An optimization system for photovoltaic array positioning

Abstract

The present invention relates to the technical field of positioning optimization, and specifically discloses an optimization system for photovoltaic array positioning, comprising: an RFID information management and identification module; a UWB drone positioning module; a drone inspection and anomaly detection module; a ground station monitoring and operation and maintenance feedback module; an offset point analysis module: analyzing the deviation regularity of the positioning point, and based on the analysis result, marking the positioning point as a stable deviation point; a stable deviation point correction module: calculating the correction vector of the stable deviation point, thereby correcting it; a stable deviation area division module; a drone path optimization module: obtaining the regional correction vector of each deviation correction area, and optimizing the drone inspection path; the present invention not only enhances the positioning accuracy and inspection operation efficiency of the drone, but also makes the fault response more rapid, thereby achieving fewer correction requirements during the operation and maintenance process, and effectively improving the overall operation and maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of power loss, and in particular to an optimization system for photovoltaic array positioning. Background Art

[0002] A photovoltaic array, also known as a photovoltaic array, is a DC power generation unit consisting of several photovoltaic modules or panels mechanically and electrically assembled together with a fixed support structure. It is the largest photovoltaic power generation system, converting sunlight into DC electricity. A photovoltaic module, consisting of multiple solar cells, converts solar energy into DC electricity. A photovoltaic array combines multiple modules and interconnects them to form a unified system to increase power output. A photovoltaic array consists of photovoltaic modules, mounting brackets, cables, and an AC / DC converter.

[0003] Chinese invention patent publication number CN118017935A discloses a fault location system and method for photovoltaic arrays, belonging to the field of photovoltaic power generation technology. The fault location system includes: M*N parameter acquisition modules, which are used to obtain photovoltaic parameters of each photovoltaic panel in a periodic or real-time manner to obtain a photovoltaic parameter array; an analysis module, which counts photovoltaic parameters of each row according to the photovoltaic parameter array, and determines whether there is a fault in the photovoltaic panel in that row based on the statistical results of each row of photovoltaic parameters; counts photovoltaic parameters of each column, and determines whether there is a fault in the photovoltaic panel in that column based on the statistical results of each column of photovoltaic parameters; and a fault location module, which receives fault row information and fault column information output by the analysis module when the analysis module determines that a fault exists, and defines the photovoltaic panel corresponding to the intersection of the fault row information and the fault column information as the faulty photovoltaic panel.

[0004] However, with the widespread use of photovoltaic energy, the operation and maintenance of photovoltaic arrays has become increasingly important. Traditional O&M methods rely on manual inspections, which are not only inefficient but also hinder real-time monitoring and rapid response of photovoltaic arrays. In recent years, although drone inspection technology has been gradually applied to the O&M of photovoltaic arrays, its positioning accuracy and inspection efficiency remain insufficient, resulting in frequent corrections during the O&M process, increasing O&M costs and time.

[0005] Therefore, how to reduce the need for corrections during drone inspections and improve positioning accuracy and inspection efficiency has become an urgent issue to be addressed in the current operation and maintenance management of photovoltaic arrays. Summary of the Invention

[0006] The object of the present invention is to provide an optimization system for photovoltaic array positioning to solve the technical problems in the above background.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] The present invention provides an optimization system for photovoltaic array positioning, comprising:

[0009] RFID information management and identification module: reads the information of RFID tags to achieve digital management and tracking of photovoltaic modules;

[0010] UWB drone positioning module: Using a drone equipped with a UWB positioning tag, when the drone is inspecting over the photovoltaic array, the distance between the drone and each UWB positioning base station is measured, and the two-dimensional coordinates of the drone are calculated in real time to achieve precise positioning of the drone;

[0011] Drone inspection and anomaly detection module: UAVs equipped with detection equipment can capture high-definition infrared images of photovoltaic arrays to detect the temperature distribution of components;

[0012] Ground station monitoring and operation and maintenance feedback module: Receives abnormal images and fault point detection coordinates sent by drones and locates faulty components;

[0013] Offset point analysis module: Based on the historical positioning data of each positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point;

[0014] Stable deviation point correction module: When a positioning point is marked as a stable deviation point, the correction vector of the stable deviation point is calculated to correct it;

[0015] Stable deviation area division module: Based on the distribution of stable deviation points in the photovoltaic array and the correction vectors of each stable deviation point, the photovoltaic array is divided into regions to form different deviation correction areas;

[0016] UAV path optimization module: obtains the regional correction vector of each deviation correction area and optimizes the UAV inspection path.

