An intelligent identification system for hot spot defects in photovoltaic modules

Abstract

This invention proposes an intelligent identification system for hot spot defects in photovoltaic modules, belonging to the technical field of unmanned aerial vehicle inspection of photovoltaic modules. The system includes a rotary-wing drone for inspecting the area where the photovoltaic modules are located; an infrared camera, mounted on the drone and operatively connected to the drone, for acquiring infrared images of the working area of ​​the photovoltaic modules; a path planning unit, mounted on the drone and using an improved butterfly optimization algorithm to fit the drone's optimal cruising path; and a hot spot defect identification unit, mounted on the drone and acquiring infrared images from the infrared camera. The system then uses a pre-trained improved GB-YOLOv8 model to detect hot spots in the infrared images and output hot spot information for the photovoltaic modules. By planning the drone's path, infrared images of different parts of the photovoltaic module can be better acquired, and the improved GB-YOLOv8 model can be used to identify and output hot spots.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle inspection of photovoltaic modules, and in particular to an intelligent recognition system for hot spot defects of photovoltaic modules. Background Art

[0002] With the rapid development of deep learning technology and intelligent, unmanned products, including drones, the use of drones equipped with infrared thermal imagers to capture infrared images of photovoltaic modules, followed by deep learning algorithms for defect detection, has gradually become the primary method for photovoltaic module defect detection. For example, the Department of Automation at North China Electric Power University and Zhangjiagang Xunjian Information Technology Co., Ltd. jointly proposed a photovoltaic hot spot detection algorithm based on a feature pyramid fusion high-resolution network. The School of Mechanical and Electrical Engineering at Anhui Jianzhu University also proposed an improved ResNet50 convolutional neural network photovoltaic hot spot detection algorithm to improve recognition accuracy in unevenly distributed samples. These efforts are evident in numerous other areas.

[0003] Traditional methods mainly rely on the capture of shallow features, which limits the ability to obtain deep image information in complex backgrounds and low-contrast environments, thereby affecting the detection accuracy and robustness. At the same time, the existing technology has high requirements for the quality of the input image and large demands on the computing resources of the hardware equipment, which limits its practical application on low-configuration devices. The processing speed is slow and lacks real-time performance, which is particularly prominent in the operation of photovoltaic power stations with high efficiency requirements. In addition, the existing methods are not adaptable enough to different photovoltaic module types and different environmental conditions and cannot meet actual needs. Therefore, an intelligent recognition system for hot spot defects of photovoltaic modules is provided, which can enhance the accurate recognition of hot spot defects of photovoltaic modules, enhance the prediction ability of small targets, reduce the amount of calculation, and improve the operational stability of the photovoltaic system. It has very good application prospects. Summary of the Invention

[0004] In view of this, the present invention proposes a photovoltaic module hot spot defect intelligent identification system that can effectively plan the inspection path of a rotary-wing drone used for photovoltaic module inspection, effectively obtain infrared images of photovoltaic panels at different positions, and reliably identify hot spot defects.

[0005] The present invention provides a photovoltaic module hot spot defect intelligent identification system, comprising:

[0006] Rotary-wing drones are used to inspect the areas where photovoltaic panels are located;

[0007] an infrared camera, mounted on the rotary-wing UAV and movably connected to the rotary-wing UAV, for acquiring infrared images of a working area of ​​the photovoltaic module;

[0008] The path planning unit is installed on the rotor UAV and uses the improved butterfly optimization algorithm to fit the optimal cruising path of the rotor UAV;

[0009] The hot spot defect recognition unit is installed on the rotorcraft drone, obtains the infrared image obtained by the infrared camera, and uses the pre-trained improved GB-YOLOv8 model to detect hot spots in the infrared image, and outputs the hot spot information of the photovoltaic module.

[0010] On the basis of the above technical solutions, preferably, the UAV includes a hollow body, an upper cover, a support rod, a universal joint, a landing gear, several rotor parts and their drive motors; an upper cover is provided on the side of the body away from the ground, and a path planning unit, a hot spot defect recognition unit and a communication device are provided inside the body; several rotor parts extending outward are provided at each vertex of the body, one end of the several rotor parts is fixedly connected to the body, and the other end extends outward in a direction away from the body, and a drive motor is provided at the end of the several rotor parts, and a propeller is provided on the output shaft of the drive motor; a support rod and a landing gear are provided at the end of the body close to the ground, and the support rod and the landing gear are both fixedly connected to the body; a universal joint is provided on the support rod, and the movable end of the universal joint is fixedly connected to the infrared camera; the communication device is used to communicate with the remote control end and transmit the output of the path planning unit and the hot spot defect recognition unit in real time.

