method, taking an inverter as an example, a camera with the best view position near the inverter is selected, the camera view angle is adjusted by a program, if the inverter is in the best position in the video provided by the camera, that is, the inverter and the surrounding reference equipment of the inverter are included in the video, the angle and focal length of the camera from the default starting position after power-on to the current position are recorded, and the photo in the best view mode is saved.

[0047] Based on the photo in the best view mode, the smallest contour rectangle of the inverter contour is obtained by pixel analysis, the starting point coordinates, length and width information of the smallest contour rectangle are recorded, and the inverter best view mode parameter set N1 is constructed.

[0048] In the detail view mode, the focal length of the camera is adjusted by a program, so that the inverter is in the most clear state under the maximum focal length mode supported by the camera, and the photo in the detail view mode is saved.

[0049] Based on the photo in the detail view mode, the smallest contour rectangle of the inverter contour is obtained by pixel analysis, the starting point coordinates, length and width information of the smallest contour rectangle are recorded, and the inverter detail mode parameter set N2 is constructed.

[0050] Further, the surrounding reference equipment of the inverter includes a cable slot box, a photovoltaic panel support and adjacent photovoltaic string.

[0051] Further, the inverter best view mode parameter set N1 includes camera number, camera rotation angle, camera focal length, starting point coordinates, length and width information of the smallest contour rectangle of the inverter in the best view mode.

[0052] Further, the inverter detail mode parameter set N2 includes camera number, camera rotation angle, camera focal length, starting point coordinates, length and width information of the smallest contour rectangle of the inverter in the detail view mode.

[0053] Specifically, the device with a display area greater than Q is monitored in the best view angle mode by using the contour sketch method. For example, a photovoltaic string is composed of multiple photovoltaic panels in series, so the contour of the photovoltaic string can be a rectangle or an irregular polygon. The rectangular contour sketch method or the polygon contour sketch method is adopted, and the best view angle camera is selected near the photovoltaic string. The camera view angle is adjusted by a program. If the photovoltaic string is in the best position in the video provided by the camera, that is, the video includes the photovoltaic string and the surrounding reference object, the angle and focal length of the camera from the default starting position after power-on to the current position are recorded, and the photo in the best view angle mode is saved.

[0054] Based on the photo in the best view angle mode, the minimum contour rectangle or the minimum contour polygon of the photovoltaic string contour is obtained by pixel analysis, the starting point coordinates and length information of each line segment of the rectangle or polygon are recorded, and the photovoltaic string best view angle mode parameter set Z1 is constructed.

[0055] In the detail view angle mode, the focal length of the camera is adjusted by a program to make the photovoltaic string in the most clear state under the maximum focal length mode supported by the camera, and the photo in the detail view angle mode is saved.

[0056] Based on the photo in the detail view angle mode, the minimum contour rectangle or the minimum contour polygon of the photovoltaic string contour is obtained by pixel analysis, the starting point coordinates and length information of each line segment of the rectangle or polygon are recorded, and the photovoltaic string detail mode parameter set Z2 is constructed.

[0057] Further, the surrounding reference object of the photovoltaic string includes a road, a street lamp, a meteorological instrument, a sky imager, a building, an inverter in a photovoltaic power station, a box transformer, and a photovoltaic support.

[0058] Further, the photovoltaic string best view angle mode parameter set Z1 includes the camera number, the camera rotation angle, the camera focal length, and the starting and ending point coordinate information of each line segment of the minimum contour rectangle or the minimum contour polygon of the photovoltaic string in the best view angle mode.

[0059] Further, the photovoltaic string detail mode parameter set Z2 includes the camera number, the camera rotation angle, the camera focal length, and the starting and ending point coordinate information of each line segment of the minimum contour rectangle or the minimum contour polygon of the photovoltaic string in the detail view angle mode.

[0060] Specifically, the monitoring of the device with the display area as a line includes the following steps: in the best view angle mode, the device with the display area as a line in the power generation area is monitored in the camera, for example, the optical fiber is composed of one or more lines, so the line segment representation is used to represent the optical fiber, the best view angle camera is selected near the route of the optical fiber, the camera view angle is adjusted by the program, if the optical fiber is in the best position of the camera video, that is, the optical fiber and the surrounding reference device are included in the video, the angle and focal length of the camera from the default starting position after power-on to the current position are recorded, and the photo in the best view angle mode is saved.

