A photovoltaic power station operation and maintenance method and system

Abstract

The present invention discloses an operation and maintenance method and system for a photovoltaic power station, which relates to the technical field of photovoltaic power stations and solves the problem that a better operation and maintenance route is not set for operation and maintenance personnel so that the operation and maintenance personnel can not only inspect and repair abnormal parts but also clean up surrounding parts to be cleaned. The present invention bundles routes for those abnormal parts or cleaning parts that do not consume time during route planning for abnormal parts or cleaning parts, and determines relevant final processing routes. Subsequent operation and maintenance personnel perform operation and maintenance processing based on the final processing route. While processing the abnormal parts, relevant cleaning is performed on the cleaning parts that appear in the surrounding areas. While processing the cleaning parts, relevant maintenance is performed on the abnormal parts that appear in the surrounding areas, so as to achieve better operation and maintenance processing effects.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power stations, and in particular to an operation and maintenance method and system of a photovoltaic power station. Background Art

[0002] Photovoltaic power station refers to a photovoltaic power generation system that uses solar energy and special materials such as crystalline silicon panels, inverters and other electronic components to connect to the power grid and transmit electricity to the grid. Photovoltaic power station is one of the green power development energy projects that the country encourages the most.

[0003] Since photovoltaic power stations need to use photovoltaic panels, in order to ensure the corresponding electricity demand, the relevant area of ​​the photovoltaic power station is generally very large. Some photovoltaic power stations may cover hundreds of square kilometers. When carrying out relevant maintenance, since the cleaning and inspection of photovoltaic power stations belong to different operation and maintenance processes, the relevant personnel only know the location of the corresponding abnormal parts when carrying out operation and maintenance processing, and are not clear about the cleaning parts existing in the surrounding area, resulting in the need for subsequent related cleaning personnel to come and clean up. This situation consumes a lot of manpower, and a better operation and maintenance route is not set up for the operation and maintenance personnel, so that the operation and maintenance personnel can not only inspect the abnormal parts, but also clean the surrounding parts to be cleaned, so as to ensure the operation and maintenance effect of the entire photovoltaic power station. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides an operation and maintenance method and system for a photovoltaic power station, which solves the problem of not setting up a better operation and maintenance route for operation and maintenance personnel, so that the operation and maintenance personnel can not only inspect and repair abnormal parts, but also clean the surrounding parts to be cleaned.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: A photovoltaic power station operation and maintenance method, comprising the following steps:

[0006] S1) Define a set of processing cycles. During this processing cycle, use relevant drones to conduct regional inspections. Based on the inspection images, confirm whether there are substandard photovoltaic panels in the corresponding area. The specific method is as follows:

[0007] S11, extracting the surface image of each different photovoltaic panel from the inspection image, comparing the surface image of a single photovoltaic panel with a preset standard image, and confirming whether there is an inconsistent area;

[0008] S12, calibrating the relevant photovoltaic panels with inconsistent areas as substandard photovoltaic panels, and not performing any calibration on the relevant photovoltaic panels without inconsistent areas;

[0009] S2) Confirm the operation data of the corresponding non-compliant photovoltaic panels in this processing cycle, and based on the comparison results of the operation data with the preset interval, mark the non-compliant photovoltaic panels as parts to be cleaned or abnormal parts, and then reprocess the parts to be cleaned or abnormal parts to determine the overtime parts to be cleaned, the low time parts to be cleaned, the over-value abnormal parts and the low-value abnormal parts. The specific method is as follows:

[0010] S21, perform mean processing on several groups of operation data generated in this processing cycle to determine the verification mean J i , where i represents different substandard photovoltaic panels, and the mean value J is identified and checked i Comparison result with the preset interval: If J i ∈ preset interval, the substandard photovoltaic panel is marked as a part to be cleaned. i ∉Preset interval, mark this substandard photovoltaic panel as an abnormal part;