[0011] Preferably, the path planning unit uses an improved butterfly optimization algorithm to fit the optimal cruise path of the rotary-wing UAV, and the fitting of the cruise path includes the following steps: 1) creating artificial butterflies and an initial swarm, and during the simulation process, the total number of butterflies remains unchanged; 2) defining the position and fitness function of the butterflies, and in the local search phase, searching for the position generated by random butterflies in the search space, calculating and storing the fitness function of the butterfly position, and then entering the iteration phase; in the iteration phase, solving the updated positions of all butterflies in the search space and re-evaluating the fitness value of the butterflies; 3) when the stopping criteria are met, the butterfly optimization algorithm finds the optimal position and fitness of each butterfly, makes the best decision, and corresponds to the optimal cruise path of the rotary-wing UAV.

[0012] Further preferably, the content of step 2) is to calculate the fitness function of all butterflies at different positions using the following formula: in and denote the solution vectors of the i-th butterfly in the t+1th iteration and the tth iteration respectively; r is a random number in [0, 1]; g * represents the optimal solution among all solutions in the current iteration; f i represents the pollen of the i-th butterfly; in the local search phase, the formula for calculating the fitness function of all butterflies at different positions is: in and are the solution vectors of the j-th butterfly and the k-th butterfly in the t-th iteration, respectively; if the j-th butterfly and the k-th butterfly belong to the same bee colony, the formula for the butterflies producing pollen smells at different positions in the local search phase means local random walk; the position of each butterfly is adjusted by the fitness function to enable the butterfly to search for food and mating partners; the butterfly's search for food and mating partners can be carried out both locally and globally; the switching probability p is used to switch between global search and local search.

[0013] More preferably, the stopping criterion is selected as one of the longest CPU time, the maximum number of iterations reached, the maximum number of iterations without improvement reached, or the specific error rate value reached; when the stopping criterion is met, the iteration stage is stopped, and the improved butterfly optimization algorithm outputs the optimal loop path.

[0014] More preferably, step 3) includes assuming that there is an intermediate point between the source and destination of the path, and outputting the optimal cycle path based on the Euclidean distances of different intermediate points and all paths passing through the intermediate point; wherein the calculation formula for the Euclidean distances of different intermediate points is: p n , p n+1 are two adjacent intermediate points in the path, n = 0, 1, 2, ..., L-1, L is the number of intermediate points; the sum of the lengths of all paths passing through the intermediate points, pl, is calculated by the following formula: It represents the sum of the path lengths of the first n-1 adjacent intermediate points. If there is no intermediate point, the sum of all path lengths pl = 0. The basis for making the optimal decision is: e is a measure of energy expenditure; OP n Represents the operating power of each intermediate point; is the cumulative value of energy consumption corresponding to the path of the first n-1 adjacent intermediate points; d(p n ,p n+1 )>OP n When the energy consumption of the rotorcraft to reach the next intermediate point is greater than the operating power of the next intermediate point, 1-[d(p n ,p n+1 )-OP n The result of the item ] is used as a penalty item. If the energy consumption of the rotorcraft to reach the next intermediate point is not greater than the operating power of the next intermediate point, the energy consumption metric e is assigned a value of 1; this prevents the rotorcraft from choosing a path whose energy consumption is greater than the operating power.

[0015] Once the position of the new midpoint is found based on the Euclidean distance and energy consumption metrics of different midpoints, the path quality is evaluated based on the number of obstacles in the path and the collision impact criterion: if the distance between the point on the path and the photovoltaic panel is less than the set photovoltaic module neighborhood radius, it means that the rotorcraft on the path will collide with the photovoltaic module. This length is added to the collision length calculated by the path, and the number of collisions is accumulated. Otherwise, it means that the rotorcraft on the path will not collide with the photovoltaic module, and the number of collisions will not increase. Similarly, the shortest path without collision with the photovoltaic module is obtained as the optimal cruise path.

[0016] On the basis of the above technical solution, preferably, the working content of the hot spot defect recognition unit is: obtaining the original data set of infrared images of the working area of ​​the photovoltaic module taken by the infrared camera, extracting the infrared images of the photovoltaic panels with hot spot images in the original data set, using the Labelme standard tool to mark the infrared images with hot spots, and using the infrared images with the hot spot areas marked as training data sets; allocating the infrared images in the training data set into training set and validation set in a ratio of 9:1; using the training set to train the improved GB-YOLOv8 model, and using the validation set to test the improved GB-YOLOv8 model, to obtain the pre-trained improved GB-YOLOv8 model for predicting hot spot defects in subsequently input infrared images.

[0017] Preferably, the improved GB-YOLOv8 model includes three parts: Backbone, Neck and Head; the Backbone part introduces several lightweight convolutional layers C2f-Ghost; the Neck part introduces several bidirectional feature pyramid network structures Concat_BiFPN, and the several bidirectional feature pyramid network structures Concat_BiFPN combine bidirectional cross-scale connections and weighted feature fusion, and are connected to the lightweight convolutional layer C2f-Ghost of the Backbone part on the one hand, and to the first C2f layer, Upsample layer or Conv layer of the preamble part of the Neck part on the other hand; the Head part samples four different small target detection heads Decet, and the four different small target detection heads Decet are respectively connected to different second C2f layers of the Neck part in a one-to-one correspondence; the first C2f layer is located on the input side of the preamble part of the several bidirectional feature pyramid network structures Concat_BiFPN; the second C2f layer is located on the output side of the several bidirectional feature pyramid network structures Concat_BiFPN.