[0061] Based on the photo in the best view angle mode, the pixel analysis is used to obtain the direction of each segment of the optical fiber, the camera number of the optical fiber direction and the start and end point coordinates of each line segment of the optical fiber direction are recorded, and the optical cable best view angle mode parameter set G1 is constructed.

[0062] In the detail view angle mode, the focal length of the camera is adjusted by the program, so that the optical fiber is in the most clear state of the maximum focal length mode supported by the camera, and the photo in the detail view angle mode is saved.

[0063] Based on the photo in the detail view angle mode, the pixel analysis is used to obtain the line segment of the optical fiber contour, the start and end point coordinates of each line segment of the optical fiber direction are recorded, and the optical cable detail view angle mode parameter set G2 is constructed, wherein if the corresponding device is a concealed project, the monitoring needs to be carried out before the corresponding concealed project is completed.

[0064] Further, the surrounding reference device of the optical fiber includes the cable slot box, the pipe trench and the photovoltaic panel support where the optical fiber or cable will be placed.

[0065] Further, the optical cable best view angle mode parameter set G1 includes the camera number, the camera rotation angle, the camera focal length and the start and end point coordinates of each line segment of the optical fiber in the best view angle mode.

[0066] Further, the optical cable detail view angle mode parameter set G2 includes the camera number, the camera rotation angle, the camera focal length and the start and end point coordinates of each line segment of the optical fiber in the detail view angle mode.

[0067] Specifically, the area threshold Q is obtained by calculating the area threshold according to the type of the photovoltaic power station device, the camera data and the observation distance, and adjusting the actual area size of the known model device.

[0068] Specifically, by adopting different monitoring methods for devices with display areas less than or equal to Q, devices with display areas greater than Q, and devices with display areas in the form of lines, the shape of the device is accurately captured, and the camera data under the best viewing angle and the detailed viewing angle is recorded, which facilitates subsequent rapid positioning and detailed inspection of the device.

[0069] S2: performing secondary processing on the camera data to obtain processed camera data.

[0070] Preferably, the secondary processing is based on the principle of perspective persistence, and pictures are extracted from the camera video according to a frame rate M, the contour of the observed object is drawn on the pictures according to a parameter set, and after the drawing is completed, the drawing is performed to the display area of the program according to the frame rate M, to obtain a video picture showing the contour of the observed object.

[0071] Specifically, for the frame rate M, pictures are extracted from the camera video based on the principle of perspective persistence, and the frame rate is not less than 25 fps, to ensure the coherence and smoothness of the video, and the frame rate is adjusted according to the requirements of the actual application scene and the operation ability of the image processing algorithm.

[0072] Specifically, by extracting pictures from the video and drawing the contour of the observed object based on the principle of perspective persistence, a video picture showing the contour of the device is generated, providing an intuitive basis for fault analysis.

[0073] S3: based on the related data of the photovoltaic power station and the processed camera data, performing fault analysis to locate the fault of the photovoltaic power station.

[0074] Preferably, the fault analysis includes device abnormalities with display areas less than or equal to Q, device abnormalities with display areas greater than Q, device abnormalities with display areas in the form of lines, and data uploading.

[0075] Specifically, as shown in Figure 2 the flow of locating the fault of the photovoltaic power station, the device abnormalities with display areas less than or equal to Q include the following steps: analyzing the devices with display areas less than or equal to Q by using multi-source data, taking the inverter as an example, if the inverter internal temperature obtained through the remote signaling signal is greater than the first threshold value, the system issues an alarm, the camera number connected to the corresponding camera according to the inverter best viewing angle mode parameter set N1 is connected to the corresponding camera, and the camera is adjusted according to the camera angle and the camera focal length of the inverter best viewing angle mode parameter set N1.

[0076] According to the minimum outline rectangle of the inverter best view angle mode parameter set N1, the starting point coordinates, length and width information are drawn on the intercepted picture, and the rectangle is displayed on the program video display area immediately after drawing is completed, and the video with the inverter outline is obtained, and then the inverter and the position of the reference device around the inverter are quickly located.