[0017] As a further solution of the present invention, based on the historical positioning data of each positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis result, the positioning point is marked as a stable deviation point. The process is as follows:

[0018] Based on the UWB UAV positioning module, the historical positioning data of the UAV is obtained, and based on the historical positioning data of the UAV, all historical deviation vectors of each positioning point are obtained;

[0019] Based on all historical deviation vectors of the positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point.

[0020] As a further solution of the present invention, based on all historical deviation vectors of any positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point. The specific process is as follows:

[0021] Based on any historical deviation vector of the positioning point, mark the historical deviation vector as the target deviation vector;

[0022] Compare and analyze the target deviation vector with all other historical deviation vectors, obtain the Euclidean distance between the target deviation vector and the historical deviation vector, and evaluate the similarity between the target deviation vector and the historical deviation vector. Based on the evaluation results, mark the historical deviation vector as a similar deviation vector.

[0023] Mark the historical deviation vector and all corresponding similar deviation vectors as a positioning point similar deviation vector group, and calculate the similarity representation value of the positioning point similar deviation vector group;

[0024] Preset a similarity representation threshold, and compare and analyze the similarity representation value with the similarity representation threshold;

[0025] If the similarity representation value is less than or equal to the similarity representation threshold, the positioning point is marked as a stable deviation point, and at the same time, the positioning point similar deviation vector group corresponding to the similarity representation value is marked as a positioning point stable deviation vector group.

[0026] As a further solution of the present invention, the process of obtaining the Euclidean distance between the target deviation vector and the historical deviation vector, thereby evaluating the similarity between the target deviation vector and the historical deviation vector, and marking the historical deviation vector as a similar deviation vector based on the evaluation result is as follows:

[0027] By formula: Calculate the Euclidean distance D, where A i is the i-th component of the target deviation vector, B i is the i-th component of the historical deviation vector, i is 1, 2, 3, ..., n, and n is the dimension of the target deviation vector and the historical deviation vector;

[0028] Preset the Euclidean distance threshold, and compare and analyze the Euclidean distance with the Euclidean distance threshold;

[0029] If the Euclidean distance is less than or equal to the Euclidean distance threshold, the Euclidean distance is used as a further solution of the present invention: the process of obtaining the similarity representation value is:

[0030] By formula: The similarity representation value EX is calculated, where F is the total number of all historical deviation vectors in the similar deviation vector group of the positioning point, and M is the total number of all historical deviation vectors. is the average value of the Euclidean distance of all historical deviation vectors in the similar deviation vector group of the positioning point, is the average value of the Euclidean distances corresponding to all historical deviation vectors of the anchor point.

[0031] As a further solution of the present invention, based on the distribution of stable deviation points in the photovoltaic array and the correction vectors of each stable deviation point, the photovoltaic array is divided into regions to form different deviation correction regions. The process is as follows:

[0032] Based on any stable deviation point, extract the stable deviation points adjacent to the stable deviation point, mark them as first points to be clustered, calculate the cluster characterization value of the stable deviation point and the first points to be clustered, and then evaluate whether the stable deviation point and the first points to be clustered are clustered based on the cluster characterization value, and obtain a first cluster area based on the evaluation result;

[0033] Extracting stable deviation points adjacent to the first clustering region and marking them as second points to be clustered, evaluating whether the stable deviation points and the second points to be clustered are clustered, and obtaining a third clustering region based on the evaluation result;

[0034] Repeat the above clustering process until all adjacent stable deviation points in the obtained a-th cluster area cannot be clustered, and mark the a-th cluster area as the deviation correction area.

[0035] As a further solution of the present invention: the process of obtaining the cluster characterization value is:

[0036] By formula: The cluster characterization value Km is calculated, where L is the cluster distance, P is the cluster deviation, and ΔL is the average distance between all positioning points.