[0018] Preferably, the hot spot defect recognition unit is further provided with preset indicators for evaluating the prediction results of the hot spot defects of the improved GB-YOLOv8 model, and the preset indicators are one or more combinations of precision rate, recall rate and average precision mean; let the prediction results of hot spots be positive samples, and the prediction results of not hot spots be negative samples, TP is the number of correct predictions of positive samples, FP is the number of incorrect predictions of positive samples; TN is the number of correct predictions of negative samples, and FN is the number of incorrect predictions of negative samples; precision rate Recall Mean Average Precision Where k is the number of categories; AP m is the mean precision, m=1,2,...,k.

[0019] Further preferably, after the hot spot defect recognition unit uses the improved GB-YOLOv8 model to predict the hot spot defect in the infrared image, it also includes an RGB image recognition process of the hot spot, and calculates the color dispersion S by extracting the RGB channel values ​​of the hot spot image according to the following formula: Y R 、Y G and Y B are the R, G, and B channel values ​​of the image of the hot spot identified by the improved GB-YOLOv8 model; min is the minimum value of the R channel; when the dispersion S of the pixels in the hot spot image is greater than 0.2, and the R channel threshold Y R ∈(230, 255), the pixel is considered to be a suspected hot spot. After color dispersion detection of each pixel in the infrared image, the pixel suspected of the hot spot is binarized to determine whether the pixel suspected of the hot spot is consistent with the prediction result of the improved GB-YOLOv8 model.

[0020] The photovoltaic module hot spot defect intelligent identification system provided by the present invention has the following beneficial effects compared with the prior art:

[0021] (1) The present invention provides a method for arranging a universal joint and an infrared camera on a rotary-wing UAV, which can realize multi-axial motion and flexibly perform high-resolution framing of the entire or partial photovoltaic modules. Compared with traditional fixed-angle framing, the method has the advantages of being more flexible and convenient.

[0022] (2) An optimized butterfly algorithm is used, and the balance relationship between the energy consumption measurement and the operating power of the path between the intermediate points is further introduced. Penalties are added to non-compliant items to prevent the rotorcraft from choosing a path with energy consumption greater than the operating power, thus maintaining the rationality of the planned path selection;

[0023] (3) The improved GB-YOLOv8 model has made adaptive improvements to the model structure and adopted four different small target detection heads, which has increased the detection speed and optimized the real-time detection capability of the model. It has a good detection effect on both large and small targets and can improve the recognition ability of photovoltaic module hot spots. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 This is a flowchart of the workflow of a photovoltaic module hot spot defect intelligent identification system according to the present invention;

[0026] Figure 2 A stereoscopic diagram of the combined state of a rotary-wing drone and an infrared camera in a photovoltaic module hot spot defect intelligent identification system according to the present invention;

[0027] Figure 3 A three-dimensional diagram of an infrared camera and a support assembly of an intelligent identification system for hot spot defects of photovoltaic modules according to the present invention;

[0028] Figure 4 This is a network structure diagram of an improved GB-YOLOv8 model of a photovoltaic module hot spot defect intelligent identification system according to the present invention;

[0029] Figure 5 This is an example of identifying hot spot images of photovoltaic modules using an intelligent hot spot defect recognition system for photovoltaic modules according to the present invention.

[0030] Figure numerals: 1. Main body; 2. Upper cover; 3. Support rod; 4. Universal joint; 5. Landing gear; 6. Rotor; 7. Drive motor; 8. Propeller; 10. Infrared camera; 101. Infrared lens; 102. Lens protector; 103. First wiring port; 104. Housing; 105. Second wiring port; 106. Support frame; 41. First clamping plate of universal joint; 42. Second clamping plate of universal joint. DETAILED DESCRIPTION

[0031] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described 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.

[0032] During the inspection of photovoltaic modules, the infrared images collected are usually fixed in angle and large in size, which is not suitable for intelligent real-time inspection of small target defects in photovoltaic modules. In addition, the traditional detection method manually extracts features such as grayscale features and then classifies them based on these features. This method can only extract shallow features, has high requirements for image data, low efficiency and poor adaptability, and cannot meet the real-time and intelligent requirements of production. In view of this, Figure 1 As shown, the present invention provides a photovoltaic module hot spot defect intelligent identification system, comprising:

[0033] Rotary-wing drones are used to inspect the areas where photovoltaic panels are located;

[0034] an infrared camera 10, which is mounted on the rotary-wing UAV and movably connected to the rotary-wing UAV and is used to obtain infrared images of the working area of ​​the photovoltaic module;

[0035] The path planning unit is installed on the rotor UAV and uses the improved butterfly optimization algorithm to fit the optimal cruising path of the rotor UAV;

[0036] The hot spot defect recognition unit is installed on the rotorcraft drone, obtains the infrared image obtained by the infrared camera, and uses the pre-trained improved GB-YOLOv8 model to detect hot spots in the infrared image, and outputs the hot spot information of the photovoltaic module.