[0011] S22, confirming the area ratio ZB of the inconsistent area of ​​the part to be cleaned, the area ratio ZB=total area of ​​the inconsistent area ÷ total area of ​​the photovoltaic panel surface, comparing ZB with the preset value Y1, if ZB≥Y1, marking the part to be cleaned as an overdue part to be cleaned, if ZB<Y1, marking the part to be cleaned as a low-time part to be cleaned, where Y1 is the preset value;

[0012] S23, check the mean value J of the abnormal parts i Check with the preset interval to determine the endpoint values ​​Qmin and Qmax of the preset interval. If J i -Qmax ≥ Y2 or Qmin-J i ≥Y2, the abnormal part is marked as an over-value abnormal part. i -Qmax<Y2 or Qmin-J i When <Y2, the abnormal part is marked as a low value abnormal part;

[0013] S3) Based on the overdue parts to be cleaned, the low-time parts to be cleaned, the overvalue abnormal parts and the low-value abnormal parts confirmed in this area, plan the final inspection route of the overvalue abnormal parts and the final cleaning route of the overdue parts to be cleaned, and based on the relevant positions of the low-time parts to be cleaned and the low-value abnormal parts, add the corresponding low-time parts to be cleaned and the low-value abnormal parts to the relevant routes as the subsidiary associated parts of the corresponding routes;

[0014] S4) Based on the final inspection route, the final cleaning route and the confirmed auxiliary related parts, the final inspection route and the final cleaning route are marked as the main line. Based on the turning points of the main line, the main line is divided into several single lines. Based on the walking direction, the associated auxiliary related parts of each single line are confirmed. Based on the number and position of the auxiliary related parts, the relevant marking of the single line is confirmed and displayed.

[0015] Preferably, in step S3, the specific method of planning the final inspection route is:

[0016] From the edge of this area, a group of abnormal parts with excessive values ​​are selected as the main part. Based on other abnormal parts with excessive values ​​near the main part, they are randomly combined and connected in sequence until all connections are completed. After confirming several groups of different inspection routes, the abnormal parts with excessive values ​​that have been connected are no longer selected. The lengths of several groups of inspection routes are sorted from small to large, and the top three groups of inspection routes are marked as the inspection routes to be selected.

[0017] For the three groups of inspection routes to be selected: confirm the position of the low-time waiting-to-be-cleaned parts and determine their centers, generate a group of radiating circles with a radius of R1, where R1 is a preset value, calibrate the number of radiating circles intersected by each group of inspection routes to be selected, select the group of inspection routes to be selected with the largest number of intersections as the final inspection routes, and use the low-time waiting-to-be-cleaned parts intersected by the final inspection routes as subsidiary associated parts.

[0018] Preferably, in step S3, the specific method of planning the final cleaning route is:

[0019] From the edge of this area, a group of overdue cleaning parts are selected as the main part, and other overdue cleaning parts adjacent to the main part are randomly combined and connected in sequence until all connections are completed. Several groups of different cleaning routes are confirmed, and the connected overdue cleaning parts are no longer selected. The lengths of several groups of cleaning routes are sorted from small to large, and the top three groups of cleaning routes are marked as the cleaning routes to be selected;

[0020] For the three groups of candidate cleaning routes, confirm the location of low-value abnormal parts and determine their centers, generate a group of radiating circles with a radius of R1, where R1 is a preset value, calibrate the number of radiating circles intersected by each group of candidate cleaning routes, and select the group of candidate inspection routes with the largest number of intersections as the final cleaning route.

[0021] Preferably, the specific method of confirming the relevant calibration of the single line is:

[0022] S41, based on the turning points appearing in the main line, the main line is classified into a plurality of groups of single lines, and the classified single lines are sorted according to the travel direction, and the single lines in the first order are preferentially traveled in the original main line;

[0023] S42, confirming the subsidiary associated parts associated with each group of single lines according to the sorted single lines, selecting the subsidiary associated parts that have been selected and no longer selecting them, and marking the single lines with the number of subsidiary associated parts ≥ 2 as logic confirmation lines;

[0024] S43. Based on the positions of the associated subsidiary parts in the logic confirmation line, generate several groups of processing routes for the subsidiary parts, select a group of processing routes with the shortest path length as the pending routes, and determine the length L1 of the pending routes, and then determine the length L2 of the current logic confirmation line, identify whether L1 and L2 satisfy: L1+L2≤L2×C1, where C1 is a preset fixed coefficient factor, and calibrate this logic confirmation line as a straight processing line. Otherwise, no calibration is performed.