[0037] As a further solution of the present invention: Based on the cluster characterization value, the process of evaluating whether the stable deviation point and the first to-be-clustered point are clustered, and obtaining the first cluster area based on the evaluation result is as follows:

[0038] Preset the cluster characterization threshold, and compare and analyze the cluster characterization value with the cluster characterization threshold;

[0039] If the cluster characterization value is less than or equal to the cluster characterization threshold, the stable deviation point and the first point to be clustered are divided into a first cluster area.

[0040] As a further solution of the present invention: the process of obtaining the regional correction vector is:

[0041] The correction vectors of all stable deviation points in the deviation correction area are extracted and the mean is calculated to obtain the regional correction vector.

[0042] As a further solution of the present invention: the process of obtaining the path priority value is:

[0043] By formula: Calculate the path priority value PRI, where SZ is the total distance of the drone inspection path, S1 r S2 is the distance from the r-1th deviation correction area to the rth deviation correction area on the UAV inspection path, r is the distance within the rth deviation correction area on the UAV inspection path, B1 r B2 is the change value of the drone’s entry correction when it enters the rth deviation correction area on the drone inspection path. r is the exit correction change value of the drone when it leaves the rth deviation correction area on the drone inspection path. The value of r is 1, 2, 3, ..., K, K is the total number of deviation correction areas on the drone inspection path, α1 is the preset first hyperparameter, α2 is the preset second hyperparameter, and both are greater than 0. e is a natural constant.

[0044] Beneficial effects of the present invention:

[0045] (1) The present invention uses RFID technology to achieve digital management and tracking of photovoltaic modules, uses UWB positioning technology to achieve precise inspection and positioning of drones, combines the detection equipment carried by drones to perform abnormality detection, and implements monitoring and operation and maintenance feedback through ground stations, thereby effectively improving the operation and maintenance efficiency and fault response speed of photovoltaic arrays;

[0046] (2) The present invention not only enhances the positioning accuracy and inspection efficiency of drones, but also provides the operation and maintenance team with more accurate and immediate fault feedback information. As a result, the operation and maintenance process of the photovoltaic array is greatly optimized, and the fault response is faster, thereby achieving fewer correction requirements during the operation and maintenance process, effectively improving the overall operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0048] Figure 1 It is a flowchart of the present invention;

[0049] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0051] Example 1:

[0052] See also Figure 1 、 Figure 2 As shown, an optimization system for photovoltaic array positioning according to an embodiment of the present invention includes:

[0053] RFID information management and identification module: Use RFID readers to read the information on RFID tags, thereby realizing digital management and tracking of photovoltaic modules;

[0054] RFID readers are deployed at key locations in the photovoltaic array. The selection of key locations should be determined based on the specific layout, component type, quantity, and management requirements of the photovoltaic array. At the same time, the performance and coverage of the RFID readers, as well as the impact of environmental factors (such as interference and occlusion) on the reading and writing effects should also be considered.

[0055] It should be explained that each photovoltaic module is installed with an RFID tag, which stores the unique identification information of the module and is used to achieve individual identification of the module;

[0056] UWB drone positioning module: Using a drone equipped with a UWB positioning tag (which can send and receive UWB signals), when the drone is inspecting over the photovoltaic array, the distance between the drone and each UWB positioning base station is measured, and the two-dimensional coordinates of the drone are calculated in real time to achieve precise positioning of the drone;

[0057] It should be explained that at least two UWB positioning base stations are installed around the photovoltaic array, and the UWB positioning base stations are connected wirelessly to form a positioning network;

[0058] In some implementation schemes, when a drone equipped with a UWB positioning tag inspects over a photovoltaic array, the UWB positioning tag continuously sends UWB signals to various UWB positioning base stations;

[0059] After receiving the UWB signal, the UWB positioning base station measures the time delay of the UWB signal transmitted from the UWB positioning tag to the UWB positioning base station (the time delay is proportional to the distance from the UWB positioning tag to the UWB positioning base station);

[0060] The actual distance from the UWB positioning tag on the drone to each UWB positioning base station is calculated using the propagation speed of the UWB signal and the measured time delay.

[0061] Using geometric positioning algorithms (including but not limited to triangulation and multilateration), the distance information from the UAV to each UWB positioning base station is used to calculate the coordinates of the UAV on a two-dimensional plane.