[0037] Combine Figure 2 It can be seen that the rotor UAV includes a hollow body 1, an upper cover 2, a support rod 3, a universal joint 4, a landing gear 5, a number of rotor parts 6 and their drive motors 7; an upper cover is provided on the side of the body away from the ground, and a path planning unit, a hot spot defect recognition unit and a communication device are provided inside the body 1; a number of rotor parts 6 extending outward are provided at each vertex of the body 1, one end of the number of rotor parts 6 is fixedly connected to the body 1, and the other end extends outward in a direction away from the body 1, and a drive motor 7 is provided at the end of the number of rotor parts 6, and a propeller 8 is provided on the output shaft of the drive motor 7; a support rod 3 and a landing gear 5 are provided at the end of the body 1 close to the ground, and the support rod 3 and the landing gear 5 are both fixedly connected to the body; a universal joint 4 is provided on the support rod 3, and the movable end of the universal joint 4 is fixedly connected to the infrared camera 10; the communication device is used to communicate with the remote control end and transmit the outputs of the path planning unit and the hot spot defect recognition unit in real time. Figure 2As shown, the area between the main body 1 and the upper cover 2 is used to protect the internal path planning unit, hot spot defect identification unit and communication device. Before the inspection begins, ensure that all components of the rotor UAV are installed intact, start each drive motor 7, and calculate the flight path of the rotor UAV through the improved butterfly optimization algorithm to cover the photovoltaic component area that needs to be inspected. The rotor UAV uses a universal joint 4 that can adjust the posture of the infrared camera 10 to ensure that when the rotor UAV flies in the area where the photovoltaic components are located, the infrared camera 10 can capture images of photovoltaic panels in different positions. When the rotor UAV completes the inspection, the rotor UAV returns to the starting point and lands smoothly, ending the inspection mission. The landing gear 5 is used to ensure that the rotor UAV takes off or lands smoothly.

[0038] As a preferred embodiment, Figure 3 As shown, the infrared camera 10 includes an infrared lens 101, a lens protector 102, a first wiring port 103, a housing 104, a second wiring port 105, and a support frame 106. The infrared lens 101 is mounted at one end of the hollow housing 104, and a waterproof and dustproof lens protector 102 is positioned in the viewing direction of the infrared lens 101. The first and second wiring ports 103 and 105 are provided through the housing 104 for connecting to an external power source and for outputting infrared images to a hot spot defect recognition unit. The infrared camera 10 has a resolution of at least 640×480.

[0039] Universal joint 4 includes a first universal joint plate 41, a second universal joint plate 42, and a pitch drive mechanism. The first universal joint plate 41 is pivotally connected to the end of the main body 1 closest to the ground. The first and second universal joint plates 41, 42 are hingedly connected. A pitch drive mechanism is provided at the hinged joint, which drives the second universal joint plate 42 to pitch and adjust a predetermined angle relative to the first universal joint plate 41. The second universal joint plate 42 is also fixedly connected to the housing 104 of the infrared camera 10 via a support frame 106. Universal joint 4 allows for a wide range of posture adjustments for the infrared camera 10, offering greater flexibility and convenience compared to traditional fixed-angle framing.

[0040] In terms of path planning, since drone inspections must account for the impact of three-dimensional terrain on drone flight, an improved butterfly optimization algorithm was developed in this paper. Unlike previous path planning algorithms, the butterfly optimization algorithm is primarily used for obstacle avoidance during drone flight. The first step of the butterfly optimization algorithm is to use a greedy approach to obtain the optimal values ​​of several key parameters, such as population size, maximum number of iterations, number of butterflies, and number of samples in each iteration. Next, initial points are randomly and uniformly distributed throughout the environment. The agent generates many intermediate points and guides the drone to new locations based on path fitting and step size parameters. In the early stages of path planning, due to the small step size, choosing a poor path will result in a lighter penalty. As the destination is approached, a more suitable path is desired, and because fewer options are available, choosing an incorrect path will result in a more severe penalty.

[0041] Based on the above ideas, the path planning unit in this application uses an improved butterfly optimization algorithm to fit the optimal cruise path of the rotorcraft UAV. The fitting of the cruise path includes the following steps: 1) creating artificial butterflies and creating an initial swarm. During the simulation, the total number of butterflies remains unchanged; 2) defining the position and fitness function of the butterfly. In the local search phase, the position generated by the random butterfly is searched in the search space, the fitness function of the butterfly position is calculated and stored, and then the iteration phase is entered; in the iteration phase, the updated positions of all butterflies in the search space are solved, and the fitness value of the butterfly is re-evaluated; 3) when the stopping criteria are met, the butterfly optimization algorithm finds the optimal position and fitness of each butterfly, makes the best decision, and corresponds to the optimal cruise path of the rotorcraft UAV.