[0025] Furthermore, it also includes:

[0026] The remaining low-value abnormal parts and low-time parts to be cleaned are randomly combined for route planning. They are all marked as parts to be selected. A group of edge parts to be selected are selected as main parts. Based on the adjacent parts to be selected around the main parts, they are randomly combined and connected in sequence until all connections are completed. Several groups of different related routes to be processed are confirmed, and the connected parts to be selected are no longer selected. Based on the route lengths of several groups of related routes to be processed, the related route to be processed with the shortest route length is selected and marked as the final route, and displayed.

[0027] An operation and maintenance system for a photovoltaic power station, comprising:

[0028] UAV image processing module: extracts the surface image of each different photovoltaic panel from the inspection image, and compares the surface image of a single photovoltaic panel with the preset standard image to confirm whether there are inconsistent areas, and calibrates the photovoltaic panels that do not meet the standards;

[0029] Operation data comparison module: compares the operation data of the substandard photovoltaic panels with the preset interval to determine the overtime parts to be cleaned, the low time parts to be cleaned, the over-valued abnormal parts and the low-valued abnormal parts;

[0030] Route planning module: plans the final inspection route for overvalued abnormal parts and the final cleaning route for overtime parts to be cleaned, and based on the relevant positions of low-time parts to be cleaned and low-valued abnormal parts, adds the corresponding low-time parts to be cleaned and low-valued abnormal parts to the relevant routes as subsidiary associated parts of the corresponding routes;

[0031] Logical determination module: Based on the final inspection route, the final cleaning route and the confirmed auxiliary related parts, the final inspection route and the final cleaning route are marked as the main line. Based on the turning points of the main line, the main line is divided into several single lines. Based on the walking direction, the associated auxiliary related parts of each single line are confirmed. Based on the number and position of the auxiliary related parts, the relevant marking of the single line is confirmed and displayed.

[0032] The present invention provides a photovoltaic power station operation and maintenance method and system. Compared with the prior art, it has the following beneficial effects:

[0033] The present invention classifies abnormal parts or cleaned parts in different situations in sequence through image analysis and numerical monitoring, which facilitates the planning of subsequent different operation and maintenance routes;

[0034] In the process of route planning for abnormal parts or cleaning parts, for those abnormal parts or cleaning parts that do not consume time, routes are bundled to determine the relevant final processing routes. Subsequent operation and maintenance personnel perform operation and maintenance based on this final processing route. While processing abnormal parts, relevant cleaning is performed on the cleaning parts that appear in the surrounding areas. While processing cleaning parts, relevant maintenance is performed on the abnormal parts that appear in the surrounding areas, which can achieve better operation and maintenance processing results. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the process of the present invention;

[0036] Figure 2 It is a schematic diagram of the system principle framework of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Embodiment 1

[0039] See also Figure 1 The present application provides an operation and maintenance method of a photovoltaic power station, comprising the following steps:

[0040] S1) Define a set of processing cycles, which are preset cycles, generally 72h or 48h. During this processing cycle, use relevant drones to conduct regional inspections, and confirm whether there are substandard photovoltaic panels in the corresponding area based on the inspection images. Specifically, the relevant drones must use high-definition cameras with 12 million pixels and a data transmission rate of 30M / S. The data transmission interface uses an AV interface. The specific method for confirmation is as follows:

[0041] S11. Extract the surface image of each different photovoltaic panel from the inspection image (because each photovoltaic panel has individual features, such as edge, shape, color, texture, etc., and then extract the features based on machine learning or deep learning algorithm to determine the relevant images of different photovoltaic panels. Since this is the prior art, it will not be described in detail), compare the surface image of a single photovoltaic panel with the preset standard image to confirm whether there is an inconsistent area (that is, during the comparison process, the relevant areas with inconsistent texture and color can be confirmed by relevant machine vision);

[0042] S12, calibrating the relevant photovoltaic panels with inconsistent areas as substandard photovoltaic panels, and not performing any calibration on the relevant photovoltaic panels without inconsistent areas;

[0043] Step S1 is mainly for the related identification of abnormalities on the surface of photovoltaic panels. It is a relatively conventional processing method, so it is relatively simple. When the surface of the photovoltaic panel is abnormal, there are only two situations. One is that there is dust on the surface of the photovoltaic panel, and the other is that there are hot spots or cracks or other situations on the surface of the photovoltaic panel. Such photovoltaic panels will definitely have data abnormalities during use, but photovoltaic panels with dust may not have data abnormalities. Generally, the data is normal.

[0044] S2) Confirm the operation data of the photovoltaic panels that do not meet the standards in this processing cycle (the operation data is power data, which can reflect the power generation capacity of the photovoltaic panels). Based on the comparison results of the operation data and the preset interval, the photovoltaic panels that do not meet the standards are marked as parts to be cleaned or abnormal parts, and then the parts to be cleaned or abnormal parts are reprocessed to determine the overtime parts to be cleaned, the low time parts to be cleaned, the over-value abnormal parts and the low-value abnormal parts. The specific method of reprocessing is as follows:

[0045] S21, perform mean processing on several groups of operation data generated in this processing cycle to determine the verification mean J i , where i represents different substandard photovoltaic panels, and the mean value J is identified and checked i Comparison result with the preset interval: If J i ∈ preset interval, the substandard photovoltaic panel is marked as a part to be cleaned. i ∉Preset interval, the substandard photovoltaic panel is marked as an abnormal part, where the endpoint values ​​of the preset interval are all preset values, which are prepared in advance by the operator. When the corresponding check mean value does not belong to this preset interval, it means that there is a deviation in this value, and it is necessary to determine the abnormal part. If it belongs to the corresponding preset interval, it means that the working performance of this part is normal, but the surface is abnormal, then there is dust on the surface, which needs to be cleaned;

[0046] S22. Confirm the area ratio ZB of the inconsistent area of ​​the part to be cleaned, which is ZB=total area of ​​inconsistent area÷total area of ​​photovoltaic panel surface. Compare ZB with the preset value Y1. If ZB≥Y1, mark the part to be cleaned as an overdue part to be cleaned (indicating that the corresponding panel surface area to be cleaned accounts for a large proportion). If ZB<Y1, mark the part to be cleaned as a low-time part to be cleaned (indicating that the corresponding panel surface area to be cleaned accounts for a small proportion). Y1 is a preset value, which is prepared in advance by the operator.

[0047] S23, check the mean value J of the abnormal parts i Check with the preset interval to determine the endpoint values ​​Qmin and Qmax of the preset interval. If J i -Qmax ≥ Y2 or Qmin-J i ≥Y2, the abnormal part is marked as an over-value abnormal part (indicating a large value deviation). i -Qmax<Y2 or Qmin-J i When <Y2, the abnormal part is marked as a low value abnormal part (indicating a small value deviation);

[0048] S3) Based on the overdue parts to be cleaned, the low-time parts to be cleaned, the overvalue abnormal parts and the low-value abnormal parts confirmed in this area, the final inspection route for the overvalue abnormal parts and the final cleaning route for the overdue parts to be cleaned are planned, and based on the relevant positions of the low-time parts to be cleaned and the low-value abnormal parts, the corresponding low-time parts to be cleaned and the low-value abnormal parts are added to the relevant routes as the subsidiary associated parts of the corresponding routes, wherein the specific method of planning the inspection route is:

[0049] S31. From the edge of the area, select a group of abnormal parts with excessive values ​​as the main part, and randomly combine them based on other abnormal parts with excessive values ​​near the main part, and connect them in sequence until all the connections are completed (if there are three groups of abnormal parts, namely A, B and C, A is the main part, and its route is ABC or ACB, which is randomly combined), confirm several groups of different inspection routes, and the connected abnormal parts with excessive values ​​are no longer selected, that is, after selecting a group of main parts, randomly combine them based on other abnormal parts in the surrounding area until all the abnormal parts are combined, confirm several groups of different inspection routes, sort the lengths of the several groups of inspection routes in ascending order, and mark the top three inspection routes as the inspection routes to be selected;

[0050] From the edge of this area, a group of overdue cleaning parts are selected as the main part, and other overdue cleaning parts adjacent to the main part are randomly combined and connected in sequence until all connections are completed. Several groups of different cleaning routes are confirmed, and the connected overdue cleaning parts are no longer selected. The lengths of several groups of cleaning routes are sorted from small to large, and the top three groups of cleaning routes are marked as the cleaning routes to be selected;

[0051] S32. For the three groups of inspection routes to be selected: confirm the positions of the parts to be cleaned during low-pressure operation and determine their centers, generate a group of radiating circles, the radius of which is R1, where R1 is a preset value, and its specific value is determined by the operator based on experience. The number of radiating circles crossed by each group of inspection routes to be selected is calibrated, and the group of inspection routes to be selected with the largest number of intersections is selected as the final inspection route (if the number is the same, it can be randomly selected), and the parts to be cleaned during low-pressure operation crossed by the final inspection route are used as auxiliary related parts (later, when executing this final inspection route for equipment maintenance, the parts to be cleaned during low-pressure operation can be cleaned at will while walking, which does not take much time and can achieve better power station operation and maintenance processing effects);

[0052] S33. For the three groups of to-be-selected cleaning routes, confirm the positions of the low-value abnormal parts, determine their centers, generate a group of radiating circles, and the radius value of the radiating circles is R1, where R1 is a preset value, and its specific value is determined by the operator based on experience. The number of radiating circles crossed by each group of to-be-selected cleaning routes is calibrated, and a group of to-be-selected operation and inspection routes with the largest number of intersections is selected as the final cleaning route (if the number is the same, it can be randomly selected, because the route needs to pass through other photovoltaic modules during the planning process, then based on the relevant positions of the corresponding to-be-cleaned parts, the relevant radiation range of the corresponding to-be-cleaned parts, that is, the radiating circles, can be confirmed. Based on the intersection situation, the best processing route is determined, which has the shortest distance and can also process the abnormal parts that are not time-consuming on the basis of cleaning the photovoltaic modules, so that the abnormal parts return to normal, so as to achieve better operation and maintenance processing effects), and the low-value abnormal parts crossed by the final cleaning route are used as subsidiary associated parts;

[0053] S4) Based on the final inspection route and the final cleaning route and the confirmed auxiliary related parts, the final inspection route and the final cleaning route are marked as the main line, and based on the turning point of the main line, the main line is divided into several single lines, and then based on the walking direction, the associated auxiliary related parts of each single line are confirmed, and based on the number and position of the auxiliary related parts, the relevant marking of the single line is confirmed and displayed, wherein the specific method of confirmation is:

[0054] S41. Based on the turning points that appear in the main line (i.e., the points where the walking direction changes, when walking to the left, it is the first group of single lines, when the direction changes upward, then the second group of single lines is generated, and so on, based on the turning points, the single lines generated subsequently are confirmed one by one), the main line is classified into several groups of single lines, and the classified single lines are sorted according to the walking direction, and the single lines sorted first are given priority in walking in the original main line;

[0055] S42, according to the sorted single lines, confirm the subsidiary associated parts associated with each group of single lines (that is, the subsidiary associated parts that cross each other), and the selected subsidiary associated parts are no longer selected (when the corresponding subsidiary associated parts are located at a corner, they will intersect with the previous and next groups of single lines. After the previous group of single lines selects this subsidiary associated part, the next group of single lines will not select this subsidiary associated part), and mark the single lines with the number of subsidiary associated parts ≥ 2 as logical confirmation lines;