[0062] The distance information between the UAV and each UWB positioning base station includes but is not limited to: the actual distance between the UWB positioning tag on the UAV and each UWB positioning base station, and the location information of each UWB positioning base station;

[0063] Among them, the specific process of calculating the coordinates of the drone on the two-dimensional plane is:

[0064] Among the deployed UWB positioning base stations, select one as a reference base station (the reference base station is usually a UWB positioning base station with a known and relatively stable location);

[0065] The position of the reference base station is used as the origin (0,0) of the two-dimensional plane coordinate system;

[0066] According to the relative position relationship between UWB positioning base stations, the directions of the X-axis and Y-axis are defined, and the relative coordinates of each UWB positioning base station are obtained;

[0067] Based on the relative coordinates of each UWB positioning base station and the distance information from the UWB positioning tag to each UWB positioning base station, the two-dimensional coordinates of the drone are calculated in real time;

[0068] Drone inspection and anomaly detection module: UAVs equipped with detection equipment can capture high-definition infrared images of photovoltaic arrays to detect the temperature distribution of components;

[0069] Among them, the detection equipment includes but is not limited to: infrared cameras;

[0070] Then, through the image recognition algorithm, anomalies in the image are identified (abnormal phenomena include: hot spots). When an abnormal component is found, the coordinates of the drone's current position are marked as the fault point detection coordinates, and the drone sends the abnormal image and fault point detection coordinates to the ground station;

[0071] Ground station monitoring and operation and maintenance feedback module: Receives abnormal images and fault point detection coordinates sent by drones and locates faulty components;

[0072] The ground station provides real-time feedback of positioning results and fault information to the operation and maintenance team, enabling them to respond quickly and take appropriate measures to fix faulty components, ensuring efficient and stable operation of the PV array.

[0073] The technical solution of the embodiment of the present invention is mainly as follows: a photovoltaic array positioning optimization system is constructed that integrates RFID information management and identification, UWB drone positioning, drone inspection and anomaly detection, and ground station monitoring and operation and maintenance feedback. The system realizes digital management and tracking of photovoltaic components through RFID technology, uses UWB positioning technology to realize precise inspection and positioning of drones, combines the detection equipment carried by drones for anomaly detection, and realizes monitoring and operation and maintenance feedback through ground stations, thereby effectively improving the operation and maintenance efficiency and fault response speed of photovoltaic arrays.

[0074] Example 2:

[0075] Based on Example 1, please refer to Figure 1 、 Figure 2 As shown, the optimization system for photovoltaic array positioning according to the embodiment of the present invention further includes:

[0076] Offset point analysis module: Based on the historical positioning data of each positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point;

[0077] In some embodiments, based on the UWB UAV positioning module, historical positioning data of the UAV is obtained, and based on the historical positioning data of the UAV, all historical deviation vectors of each positioning point are obtained;

[0078] The historical positioning data includes: the position information measured by the UWB positioning base station during the inspection process (i.e., the measured position), as well as the actual position information of the drone obtained through other means (such as GPS or on-site verification by operation and maintenance personnel);

[0079] It should be explained that there are various methods for obtaining the actual position of a drone, including but not limited to: GPS positioning: In an open and unobstructed environment, GPS can provide relatively accurate drone location information as a reference for the actual position; on-site verification by operators: In certain circumstances, operators can correct the actual position of a drone through on-site observation and measurement. This method is particularly suitable for verifying positioning accuracy in complex environments;

[0080] It should be explained that the process of obtaining the historical deviation vector of the positioning point is as follows: during the historical inspection process of the UAV, every time the UAV reaches a positioning point, two pieces of position information are recorded at the same time: one is the UAV position measured by the UWB positioning base station, and the other is the actual position of the UAV obtained by other reliable means. Subsequently, based on the difference between the two positions, that is, the deviation between the actual position of the UAV and the position measured by the UWB positioning base station, the historical deviation vector of the positioning point is obtained;

[0081] It should be explained that positioning points are specific coordinate locations on the flight path of a drone inspecting a photovoltaic array. These locations are pre-set or dynamically determined by the drone during the inspection process based on actual conditions and serve as reference points for the drone's positioning, navigation, and inspection.