[0077] If the details of the inverter need to be viewed through the camera, the corresponding camera is connected according to the camera number of the inverter detail mode parameter set N2, and the camera is adjusted according to the camera rotation angle and the camera focal length of the inverter detail mode parameter set N2.

[0078] According to the minimum outline rectangle of the inverter detail mode parameter set N2, the starting point coordinates, length and width information are drawn on the intercepted picture, and the rectangle is displayed on the program video display area immediately after drawing is completed, and the video with the inverter outline is obtained, and then the details of the inverter are viewed.

[0079] Specifically, the device abnormality of the display area greater than Q includes the following steps: using multi-source data to analyze the device with the display area greater than Q, taking a photovoltaic string as an example, if the voltage or current of the photovoltaic string is less than half of the historical voltage or current under the same weather, it is determined that the photovoltaic string is faulty, that is, the photovoltaic panel is damaged, blocked or contaminated, the corresponding camera is connected according to the camera number of the photovoltaic string best view angle mode parameter set Z1, and the camera is adjusted according to the camera rotation angle and the camera focal length of the photovoltaic string best view angle mode parameter set Z1.

[0080] According to the minimum outline rectangle of the inverter best view angle mode parameter set N1, the starting point coordinates, length and width information are drawn on the intercepted picture, and the rectangle is displayed on the program video display area immediately after drawing is completed, and the video with the inverter outline is obtained, and then the inverter and the position of the reference device around the inverter are quickly located.

[0081] If the photovoltaic string needs to be remotely surveyed, the corresponding camera is connected according to the camera number of the photovoltaic string detail mode parameter set Z2, and the camera is adjusted according to the camera rotation angle and the camera focal length of the photovoltaic string detail mode parameter set Z2.

[0082] According to the frame rate of not less than 25fps, the picture is intercepted from the camera video picture, the start and end point coordinate information of each line segment of the minimum outline rectangle or the minimum outline polygon of the photovoltaic string detail mode parameter set Z2 is drawn in the intercepted picture, and the line segment is displayed in the program video display area immediately after drawing is completed, and then the remote reconnaissance of the photovoltaic string is realized in the camera.

[0083] Specifically, the device abnormality of the display area as a line includes the following steps: analyzing the display area as a line device by using multi-source data, taking the optical fiber as an example, if the link of the optical fiber fails, connecting to the corresponding camera according to the camera number of the optical cable best view angle mode parameter set G1, and adjusting the camera according to the camera corner and camera focal length of the optical cable best view angle mode parameter set G1.

[0084] According to the frame rate of not less than 25fps, the picture is intercepted from the camera video picture, the start and end point coordinate information of each line segment of the minimum outline rectangle or the minimum outline polygon of the photovoltaic string detail mode parameter set Z2 is drawn in the intercepted picture, and the line segment is displayed in the program video display area immediately after drawing is completed, and then the remote reconnaissance of the photovoltaic string is realized in the camera.

[0085] If it is necessary to remotely view the wiring details of the optical fiber, connect to the corresponding camera according to the camera number of the optical cable detail view angle mode parameter set G2, and adjust the camera according to the camera corner and camera focal length of the optical cable detail view angle mode parameter set G2.

[0086] According to the frame rate of not less than 25fps, the picture is intercepted from the camera video picture, the start and end point coordinate information of each line segment of the minimum outline rectangle or the minimum outline polygon of the photovoltaic string detail mode parameter set Z2 is drawn in the intercepted picture, and the line segment is displayed in the program video display area immediately after drawing is completed, and then the remote reconnaissance of the photovoltaic string is realized in the camera.

[0087] Specifically, the data upload is analyzed by using multi-source data, if the monitoring device has no data upload in two continuous collection time periods, it is determined that the corresponding device or the link between the corresponding device and the upper device fails, and the fault analysis is performed according to the priority of parent-child.

[0088] Further, the absence of data uploading in two consecutive acquisition time periods includes: when the current or voltage data of a certain photovoltaic string is not uploaded in two consecutive acquisition time periods, it is judged whether the data of other photovoltaic strings under the same inverter as the corresponding photovoltaic string can be correctly uploaded, if the data can be correctly uploaded, it is determined that the fault range is the photovoltaic panel under the corresponding photovoltaic string or the link between the corresponding photovoltaic string and the inverter, the device abnormality step of displaying an area greater than Q is executed, the outline of the corresponding photovoltaic string is located, and the device abnormality step of displaying a line is executed, and the link direction of the photovoltaic string to the inverter is located in the video.