[0042] The content of step 2) is to calculate the fitness function of all butterflies at different positions using the following formula: in and denote the solution vectors of the i-th butterfly in the t+1th iteration and the tth iteration respectively; r is a random number in [0, 1]; g * represents the optimal solution among all solutions in the current iteration; f i represents the pollen of the i-th butterfly; in the local search phase, the formula for calculating the fitness function of all butterflies at different positions is: in and are the solution vectors of the j-th butterfly and the k-th butterfly in the t-th iteration, respectively; if the j-th butterfly and the k-th butterfly belong to the same bee colony, the formula for the butterflies producing pollen smells at different positions in the local search phase means local random walk; the position of each butterfly is adjusted by the fitness function to enable the butterfly to search for food and mating partners; the butterfly's search for food and mating partners can be carried out both locally and globally; the switching probability p is used to switch between global search and local search.

[0043] The butterfly optimization algorithm involves global and local search steps for the photovoltaic panels. To select the global optimal path and avoid local optimal paths, an agent is required to make intelligent decisions based on path fitting. This intelligent agent removes erroneous points and controls the drone as needed. This agent helps the drone perform path fitting and path selection for photovoltaic panel inspections, avoiding local optimal paths. The agent here is the switching probability p, a hyperparameter.

[0044] The stopping criterion in step 3) is selected as one of the longest CPU time, the maximum number of iterations reached, the maximum number of iterations without improvement reached, or the specific error rate value reached; when the stopping criterion is met, the iteration phase is stopped, and the improved butterfly optimization algorithm outputs the optimal loop path.

[0045] The specific content of step 3) is to assume that there is an intermediate point between the source and destination of the path, and output the optimal cycle path based on the Euclidean distance of different intermediate points and all paths passing through the intermediate point; where the calculation formula of the Euclidean distance of different intermediate points is: P n , p n+1 are two adjacent intermediate points in the path, n = 0, 1, 2, ..., L-1, L is the number of intermediate points; the sum of the lengths of all paths passing through the intermediate points, pl, is calculated by the following formula: It represents the sum of the path lengths of the first n-1 adjacent intermediate points. If there is no intermediate point, the sum of all path lengths pl = 0. The basis for making the optimal decision is: e is a measure of energy expenditure; OP n Represents the operating power of each intermediate point; is the cumulative value of energy consumption corresponding to the path of the first n-1 adjacent intermediate points; d(p n , p n+1 )>OPn indicates that the energy consumption of the rotorcraft to reach the next intermediate point is greater than the operating power of the next intermediate point. n , p n+1 )-OP n] is used as the penalty item. If the energy consumption of the rotorcraft when reaching the next intermediate point is not greater than the operating power of the next intermediate point, the energy consumption metric e is assigned a value of 1. The purpose of the decision is to prevent the rotorcraft from choosing a path whose energy consumption is greater than the operating power.

[0046] Once the new midpoint location is found based on the Euclidean distance and energy consumption metrics of different midpoints, the path quality is evaluated based on the number of obstacles in the path and the collision impact criterion: If the distance between a point on the path and the photovoltaic panel is less than the set PV panel neighborhood radius, it indicates that the rotorcraft on the path will collide with the PV panel. This distance is added to the collision length calculated for the path as the lower limit reference value of the path length, indicating that the rotorcraft will collide with the PV panel at this path length and the collision count will be accumulated. Otherwise, the rotorcraft on the path will not collide with the PV panel and the collision count will not be increased. After making the above judgments for each path between the midpoints, the shortest path without collision with the PV panel is obtained as the optimal cruise path. These three steps constitute the complete content of the butterfly optimization algorithm.

[0047] As a preferred embodiment, the working content of the hot spot defect recognition unit in the present invention is: obtaining the original data set of infrared images of the working area of ​​the photovoltaic module taken by the infrared camera, extracting the infrared images of the photovoltaic panels with hot spot images in the original data set, using the Labelme standard tool to mark the infrared images with hot spots, and using the infrared images with the hot spot areas marked as training data sets; allocating the infrared images in the training data set into training set and validation set in a ratio of 9:1; using the training set to train the improved GB-YOLOv8 model, and using the validation set to test the improved GB-YOLOv8 model, to obtain the pre-trained improved GB-YOLOv8 model for predicting hot spot defects in subsequently input infrared images.