[0082] Based on all historical deviation vectors of the positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point;

[0083] The specific process of analyzing the deviation regularity of any positioning point based on all historical deviation vectors of the positioning point and marking the positioning point as a stable deviation point based on the analysis results is as follows:

[0084] Based on any historical deviation vector of the positioning point, mark the historical deviation vector as the target deviation vector;

[0085] Compare and analyze the target deviation vector with all other historical deviation vectors, obtain the Euclidean distance between the target deviation vector and the historical deviation vector, and evaluate the similarity between the target deviation vector and the historical deviation vector. Based on the evaluation results, mark the historical deviation vector as a similar deviation vector.

[0086] Exemplarily, the process of obtaining the Euclidean distance between the target deviation vector and the historical deviation vector, thereby evaluating the similarity between the target deviation vector and the historical deviation vector, and marking the historical deviation vector as a similar deviation vector based on the evaluation result is as follows:

[0087] By formula: Calculate the Euclidean distance D, where A i is the i-th component of the target deviation vector, B i is the i-th component of the historical deviation vector, i is 1, 2, 3, ..., n, and n is the dimension of the target deviation vector and the historical deviation vector;

[0088] Preset the Euclidean distance threshold, and compare and analyze the Euclidean distance with the Euclidean distance threshold;

[0089] If the Euclidean distance is less than or equal to the Euclidean distance threshold, it is determined that the similarity between the target deviation vector corresponding to the Euclidean distance and the historical deviation vector is high, and the historical deviation vector corresponding to the Euclidean distance is marked as a similar deviation vector;

[0090] If the Euclidean distance is greater than the Euclidean distance threshold, it is determined that the similarity between the target deviation vector and the historical deviation vector corresponding to the Euclidean distance is low;

[0091] Mark the historical deviation vector and all corresponding similar deviation vectors as a positioning point similar deviation vector group, and calculate the similarity representation value of the positioning point similar deviation vector group;

[0092] Exemplarily, the process of obtaining the similarity representation value is as follows:

[0093] By formula: The similarity representation value EX is calculated, where F is the total number of all historical deviation vectors in the positioning point similar deviation vector group, M is the total number of all historical deviation vectors, is the average value of the Euclidean distance of all historical deviation vectors in the similar deviation vector group of the positioning point, is the average value of the Euclidean distances corresponding to all historical deviation vectors of the positioning point;

[0094] It needs to be explained that It reflects the representativeness of the similar deviation vector group relative to the overall deviation. The closer the ratio is to 1, the more representative the deviation of the similar deviation vector group is of the overall deviation, which means the deviation stability of the positioning point is higher.

[0095] Preset a similarity representation threshold, and compare and analyze the similarity representation value with the similarity representation threshold;

[0096] If the similarity representation value is less than or equal to the similarity representation threshold, it means that the deviation of the positioning point has a high regularity, and it is marked as a stable deviation point. At the same time, the positioning point similar deviation vector group corresponding to the similarity representation value is marked as the positioning point stable deviation vector group;

[0097] If the similarity representation value is greater than the similarity representation threshold, it means that the deviation of the positioning point has a low regularity;

[0098] Stable deviation point correction module: When a positioning point is marked as a stable deviation point, the correction vector of the stable deviation point is calculated to correct it;

[0099] In some embodiments, after the positioning point is marked as a stable deviation point, all historical deviation vectors in the stable deviation vector group are extracted and averaged to obtain a correction vector;

[0100] Based on the correction vector, the positioning result of the UAV at the stable deviation point is corrected (for example, when the UAV reaches the stable deviation point, its position information can be adjusted to eliminate the known deviation);

[0101] Stable deviation area division module: Based on the distribution of stable deviation points in the photovoltaic array and the correction vectors of each stable deviation point, the photovoltaic array is divided into regions to form different deviation correction areas;

[0102] In some embodiments, based on any stable deviation point, stable deviation points adjacent to the stable deviation point are extracted and marked as first points to be clustered, cluster characterization values of the stable deviation point and the first points to be clustered are calculated, and then based on the cluster characterization values, whether the stable deviation point and the first points to be clustered are evaluated, and based on the evaluation result, a first clustering area is obtained;