[0056] S43. Based on the positions of the associated subsidiary parts in the logic confirmation line, generate several groups of processing routes for the subsidiary parts, select a group of processing routes with the shortest path length as the pending routes, and determine the length L1 of the pending routes, and then determine the length L2 of the current logic confirmation line, and identify whether L1 and L2 satisfy: L1+L2≤L2×C1, where C1 is a preset fixed coefficient factor, which is formulated by the operator based on experience and generally takes a value of 1.3, that is, after the subsidiary parts are processed simultaneously, the route length distance for processing the abnormal parts does not exceed 1.3 times the original distance, and this logic confirmation line is calibrated as a straight processing line. Otherwise, no calibration is performed. When the operator displays that the single line is a straight processing line through relevant equipment during the processing process, the single line is processed in sequence first, and when the subsidiary parts are processed, all the subsidiary parts are processed. , and then process the single line, which will not affect the processing rate. On the contrary, it improves the processing rate of the corresponding single line and reduces the time for changing equipment. Because in the actual processing process, the equipment for handling abnormal parts or cleaning parts is not the same, the operator needs to go to the front of the vehicle to replace the equipment. Because the photovoltaic panel area of ​​the photovoltaic station is large, it is generally necessary to drive to carry out equipment maintenance. For example: the original single line is the final inspection route, and there are two abnormal parts, A and B, which are located at the two end points of this final inspection route. Based on this single line, there are three groups of C, D and E low-time parts to be cleaned around it. Then, when cleaning this single line, the three groups of C, D and E low-time parts to be cleaned can be cleaned first, and then the maintenance of abnormal parts can be carried out, which can reduce the overall processing time to achieve the corresponding processing efficiency. For other single lines, no display is performed, and the operator decides how to deal with them;

[0057] S5) Perform route planning for the remaining low-value abnormal parts and low-time to-be-cleaned parts through random combination, mark them as to-be-selected parts, select a group of edge to-be-selected parts as main parts, and randomly combine them based on the adjacent to-be-selected parts around the main parts, connect them in sequence until all connections are completed, confirm several groups of different to-be-processed related routes, and the connected to-be-selected parts will no longer be selected. Based on the route lengths of several groups of to-be-processed related routes, select the to-be-processed related route with the shortest route length to be marked as the final route, and display it.

[0058] Embodiment 2

[0059] A photovoltaic power station operation and maintenance system, combined with Figure 2 ,include:

[0060] UAV image processing module: extracts the surface image of each different photovoltaic panel from the inspection image, and compares the surface image of a single photovoltaic panel with the preset standard image to confirm whether there are inconsistent areas, and calibrates the photovoltaic panels that do not meet the standards;

[0061] Operation data comparison module: compares the operation data of the substandard photovoltaic panels with the preset interval to determine the overtime parts to be cleaned, the low time parts to be cleaned, the over-valued abnormal parts and the low-valued abnormal parts;

[0062] Route planning module: plans the final inspection route for overvalued abnormal parts and the final cleaning route for overtime parts to be cleaned, and based on the relevant positions of low-time parts to be cleaned and low-valued abnormal parts, adds the corresponding low-time parts to be cleaned and low-valued abnormal parts to the relevant routes as subsidiary associated parts of the corresponding routes;

[0063] Logical determination module: Based on the final inspection route, the final cleaning route and the confirmed auxiliary related parts, the final inspection route and the final cleaning route are marked as the main line. Based on the turning points of the main line, the main line is divided into several single lines. Based on the walking direction, the associated auxiliary related parts of each single line are confirmed. Based on the number and position of the auxiliary related parts, the relevant marking of the single line is confirmed and displayed.