[0089] When the running state data of a certain inverter is not uploaded in two consecutive acquisition time periods, it is judged whether the running state data of other inverters connected to the same transformer can be uploaded in real time, if the running state data can be uploaded in real time, it is determined that the fault range is the corresponding inverter or the link between the corresponding inverter and the transformer, the device abnormality step of displaying an area less than or equal to Q is executed, the outline of the corresponding inverter is located, and the device abnormality step of displaying a line is executed, and the link direction of the inverter to the transformer is located in the video.

[0090] Specifically, for the first threshold value for judging whether the inverter is overheated, the upper limit of the safe operating temperature range specified by the manufacturer is obtained according to the inverter product manual to obtain an initial first threshold value, and the relationship between the inverter internal temperature and the fault occurrence rate is analyzed combined with the actual operation data to adjust the threshold value to obtain a reasonable temperature threshold value as the first threshold value.

[0091] Specifically, by comprehensively analyzing the device data and the processed camera data, the fault type and the approximate location are quickly judged, and different analysis methods are adopted according to different fault types, the parameters recorded from multiple perspectives are fully utilized, the outline of the fault device and the link direction of the fault are accurately marked in the video, important clues are provided for on-site troubleshooting, and precise fault positioning is realized.

[0092] The embodiment also provides a photovoltaic power station fault positioning system based on multi-camera and multi-source data fusion, comprising: a data acquisition module for acquiring device data of a photovoltaic power station and performing judgment, deploying cameras and collecting video pictures by using a best perspective mode and a detail perspective mode, and constructing a parameter set; a secondary processing module for intercepting pictures from the video pictures of the cameras according to a frame rate and drawing in a specified area at the same frequency to obtain a video with an outline of an observed object; and a fault positioning module for performing fault analysis according to running data and state information of the observed object and a topological relationship between devices of the photovoltaic power station, locating a fault position by using the video with the outline of the observed object, and marking a cable direction related to the fault in the video.

[0093] The embodiment also provides a computer device suitable for the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion proposed in the above embodiment.

[0094] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0095] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to realize the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion proposed in the above embodiment. The storage medium can be realized by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0096] In summary, the application provides two viewing angle modes, in the best viewing angle mode, the observed object and the surrounding environment are quickly located through the video, and in the detail viewing angle mode, the details of the observed object are maximally understood through the video, so that the efficiency of fault positioning is improved; through secondary processing of the camera video stream, pictures are intercepted from the camera video according to the frequency, and the observed object is marked in the intercepted pictures, then the pictures are drawn in the specified area at the same frequency, according to the principle of visual persistence, a continuous video is obtained, and the staff can quickly view the object through the outline of the observed object marked in the video to analyze the cause of the fault; the working parameters of the camera in the best viewing angle mode and the detail viewing angle mode are recorded through the program, one-key fast switching between the two modes is realized, and the problem that manual adjustment is time-consuming and laborious and prone to errors is avoided; the running data and state information of the observed object collected by the photovoltaic power station monitoring system and the topological relationship between the photovoltaic power station equipment are combined, the cause of the fault is analyzed, and the photovoltaic cable direction related to the fault is marked in the video, so that the photovoltaic cable direction in the concealed project is positioned.

[0097] Example 2

[0098] Referring to Tables 1 and 2, for the second embodiment of the application, which is different from the first embodiment, in order to verify its beneficial effects, experimental comparison data of the application and the prior art are provided.

[0099] This example simulates a certain photovoltaic power station, including 1 power generation area, 1 box transformer is installed in each area, each box transformer is connected with 20 inverters, each inverter is connected with 28 photovoltaic strings, 24 high-definition cameras supporting rotation and zoom are deployed in each power generation area to realize panoramic monitoring of the power generation area; since the photovoltaic string is composed of multiple photovoltaic panels, the area of a single photovoltaic panel is more than 2m 2 , the number of photovoltaic panels of a photovoltaic string is usually between 13 and 26, so the area of a photovoltaic string is usually between 26m 2 and 53m 2 , and the threshold Q in this example is calculated to be 10m 2 .