[0103] It should be explained that the adjacent stable deviation points in the stable deviation points adjacent to each other indicate that the positioning points that appear consecutively on the inspection path of the UAV are all stable deviation points;

[0104] The process of obtaining the cluster representation value is as follows:

[0105] By formula: Calculate the cluster representation value Km, where L is the cluster distance, P is the cluster deviation, and ΔL is the average distance between all positioning points;

[0106] It should be explained that the clustering distance is the distance between the stable deviation point and the first point to be clustered. It should be particularly explained that in the subsequent clustering process, the clustering distance is the distance between the center of the clustering area and the point to be clustered. For example, the clustering distance between the first clustering area and the second point to be clustered is the distance between the center of the first clustering area and the second point to be clustered.

[0107] It should be explained that the clustering deviation is the difference between the correction vector of the stable deviation point and the correction vector of the first point to be clustered. It should be particularly explained that in the subsequent clustering process, the clustering deviation is the difference between the mean of all correction vectors in the clustering area and the correction vector of the point to be clustered. For example, the clustering deviation between the first clustering area and the second point to be clustered is the difference between the mean of all correction vectors in the first clustering area and the correction vector of the second point to be clustered.

[0108] Based on the cluster characterization value, the process of evaluating whether the stable deviation point and the first point to be clustered are clustered and obtaining the first cluster area based on the evaluation result is as follows:

[0109] Preset the cluster characterization threshold, and compare and analyze the cluster characterization value with the cluster characterization threshold;

[0110] If the cluster characterization value is less than or equal to the cluster characterization threshold, it is determined that the stable deviation point and the first point to be clustered can be clustered, that is, the stable deviation point and the first point to be clustered are divided into a first clustering area;

[0111] Extracting stable deviation points adjacent to the first clustering region and marking them as second points to be clustered, evaluating whether the stable deviation points and the second points to be clustered are clustered, and obtaining a third clustering region based on the evaluation result;

[0112] Repeat the above clustering process until all adjacent stable deviation points in the obtained cluster area a cannot be clustered, and mark the cluster area a as the deviation correction area;

[0113] UAV path optimization module: obtains the regional correction vector of each deviation correction area and optimizes the UAV inspection path;

[0114] In some embodiments, the process of obtaining the region correction vector is:

[0115] Extract the correction vectors of all stable deviation points in the deviation correction area and perform mean calculation to obtain the regional correction vector;

[0116] The process of optimizing the drone inspection path is as follows:

[0117] Preset multiple drone inspection paths, calculate the path priority value of each drone inspection path, and select the drone inspection path corresponding to the path priority value with the smallest value as the optimized drone inspection path;

[0118] Exemplarily, the process of obtaining the path priority value is as follows:

[0119] By formula: Calculate the path priority value PRI, where SZ is the total distance of the drone inspection path, S1 r S2 is the distance from the r-1th deviation correction area to the rth deviation correction area on the UAV inspection path, r is the distance within the rth deviation correction area on the UAV inspection path, B1 r B2 is the change value of the drone’s entry correction when it enters the rth deviation correction area on the drone inspection path. r is the exit correction change value of the drone when it leaves the rth deviation correction area on the drone inspection path. The value of r is 1, 2, 3, ..., K, K is the total number of deviation correction areas on the drone inspection path, α1 is the preset first hyperparameter, α2 is the preset second hyperparameter, and both are greater than 0. e is a natural constant.

[0120] Exemplarily, the process of obtaining the entry correction change value and the exit correction change value is as follows:

[0121] Based on the correction vector of the UAV before entering the r-th deviation correction area and the correction vector after entering the r-th deviation correction area, the correction change of the UAV is obtained and marked as the entry correction change value;

[0122] Based on the correction vector of the UAV before leaving the r-th deviation correction area and the correction vector after leaving the r-th deviation correction area, a correction change of the UAV is obtained, which is marked as an exit correction change value;

[0123] It needs to be explained that represents the first product corresponding to the rth deviation correction area, The first eigenvalue corresponding to the r-th deviation correction area is represented by , and the first eigenvalue is used as the weight value. The longer the distance on the drone inspection path that is not corrected before entering the r-th deviation correction area is, the smaller the corresponding weight value is;

[0124] It should be further explained that the path priority value combines the actual length of the drone inspection path and the changes in the drone's behavior in the deviation correction area. By introducing weight values and eigenvalues, the performance of the drone on different paths is quantitatively evaluated. A smaller path priority value means that the path has better overall performance after considering deviation correction, which may be a shorter distance, less correction requirements, or smoother path transition.