[0048] refer to Figure 4In this embodiment, the improved GB-YOLOv8 model includes three parts: Backbone, Neck, and Head. The Backbone part introduces several lightweight convolutional layers C2f-Ghost. The Neck part introduces several bidirectional feature pyramid network structures Concat_BiFPN. The several bidirectional feature pyramid network structures Concat_BiFPN combine bidirectional cross-scale connections and weighted feature fusion. On the one hand, they are connected to the lightweight convolutional layer C2f-Ghost of the Backbone part, and on the other hand, they are connected to the first C2f layer, Upsample layer, or Conv layer of the front part of the Neck part. The Head part samples four different small target detection heads Decet, and the four different small target detection heads Decet are respectively connected to different second C2f layers of the Neck part in a one-to-one correspondence. The first C2f layer is located at the input side of the front part of the several bidirectional feature pyramid network structures Concat_BiFPN. The second C2f layer is located at the output side of the several bidirectional feature pyramid network structures Concat_BiFPN.

[0049] Figure 4 The left side is the Backbone part. In addition to the input and output parts, the middle link is continuously set with a combination structure of multiple sequentially connected convolutional layers Conv-lightweight convolutional layers C2f-Ghost. Figure 4 The center is the Neck section, which includes several upsampling layers (Upsample), a bidirectional feature pyramid network (Concat_BiFPN), and a C2f layer structure, as well as a convolutional layer (Conv), a bidirectional feature pyramid network (Concat_BiFPN), and a C2f layer structure. Different C2f layers are connected to different small object detection heads (Dectets) in the Head section. The Concat_BiFPN bidirectional feature pyramid network (Concat_BiFPN) combines bidirectional cross-scale connections with weighted feature fusion, enhancing the model's feature extraction capabilities while reducing computational resource consumption. Furthermore, by adjusting weights and fine-tuning the contribution of each scale to the feature fusion network, it accelerates detection and optimizes the model's real-time detection capabilities. The head uses four different small target detection heads, Dectets, which are more suitable for infrared image analysis tasks. The traditional YOLOv8 model only has three detection heads, Dectets, and their detection feature sizes are: 20×20, 40×40, and 80×80, respectively. These heads have good detection effects on large targets, but poor detection effects on small targets. Therefore, a new detection feature size of 160×160 is added to detect targets larger than 4×4, improving detection accuracy.

[0050] As a preferred embodiment, the hot spot defect recognition unit is further provided with preset indicators for evaluating the prediction results of the hot spot defects of the improved GB-YOLOv8 model. The preset indicators are one or more combinations of precision rate, recall rate and average precision mean; let the prediction results of hot spots be positive samples, and the prediction results of not hot spots be negative samples, TP is the number of correct positive sample predictions, FP is the number of incorrect positive sample predictions; TN is the number of correct negative sample predictions, and FN is the number of incorrect negative sample predictions; precision rate Recall Mean Average Precision Where k is the number of categories; AP m is the mean precision, m = 1, 2, ..., k.

[0051] As a further improvement to this solution, the hot spot defect recognition unit, after using the improved GB-YOLOv8 model to predict hot spot defects in infrared images, also includes an RGB image recognition process for hot spots. By extracting the RGB channel values ​​of the hot spot image, the color dispersion S is calculated according to the following formula: Y R 、Y G and Y B are the R, G, and B channel values ​​of the image of the hot spot identified by the improved GB-YOLOv8 model; min is the minimum value of the R channel; when the dispersion S of the pixels in the hot spot image is greater than 0.2, and the R channel threshold Y R ∈(230, 255), the pixel is considered to be a suspected hot spot. After the color dispersion of each pixel in the infrared image is detected, the pixel suspected of the hot spot is binarized to determine whether the pixel suspected of the hot spot is consistent with the result predicted by the improved GB-YOLOv8 model. The above color dispersion is to avoid the situation where the R channel is within the threshold but is not actually a hot spot. If the RGR value is (255, 255, 255), S=0, then the R channel is within the hot spot detection threshold, but the pixel is actually detected as pure white, that is, a false detection occurs. Therefore, S>0.2 and Y R ∈(230, 255). The suspected hot spot pixels obtained from the above screening can be further processed through Gaussian filtering, image dilation, erosion, and other operations to binarize the image based on the suspected hot spot pixels and other pixels. The improved GB-YOLOv8 model is then used to detect and identify hot spots.

[0052] The communication device mentioned above can receive hot spot identification data fed back by the rotorcraft UAV at the remote control end. Using PyQt5 and MySQL database combined with an improved GB-YOLOv8 model, it can realize the collection, identification, and visual output of hot spot images of the rotorcraft UAV's current location, and better alarm prompts for logged-in users, making it easier for staff to detect hot spots in a timely manner and provide data support for replacement and maintenance of corresponding photovoltaic modules.