[0100] For the above threshold Q, this example adopts different monitoring methods for inverters, photovoltaic strings and optical fibers, and the camera data collected is shown in Table 1, from which it can be seen that for the inverters with an area less than the threshold Q, the rectangular contour method is used to record the camera parameters and the contour information of the inverter in the best viewing angle and the detail viewing angle, for the photovoltaic strings with an area greater than the threshold Q, the polygon contour method is used to record the camera data and the coordinate information of each edge in the polygon, and for the line-shaped optical fibers, the line segment method is used to record all the line segment coordinates of the optical fiber direction.

[0101] Table 1 best view mode and detail view mode data table

[0102]

[0103] As can be seen from the above table, by using the best view mode and the detail view mode to collect camera data and device data respectively, the application can maximize the understanding of the details of the observed object, reduce the number of times that the staff goes to the site, and realize fast switching between the two modes, thereby reducing the error rate and reducing resource consumption; for the camera data obtained from the above table, by using a single camera monitoring system and the application in the same environment, the comparison is shown in Table 2.

[0104] Table 2 comparison data table of the application and the prior art

[0105]

[0106] In summary, by comparing the single camera monitoring system with the application, it is shown that the application can more comprehensively monitor the equipment, thereby providing more accurate device data, effectively improving the accuracy of fault positioning, and by recording the parameters of the best view mode and the detail view mode, fast switching between the two modes is realized, and the calling efficiency of the camera is improved.

[0107] It should be noted that the above embodiments are only used to illustrate the technical solutions of the application and are not limiting, and although the application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the application, and they should be covered in the scope of the claims of the application.

Claims

1. A photovoltaic power station fault location method based on multi-camera and multi-source data fusion, characterized in that: The method comprises the following steps: deploying a camera and monitoring a photovoltaic power station to collect equipment data and camera data of the photovoltaic power station in real time; performing secondary processing on the camera data to obtain processed camera data; performing fault analysis based on the related data of the photovoltaic power station and the processed camera data to locate the fault of the photovoltaic power station; the camera comprises an optimal view angle mode and a detailed view angle mode; the monitoring of the photovoltaic power station comprises equipment with a display area less than or equal to Q, equipment with a display area greater than Q, and equipment with a display area in the form of lines in the power generation area, wherein the equipment with a display area less than or equal to Q is an inverter, the equipment with a display area greater than Q is a photovoltaic string, and the equipment with a display area in the form of lines is an optical fiber; the fault analysis comprises equipment anomaly with a display area less than or equal to Q, equipment anomaly with a display area greater than Q, equipment anomaly with a display area in the form of lines, and data uploading; the monitoring of the equipment with a display area in the form of lines comprises the following steps: based on a photo in the optimal view angle mode, the pixel analysis is used to obtain the direction of each segment of the optical fiber, the camera number constituting the direction of the optical fiber and the start and end point coordinates of each line segment constituting the direction of the optical fiber are recorded, and an optimal view angle mode parameter set G1 of the optical cable is constructed; based on a photo in the detailed view angle mode, the pixel analysis is used to obtain the line segment of the outline of the optical fiber, the start and end point coordinates of each line segment constituting the direction of the optical fiber are recorded, and a detailed view angle mode parameter set G2 of the optical cable is constructed, wherein if the corresponding equipment is a concealed project, the monitoring is performed before the corresponding concealed project is completed; the equipment anomaly with a display area in the form of lines comprises the following steps: the equipment with a display area in the form of lines is analyzed by using multi-source data, if the link fault of the optical fiber occurs, the corresponding camera is connected according to the optimal view angle mode parameter set G1 of the optical cable, and the camera is adjusted; pictures are captured from the camera video screen at a frame rate not less than M, the line segment is drawn in the captured pictures according to the optimal view angle mode parameter set G1 of the optical cable, and the drawing is displayed in the program video display area after the drawing is completed; if it is necessary to remotely view the wiring details of the optical fiber, the corresponding camera is connected according to the detailed view angle mode parameter set G2 of the optical cable, and the camera is adjusted; pictures are captured from the camera video screen at a frame rate not less than M, the line segment is drawn in the captured pictures according to the detailed view angle mode parameter set G2 of the optical cable, and the drawing is displayed in the program video display area after the drawing is completed, and whether the cable slot box through by the optical fiber or the wire is damaged is observed in the camera, wherein if the optical fiber is buried underground or put into the cable slot box after the step of monitoring the equipment with a display area in the form of lines is performed, the direction of the optical fiber in the concealed project is marked in the video; the data uploading is analyzed by using multi-source data, if there is no data uploading of the monitoring equipment in a continuous T acquisition time period, it is determined that the corresponding equipment or the link fault between the corresponding equipment and the upper equipment occurs; there is no data uploading in the continuous T acquisition time period comprises: When the current or voltage data of a certain photovoltaic string fails to be uploaded for continuous T acquisition time periods, it is determined whether the data of other photovoltaic strings under the same inverter as the corresponding photovoltaic string can be correctly uploaded, if the data can be correctly uploaded, it is determined that the fault range is the photovoltaic panel under the corresponding photovoltaic string or the link between the corresponding photovoltaic string and the inverter, the device abnormality step of displaying an area greater than Q is performed, the profile of the corresponding photovoltaic string is located, and the device abnormality step of displaying a line is performed to locate the link direction of the photovoltaic string to the inverter in the video; When the running state data of a certain inverter fails to be uploaded for continuous T acquisition time periods, it is determined whether the running state data of other inverters connected to the same transformer can be uploaded in real time, if the running state data can be uploaded in real time, it is determined that the fault range is the corresponding inverter or the link between the corresponding inverter and the transformer, the device abnormality step of displaying an area less than or equal to Q is performed, the profile of the corresponding inverter is located, and the device abnormality step of displaying a line is performed to locate the link direction of the inverter to the transformer in the video.