[0064] Some of the data in the above formulas are numerically calculated by removing their dimensions. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0065] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A photovoltaic power station operation and maintenance method, characterized in that: The following steps are involved: S1) Define a set of processing cycles, during which relevant drones are used to conduct regional inspections, and confirm whether there are substandard photovoltaic panels in the corresponding area based on the inspection images; S2) Confirm the operation data of the corresponding non-compliant photovoltaic panels in this processing cycle, and based on the comparison results of the operation data with the preset interval, mark the non-compliant photovoltaic panels as parts to be cleaned or abnormal parts, and then reprocess the parts to be cleaned or abnormal parts to determine the overtime parts to be cleaned, the low time parts to be cleaned, the over-value abnormal parts and the low-value abnormal parts. The specific method is as follows: S21, perform mean processing on several groups of operation data generated in this processing cycle to determine the verification mean J i , where i represents different substandard photovoltaic panels, and the mean value J is identified and checked i Comparison result with the preset interval: If J i If the panel is within the preset range, the panel will be marked as a part to be cleaned. i If it does not fall within the preset range, the substandard photovoltaic panel will be marked as an abnormal part; S22, confirming the area ratio ZB of the inconsistent area of ​​the part to be cleaned, the area ratio ZB=total area of ​​the inconsistent area ÷ total area of ​​the photovoltaic panel surface, comparing ZB with the preset value Y1, if ZB≥Y1, marking the part to be cleaned as an overdue part to be cleaned, if ZB<Y1, marking the part to be cleaned as a low-time part to be cleaned, where Y1 is the preset value; S23, check the mean value J of the abnormal parts i Check with the preset interval to determine the endpoint values ​​Qmin and Qmax of the preset interval. If J i -Qmax ≥ Y2 or Qmin-J i ≥Y2, the abnormal part is marked as an over-value abnormal part. i -Qmax<Y2 or Qmin-J i When <Y2, the abnormal part is marked as a low value abnormal part; S3) Based on the overdue parts to be cleaned, the low-time parts to be cleaned, the overvalue abnormal parts and the low-value abnormal parts confirmed in this area, plan the final inspection route of the overvalue abnormal parts and the final cleaning route of the overdue parts to be cleaned, and based on the relevant positions of the low-time parts to be cleaned and the low-value abnormal parts, add the corresponding low-time parts to be cleaned and the low-value abnormal parts to the relevant routes as the subsidiary associated parts of the corresponding routes; S4) Based on the final inspection route and the final cleaning route and the confirmed auxiliary related parts, the final inspection route and the final cleaning route are marked as the main line. Based on the turning point of the main line, the main line is divided into several single lines. Then, based on the walking direction, the associated auxiliary related parts of each single line are confirmed. Based on the number and position of the auxiliary related parts, the relevant marking of the single line is confirmed and displayed. The specific method is as follows: S41, based on the turning points appearing in the main line, the main line is classified into a plurality of groups of single lines, and the classified single lines are sorted according to the travel direction, and the single lines in the first order are preferentially traveled in the original main line; S42, confirming the subsidiary associated parts associated with each group of single lines according to the sorted single lines, selecting the subsidiary associated parts that have been selected and no longer selecting them, and marking the single lines with the number of subsidiary associated parts ≥ 2 as logic confirmation lines; S43. Based on the positions of the associated subsidiary parts in the logic confirmation line, generate several groups of processing routes for the subsidiary parts, select a group of processing routes with the shortest path length as the pending routes, and determine the length L1 of the pending routes, and then determine the length L2 of the current logic confirmation line, identify whether L1 and L2 satisfy: L1+L2≤L2×C1, where C1 is a preset fixed coefficient factor, and calibrate this logic confirmation line as a straight processing line. Otherwise, no calibration is performed.

2. A photovoltaic power station operation and maintenance method according to claim 1, characterized in that: In step S1, the specific method of confirming the photovoltaic panels that do not meet the standards is: S11, extracting the surface image of each different photovoltaic panel from the inspection image, comparing the surface image of a single photovoltaic panel with a preset standard image, and confirming whether there is an inconsistent area; S12, calibrating the relevant photovoltaic panels with inconsistent areas as substandard photovoltaic panels, and not performing any calibration on the relevant photovoltaic panels without inconsistent areas.