[0053] In one example, the improved GB-YOLOv8 model and the current mainstream target detection model YOLOv8 were used to simultaneously identify infrared hot spots on photovoltaic modules in the same environment. The experimental environment is shown in Table 1:

[0054] Table 1 Experimental environment

[0055]

[0056]

[0057] The inspection results of the light spots of photovoltaic modules by traditional YOLOv8 and the improved GB-YOLOv8 adopted in this invention are shown in Table 2:

[0058]

[0059] As shown in Table 2, the improved GB-YOLOv8 model used in this invention has 56.7% of the parameters of the YOLOv8 model, 81.18% of the computational complexity, and 52.24% of the model size compared with the YOLOv8 model. The recall rate is improved by 3.2%, the recall rate is improved by 3.33%, and the mAP is improved by 3.45%. The specific detection images are as follows: Figure 5 As shown in the figure, the improved GB-YOLOv8 not only improves detection accuracy, but also reduces the number of parameters, computational complexity, and model size. It has strong applicability and can be applied to the real-time monitoring of infrared hot spots in photovoltaic modules.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A photovoltaic module hot spot defect intelligent identification system, characterized in that: include: Rotary-wing drones are used to inspect the areas where photovoltaic panels are located; an infrared camera, mounted on the rotary-wing UAV and movably connected to the rotary-wing UAV, for acquiring infrared images of a working area of ​​the photovoltaic module; The path planning unit is installed on the rotor UAV and uses the improved butterfly optimization algorithm to fit the optimal cruising path of the rotor UAV; The hot spot defect recognition unit is installed on the rotorcraft drone. It obtains the infrared image obtained by the infrared camera and uses the pre-trained improved GB-YOLOv8 model to detect hot spots in the infrared image and output the hot spot information of the photovoltaic module. The path planning unit uses an improved butterfly optimization algorithm to fit the optimal cruise path of the rotorcraft. The cruise path fitting includes the following steps: 1) creating artificial butterflies and an initial swarm. During the simulation, the total number of butterflies remains unchanged; 2) Define the butterfly position and fitness function. In the local search phase, find the position of the randomly generated butterfly in the search space, calculate and store the fitness function of the butterfly position, and then enter the iteration phase. In the iterative phase, the updated positions of all butterflies in the search space are solved and the fitness values ​​of the butterflies are re-evaluated. 3) When the stopping criteria are met, the butterfly optimization algorithm finds the best position and fitness of each butterfly and makes the best decision, which corresponds to the optimal cruising path of the rotorcraft. In step 3), an intermediate point is assumed to exist between the source and destination of the path. The optimal cruise path is output based on the Euclidean distances of different intermediate points and all paths passing through the intermediate point. The energy consumption metric is used as the basis for determining the best decision. Once the position of the new intermediate point is found based on the Euclidean distances of different intermediate points and the energy consumption metric, the path quality is evaluated based on the number of obstacles in the path and the collision impact criterion: if the distance between a point on the path and the photovoltaic panel is less than the set photovoltaic module neighborhood radius, it means that the rotorcraft on the path has collided with the photovoltaic module. This length is added to the collision length calculated for the path, and the number of collisions is accumulated. Otherwise, it means that the rotorcraft on the path will not collide with the photovoltaic module, and the number of collisions will not increase. Similarly, the shortest path without collision with the photovoltaic module is obtained as the optimal cruise path.

2. The photovoltaic module hot spot defect intelligent identification system according to claim 1, characterized in that: The UAV includes a hollow body, an upper cover, a support rod, a universal joint, a landing gear, several rotor parts and their drive motors; an upper cover is provided on the side of the body away from the ground, and a path planning unit, a hot spot defect recognition unit and a communication device are provided inside the body; several rotor parts extending outward are provided at each vertex of the body, one end of the several rotor parts is fixedly connected to the body, and the other end extends outward in a direction away from the body, and a drive motor is provided at the end of the several rotor parts, and a propeller is provided on the output shaft of the drive motor; a support rod and a landing gear are provided at the end of the body close to the ground, and the support rod and the landing gear are both fixedly connected to the body; a universal joint is provided on the support rod, and the movable end of the universal joint is fixedly connected to the infrared camera; the communication device is used to communicate with the remote control end and transmit the output of the path planning unit and the hot spot defect recognition unit in real time.

3. The photovoltaic module hot spot defect intelligent identification system according to claim 1, characterized in that: Step 2) is to calculate the fitness function of all butterflies at different positions using the following formula: in and denote the solution vectors of the i-th butterfly in the t+1th iteration and the tth iteration respectively; r is a random number in [0, 1]; g * represents the optimal solution among all solutions in the current iteration; f i represents the pollen of the i-th butterfly; in the local search phase, the formula for calculating the fitness function of all butterflies at different positions is: in and are the solution vectors of the jth butterfly and the kth butterfly in the tth iteration respectively; If the j-th butterfly and the k-th butterfly belong to the same bee colony, the formula for the butterflies producing pollen smells at different locations in the local search phase means local random walk; the position of each butterfly is adjusted by the fitness function to enable the butterfly to search for food and mating partners; the butterfly's search for food and mating partners can be carried out both locally and globally; the switching probability p is used to switch between global search and local search.