[0125] The technical solution of the embodiment of the present invention is mainly as follows: on the basis of the original photovoltaic array positioning optimization system, further introducing modules such as offset point analysis, stable deviation point correction, stable deviation area division and UAV path optimization, thereby realizing the refined management and optimization of the UAV positioning inspection process; through the offset point analysis module, through in-depth mining of historical positioning data, the positioning points with stable deviations are identified, which provides a basis for subsequent corrections; the stable deviation point correction module calculates the correction vectors of these stable deviation points and corrects them, thereby improving the positioning accuracy of the UAV; further, the stable deviation area division module divides the photovoltaic array into regions based on the distribution of stable deviation points and correction vectors, forming different deviation correction areas. This step not only helps to better understand the positioning deviation in the photovoltaic array The difference situation also provides a basis for the subsequent UAV path optimization. The UAV path optimization module optimizes the UAV inspection path according to the regional correction vector of each deviation correction area. By presetting multiple inspection paths and calculating their path priority values, the path with the smallest value is finally selected as the optimized inspection path. The optimization process comprehensively considers the actual length of the UAV inspection path and the behavior changes of the UAV in the deviation correction area, thereby ensuring that the UAV can perform inspections in a more efficient and accurate manner; the present invention not only enhances the positioning accuracy and inspection efficiency of the UAV, but also brings more accurate and immediate fault feedback information to the operation and maintenance team. As a result, the operation and maintenance process of the photovoltaic array is greatly optimized, and the fault response is faster, thereby achieving fewer correction requirements during the operation and maintenance process, and effectively improving the overall operation and maintenance efficiency.

[0126] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An optimization system for photovoltaic array positioning, characterized in that: include: RFID information management and identification module: reads the information of RFID tags to achieve digital management and tracking of photovoltaic modules; UWB drone positioning module: Using a drone equipped with a UWB positioning tag, when the drone is inspecting over the photovoltaic array, the distance between the drone and each UWB positioning base station is measured, and the two-dimensional coordinates of the drone are calculated in real time to achieve precise positioning of the drone; Drone inspection and anomaly detection module: UAVs equipped with detection equipment can capture high-definition infrared images of photovoltaic arrays to detect the temperature distribution of components; Ground station monitoring and operation and maintenance feedback module: Receives abnormal images and fault point detection coordinates sent by drones and locates faulty components; Offset point analysis module: Based on the historical positioning data of each positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point; Based on any historical deviation vector of the positioning point, mark the historical deviation vector as the target deviation vector; Compare and analyze the target deviation vector with all other historical deviation vectors, obtain the Euclidean distance between the target deviation vector and the historical deviation vector, and evaluate the similarity between the target deviation vector and the historical deviation vector. Based on the evaluation results, mark the historical deviation vector as a similar deviation vector. Mark the historical deviation vector and all corresponding similar deviation vectors as a positioning point similar deviation vector group, and calculate the similarity representation value of the positioning point similar deviation vector group; Preset a similarity representation threshold, and compare and analyze the similarity representation value with the similarity representation threshold; If the similarity representation value is less than or equal to the similarity representation threshold, the positioning point is marked as a stable deviation point, and at the same time, the positioning point similarity deviation vector group corresponding to the similarity representation value is marked as a positioning point stable deviation vector group; The process of obtaining the similarity representation value is as follows: By formula: The similarity representation value EX is calculated, where F is the total number of all historical deviation vectors in the similar deviation vector group of the positioning point, and M is the total number of all historical deviation vectors. is the average value of the Euclidean distance of all historical deviation vectors in the similar deviation vector group of the positioning point, is the average value of the Euclidean distances corresponding to all historical deviation vectors of the positioning point; Stable deviation point correction module: When a positioning point is marked as a stable deviation point, the correction vector of the stable deviation point is calculated to correct it; Stable deviation area division module: Based on the distribution of stable deviation points in the photovoltaic array and the correction vectors of each stable deviation point, the photovoltaic array is divided into regions to form different deviation correction areas; UAV path optimization module: obtains the regional correction vector of each deviation correction area and optimizes the UAV inspection path.