3. The operation and maintenance method of a photovoltaic power station according to claim 1, characterized in that: In step S3, the specific method of planning the final inspection route is: From the edge of this area, a group of abnormal parts with excessive values ​​are selected as the main part. Based on other abnormal parts with excessive values ​​near the main part, they are randomly combined and connected in sequence until all connections are completed. After confirming several groups of different inspection routes, the abnormal parts with excessive values ​​that have been connected are no longer selected. The lengths of several groups of inspection routes are sorted from small to large, and the top three groups of inspection routes are marked as the inspection routes to be selected. For the three groups of inspection routes to be selected: confirm the position of the low-time waiting-to-be-cleaned parts and determine their centers, generate a group of radiating circles with a radius of R1, where R1 is a preset value, calibrate the number of radiating circles intersected by each group of inspection routes to be selected, select the group of inspection routes to be selected with the largest number of intersections as the final inspection routes, and use the low-time waiting-to-be-cleaned parts intersected by the final inspection routes as subsidiary associated parts.

4. The operation and maintenance method of a photovoltaic power station according to claim 1, characterized in that: In step S3, the specific method of planning the final cleaning route is: From the edge of this area, a group of overdue cleaning parts are selected as the main part, and other overdue cleaning parts adjacent to the main part are randomly combined and connected in sequence until all connections are completed. Several groups of different cleaning routes are confirmed, and the connected overdue cleaning parts are no longer selected. The lengths of several groups of cleaning routes are sorted from small to large, and the top three groups of cleaning routes are marked as the cleaning routes to be selected; For the three groups of candidate cleaning routes, confirm the location of low-value abnormal parts and determine their centers, generate a group of radiating circles with a radius of R1, where R1 is a preset value, calibrate the number of radiating circles intersected by each group of candidate cleaning routes, and select the group of candidate inspection routes with the largest number of intersections as the final cleaning route.

5. The photovoltaic power station operation and maintenance method according to claim 1, characterized in that: Also includes: The remaining low-value abnormal parts and low-time parts to be cleaned are randomly combined for route planning. They are all marked as parts to be selected. A group of edge parts to be selected are selected as main parts. Based on the adjacent parts to be selected around the main parts, they are randomly combined and connected in sequence until all connections are completed. Several groups of different related routes to be processed are confirmed, and the connected parts to be selected are no longer selected. Based on the route lengths of several groups of related routes to be processed, the related route to be processed with the shortest route length is selected and marked as the final route, and displayed.

6. An operation and maintenance system for a photovoltaic power station, the operation and maintenance system is operated based on an operation and maintenance method for a photovoltaic power station according to any one of claims 1 to 5, characterized in that: include: UAV image processing module: extracts the surface image of each different photovoltaic panel from the inspection image, and compares the surface image of a single photovoltaic panel with the preset standard image to confirm whether there are inconsistent areas, and calibrates the photovoltaic panels that do not meet the standards; Operation data comparison module: compares the operation data of the substandard photovoltaic panels with the preset interval to determine the overtime parts to be cleaned, the low time parts to be cleaned, the over-valued abnormal parts and the low-valued abnormal parts; Route planning module: plans the final inspection route for overvalued abnormal parts and the final cleaning route for overtime parts to be cleaned, and based on the relevant positions of low-time parts to be cleaned and low-valued abnormal parts, adds the corresponding low-time parts to be cleaned and low-valued abnormal parts to the relevant routes as subsidiary associated parts of the corresponding routes; Logical determination module: Based on the final inspection route, the final cleaning route and the confirmed auxiliary related parts, the final inspection route and the final cleaning route are marked as the main line. Based on the turning points of the main line, the main line is divided into several single lines. Based on the walking direction, the associated auxiliary related parts of each single line are confirmed. Based on the number and position of the auxiliary related parts, the relevant marking of the single line is confirmed and displayed.

Images