4. The photovoltaic module hot spot defect intelligent identification system according to claim 1, characterized in that: The stopping criterion is to select one of the longest CPU time, the maximum number of iterations reached, the maximum number of iterations without improvement, or the specific error rate value reached as the stopping criterion; when the stopping criterion is met, the iteration stage is stopped and the improved butterfly optimization algorithm outputs the optimal cruise path.

5. The photovoltaic module hot spot defect intelligent identification system according to claim 1, characterized in that: In step 3, the Euclidean distance between different midpoints is calculated as follows: p n ,p n+1 are two adjacent midpoints in the path, n = 0, 1, 2, ..., L-1, L is the number of midpoints; the sum of the lengths of all paths passing through the midpoints, pl, is calculated using the following formula: It represents the sum of the path lengths of the first n-1 adjacent intermediate points. If there is no intermediate point, the sum of all path lengths pl = 0. The basis for making the optimal decision is: e is a measure of energy expenditure; OP n Represents the operating power of each intermediate point; is the cumulative value of energy consumption corresponding to the path of the first n-1 adjacent intermediate points; d(p n , p n+1 )>OP n When the energy consumption of the rotorcraft to reach the next intermediate point is greater than the operating power of the next intermediate point, 1-[d(p n , p n+1 )-OP n The result of the item ] is used as a penalty item. If the energy consumption of the rotorcraft to reach the next intermediate point is not greater than the operating power of the next intermediate point, the energy consumption metric e is assigned a value of 1; Prevent rotary-wing drones from choosing paths where energy consumption is greater than their operating power.

6. The photovoltaic module hot spot defect intelligent identification system according to claim 1, characterized in that: The work content of the hot spot defect recognition unit is as follows: obtain the original data set of infrared images of the working area of ​​the photovoltaic module taken by the infrared camera, extract the infrared images of the photovoltaic panels with hot spot images in the original data set, use the Labelme standard tool to annotate the infrared images with hot spots, and use the infrared images with the hot spot areas annotated as the training data set; divide the infrared images in the training data set into training set and validation set in a ratio of 9:1; use the training set to train the improved GB-YOLOv8 model, and use the validation set to test the improved GB-YOLOv8 model, and obtain the pre-trained improved GB-YOLOv8 model for predicting hot spot defects in subsequently input infrared images.

7. The photovoltaic module hot spot defect intelligent identification system according to claim 6, characterized in that: The improved GB-YOLOv8 model consists of three parts: Backbone, Neck and Head; the Backbone part introduces several lightweight convolutional layers C2f-Ghost; the Neck part introduces several bidirectional feature pyramid network structures Concat_BiFPN, which combine bidirectional cross-scale connections and weighted feature fusion. On the one hand, they are connected to the lightweight convolutional layer C2f-Ghost of the Backbone part, and on the other hand, they are connected to the first C2f layer, Upsample layer or Conv layer of the preamble part of the Neck part; the Head part samples four different small target detection heads Decet, and the four different small target detection heads Decet are respectively connected one-to-one with the different second C2f layers of the Neck part; the first C2f layer is located on the input side of the preamble part of the several bidirectional feature pyramid network structures Concat_BiFPN; the second C2f layer is located on the output side of the several bidirectional feature pyramid network structures Concat_BiFPN.

8. The photovoltaic module hot spot defect intelligent identification system according to claim 6, characterized in that: The hot spot defect recognition unit is also provided with preset indicators for evaluating the prediction results of the hot spot defects of the improved GB-YOLOv8 model. The preset indicators are one or more combinations of precision rate, recall rate and average precision mean; let the prediction results of hot spots be positive samples, and the prediction results of not hot spots be negative samples, TP is the number of correct positive sample predictions, FP is the number of incorrect positive sample predictions; TN is the number of correct negative sample predictions, and FN is the number of incorrect negative sample predictions; precision rate Recall Mean Average Precision Where k is the number of categories; AP m is the mean precision, m = 1, 2, ..., k.

9. The photovoltaic module hot spot defect intelligent identification system according to claim 8, characterized in that: After using the improved GB-YOLOv8 model to predict hot spot defects in infrared images, the hot spot defect recognition unit also includes an RGB image recognition process for hot spots. By extracting the RGB channel values ​​of the hot spot image, the color dispersion S is calculated according to the following formula: Y R 、Y G and Y B are the R, G, and B channel values ​​of the image of the hot spot identified by the improved GB-YOLOv8 model; min is the minimum value of the R channel; When the dispersion of the pixels in the hot spot image S>0.2, and the R channel threshold Y R ∈(230, 255), the pixel is considered to be a suspected hot spot. After color dispersion detection of each pixel in the infrared image, the pixel suspected of the hot spot is binarized to determine whether the pixel suspected of the hot spot is consistent with the prediction result of the improved GB-YOLOv8 model.

Images