2. The method for photovoltaic power station fault location based on multi-camera and multi-source data fusion according to claim 1, characterized in that: The device data of the photovoltaic power station includes a photovoltaic power station power generation area device hardware installation state, a camera to control center machine room network state, a photovoltaic power station monitoring system state, and a device remote signaling signal. The camera data includes a camera number, a camera corner, a camera focal length, and a video picture. 3.The photovoltaic power station fault location method based on multi-camera and multi-source data fusion of claim 2, characterized in that: The device with the display area less than or equal to Q is monitored by the following steps: In the best view angle mode, the device with the display area less than or equal to Q in the camera in the power generation area is monitored by using the rectangular profile method, the best view angle position camera is selected in the camera near the inverter, the camera view angle is adjusted by the program, if the inverter is in the best position in the video picture provided by the camera, the angle and focal length of the camera rotated from the default starting position after power-on to the current position are recorded, and the photo in the best view angle mode is saved; Based on the photo in the best view angle mode, the minimum profile rectangle of the inverter profile is obtained by pixel analysis, the starting point coordinates, length and width information of the minimum profile rectangle are recorded, and the inverter best view angle mode parameter set N1 is constructed; In the detail view angle mode, the focal length of the camera is adjusted by the program, so that the inverter is in the most clear state in the maximum focal length mode supported by the camera, and the photo in the detail view angle mode is saved; Based on the photo in the detail view angle mode, the minimum profile rectangle of the inverter profile is obtained by pixel analysis, the starting point coordinates, length and width information of the minimum profile rectangle are recorded, and the inverter detail mode parameter set N2 is constructed. 4.The photovoltaic power station fault location method based on multi-camera and multi-source data fusion of claim 3, characterized in that: The device with the display area greater than Q is monitored by the following steps: In the best view angle mode, the devices in the power generation area whose display area in the camera is greater than Q are monitored by using the outline sketch method, the rectangular outline sketch method or the polygonal outline sketch method is adopted, the best view angle camera is selected near the photovoltaic string in the camera, the camera view angle is adjusted by the program, if the photovoltaic string is in the best position in the video provided by the camera, the angle and the focal length of the camera rotated from the default starting position after being powered on to the current position are recorded, and the photo in the best view angle mode is saved; Based on the photo in the best view angle mode, the minimum outline rectangle or the minimum outline polygon of the photovoltaic string contour is obtained by using pixel analysis, the starting point coordinates and the length information of each line segment of the rectangle or the polygon are recorded, and the photovoltaic string best view angle mode parameter set Z1 is constructed; In the detail view angle mode, the focal length of the camera is adjusted by the program, so that the photovoltaic string is in the most clear state under the maximum focal length mode supported by the camera, and the photo in the detail view angle mode is saved; Based on the photo in the detail view angle mode, the minimum outline rectangle or the minimum outline polygon of the photovoltaic string contour is obtained by using pixel analysis, the starting point coordinates and the length information of each line segment of the rectangle or the polygon are recorded, and the photovoltaic string detail mode parameter set Z2 is constructed.