2. The optimization system for photovoltaic array positioning according to claim 1, characterized in that: Based on the historical positioning data of each positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point. The process is as follows: Based on the UWB UAV positioning module, the historical positioning data of the UAV is obtained, and based on the historical positioning data of the UAV, all historical deviation vectors of each positioning point are obtained; Based on all historical deviation vectors of the positioning point, the deviation regularity of the positioning point is analyzed, and based on the analysis results, the positioning point is marked as a stable deviation point.

3. The optimization system for photovoltaic array positioning according to claim 1, characterized in that: The process of obtaining the Euclidean distance between the target deviation vector and the historical deviation vector to evaluate the similarity between the target deviation vector and the historical deviation vector and marking the historical deviation vector as a similar deviation vector based on the evaluation result is as follows: By formula: Calculate the Euclidean distance D, where A i is the i-th component of the target deviation vector, B i is the i-th component of the historical deviation vector, i is 1, 2, 3, ..., n, and n is the dimension of the target deviation vector and the historical deviation vector; Preset the Euclidean distance threshold, and compare and analyze the Euclidean distance with the Euclidean distance threshold; If the Euclidean distance is less than or equal to the Euclidean distance threshold, the historical deviation vector corresponding to the Euclidean distance is marked as a similar deviation vector.

4. The optimization system for photovoltaic array positioning according to claim 1, characterized in that: Based on the distribution of stable deviation points in the photovoltaic array and the correction vectors of each stable deviation point, the photovoltaic array is divided into regions to form different deviation correction regions. The process is as follows: Based on any stable deviation point, extract the stable deviation points adjacent to the stable deviation point, mark them as first points to be clustered, calculate the cluster characterization values of the stable deviation points and the first points to be clustered, and then evaluate whether the stable deviation points and the first points to be clustered are clustered based on the cluster characterization values, and obtain a first clustering area based on the evaluation result; Extracting stable deviation points adjacent to the first clustering region and marking them as second points to be clustered, evaluating whether the stable deviation points and the second points to be clustered are clustered, and obtaining a third clustering region based on the evaluation result; Repeat the above clustering process until all adjacent stable deviation points in the obtained a-th cluster area cannot be clustered, and mark the a-th cluster area as the deviation correction area.

5. The optimization system for photovoltaic array positioning according to claim 4, characterized in that: The process of obtaining cluster characterization values is as follows: By formula: The cluster characterization value Km is calculated, where L is the cluster distance, P is the cluster deviation, and ΔL is the average distance between all positioning points.

6. The optimization system for photovoltaic array positioning according to claim 4, characterized in that: Based on the cluster characterization value, the process of evaluating whether the stable deviation point and the first point to be clustered are clustered and obtaining the first cluster area based on the evaluation result is as follows: Preset the cluster characterization threshold, and compare and analyze the cluster characterization value with the cluster characterization threshold; If the cluster characterization value is less than or equal to the cluster characterization threshold, the stable deviation point and the first point to be clustered are divided into a first cluster area.

7. The optimization system for photovoltaic array positioning according to claim 6, characterized in that: The process of obtaining the regional correction vector is: The correction vectors of all stable deviation points in the deviation correction area are extracted and averaged to obtain the regional correction vector.

8. The optimization system for photovoltaic array positioning according to claim 6, characterized in that: The process of obtaining the path priority value is as follows: By formula: Calculate the path priority value PRI, where SZ is the total distance of the drone inspection path, S1 r S2 is the distance from the r-1th deviation correction area to the rth deviation correction area on the UAV inspection path, r is the distance within the rth deviation correction area on the UAV inspection path, B1 r B2 is the change value of the drone’s entry correction when it enters the rth deviation correction area on the drone inspection path. r is the exit correction change value of the drone when it leaves the rth deviation correction area on the drone inspection path. The value of r is 1, 2, 3, ..., K, K is the total number of deviation correction areas on the drone inspection path, α1 is the preset first hyperparameter, α2 is the preset second hyperparameter, and both are greater than 0. e is a natural constant.

Images