5. The method for photovoltaic power station fault location based on multi-camera and multi-source data fusion according to claim 4, characterized in that: The device whose display area is a line in the monitoring area further includes the following steps: In the best view angle mode, the devices in the power generation area whose display area in the camera is greater than Q are monitored, the best view angle camera is selected near the fiber route in the camera, the camera view angle is adjusted by the program, if the fiber is in the best position in the video provided by the camera, the angle and the focal length of the camera rotated from the default starting position after being powered on to the current position are recorded, and the photo in the best view angle mode is saved; In the detail view angle mode, the focal length of the camera is adjusted by the program, so that the fiber is in the most clear state under the maximum focal length mode supported by the camera, and the photo in the detail view angle mode is saved. 6.The photovoltaic power station fault location method based on multi-camera and multi-source data fusion of claim 5, characterized in that: The device whose display area is less than or equal to Q includes the following steps: Multi-source data is used to analyze the device whose display area is less than or equal to Q, if the inverter internal temperature obtained through the remote signaling signal is greater than the first threshold value, the system issues an alarm, and the corresponding camera is connected according to the inverter best view angle mode parameter set N1 and the camera is adjusted; Pictures are intercepted from the camera video according to a frame rate not lower than M, a rectangle is drawn on the intercepted pictures according to the inverter best view angle mode parameter set N1, and a video with the inverter contour is obtained; If the details of the inverter need to be viewed through the camera, the corresponding camera is connected according to the inverter detail mode parameter set N2 and the camera is adjusted; Pictures are intercepted from the camera video according to a frame rate not lower than M, a rectangle is drawn on the intercepted pictures according to the inverter detail mode parameter set N2, and a video with the inverter contour is obtained.

7. The method for photovoltaic power station fault location based on multi-camera and multi-source data fusion according to claim 6, characterized in that: The device whose display area is greater than Q includes the following steps: Using multi-source data to analyze the device with display area greater than Q, if the voltage or current of the photovoltaic string is less than the second threshold or the third threshold, it is determined that the photovoltaic string is faulty, and according to the photovoltaic string optimal viewing angle mode parameter set Z1, it is connected to the corresponding camera and the camera is adjusted; According to the frame rate of not less than M, the picture is intercepted from the camera video picture, the line segment is drawn in the intercepted picture according to the photovoltaic string optimal viewing angle mode parameter set Z1, and the line segment is displayed in the program video display area after drawing is completed; If it is necessary to carry out remote reconnaissance on the photovoltaic string, according to the photovoltaic string detail mode parameter set Z2, it is connected to the corresponding camera and the camera is adjusted; According to the frame rate of not less than M, the picture is intercepted from the camera video picture, the line segment is drawn in the intercepted picture according to the photovoltaic string optimal viewing angle mode parameter set Z1, and the line segment is displayed in the program video display area after drawing is completed.

8. The photovoltaic power station fault location system based on multi-camera and multi-source data fusion, according to any one of claims 1-7, characterized in that: Including, The data acquisition module is used for collecting the equipment data of the photovoltaic power station and judging, deploying the camera and collecting the video picture by using the optimal viewing angle mode and the detail viewing angle mode, and constructing the parameter set; The secondary processing module is used for intercepting the picture from the video picture of the camera according to the frame rate and drawing in the specified area at the same frequency, and obtaining the video with the outline of the observed object; The fault positioning module is used for fault analysis according to the running data, state information and topological relationship between the photovoltaic power station equipment, and positioning the position of the fault by using the video with the outline of the observed object and marking the cable direction related to the fault in the video. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion of any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the photovoltaic power station fault positioning method based on multi-camera and multi-source data fusion of any one of claims 1-7.

Images