A monitoring and management method and system for photovoltaic power generation

Abstract

The present application relates to the field of photovoltaic power generation monitoring, and in particular to a monitoring and management method and system for photovoltaic power generation, the method comprising the steps of: marking the cell area and the gate area in the grayscale image of the photovoltaic panel to obtain a number of initial windows; screening the initial windows to obtain the optimal window; in the optimal window, calculating the damage evaluation, and when the damage evaluation is greater than 0, generating an alarm signal indicating the presence of damage; wherein the calculation method of the damage evaluation is: constructing the entropy value of the grayscale co-occurrence matrix in multiple directions, taking the maximum entropy value as the texture density value, the direction corresponding to the maximum entropy value is the texture direction of the crack, and each optimal window corresponds to a texture density value; constructing a texture density matrix according to the texture density value, taking any point in the texture density matrix as the center point, and taking the ratio of the texture density mean in the small neighborhood of the center point to the texture density mean in the large neighborhood as the damage evaluation. The present application has the effect of improving the accuracy of damaged area recognition.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic power generation monitoring, and in particular to a photovoltaic power generation monitoring and management method and system. Background Art

[0002] Photovoltaic power generation refers to a method of generating electricity by using solar panels (photovoltaic panels) to convert sunlight into electrical energy. The photovoltaic power generation system is mainly composed of photovoltaic modules, inverters, AC distribution cabinets, etc. The photovoltaic modules generate direct current, and the inverter is responsible for converting direct current into alternating current for use.

[0003] In order to monitor photovoltaic panels, in the prior art, such as the solar monitor disclosed in Chinese patent publication No. CN203574742U, a camera can be set to collect video or image information of the photovoltaic panels.

[0004] By monitoring the photovoltaic panels, it is possible to identify whether the photovoltaic panels are damaged. Damaged photovoltaic panels will cause the photoelectric conversion efficiency to decrease due to the damage of the cells, reducing the power generation. It will also produce hot spot effects, which will reduce the power generation efficiency of the components, accelerate the aging of the components, and even burn the entire component. If the damage causes hidden cracks, it may change the direction of current collection and affect the electrical performance of the cells. In severe cases, it may even cause the cells to be in an independent parallel state, destroying the circuit structure. Therefore, it is necessary to identify the damaged position of the photovoltaic panel in time, cut off the damaged module, and reduce the damage.

[0005] For example, a solar photovoltaic panel surface defect detection method disclosed in a Chinese patent application document with publication number CN114565600A includes: a fault detection module, which detects cracks and bending of the panel surface through Hough line detection, and the fault detection module performs edge detection on damage and dents through a canny edge detection algorithm.

[0006] The Canny algorithm relies on two thresholds to determine the strength of the edge. The low threshold is used to determine the candidate pixels of the edge, and the high threshold is used to finally determine the edge. These two thresholds are usually set manually and need to be adjusted according to the specific application scenario. If the threshold is set improperly, it may lead to poor edge detection, such as missed edge detection or false detection (for example, dust accumulates on the surface of photovoltaic panels during long-term outdoor use, and dust may be mistakenly identified as cracks). Therefore, the existing technology has the technical problem of low accuracy in identifying damaged areas. Summary of the invention

[0007] In order to solve the technical problem of low accuracy in identifying damaged areas, the present application provides a photovoltaic power generation monitoring and management method and system.

[0008] In a first aspect, the present application provides a photovoltaic power generation monitoring and management method, which adopts the following technical solution:

[0009] A monitoring and management method for photovoltaic power generation comprises the following steps: marking a cell region and a grid region in a grayscale image of a photovoltaic panel to obtain a plurality of initial windows; screening the initial windows to obtain an optimal window; calculating a damage evaluation in the optimal window, and generating an alarm signal indicating the presence of damage when the damage evaluation is greater than 0; wherein the damage evaluation is calculated by: constructing entropy values ​​of a grayscale co-occurrence matrix in multiple directions, taking the maximum entropy value as a texture density value, the direction corresponding to the maximum entropy value is the texture direction of a crack, and each optimal window corresponds to a texture density value; constructing a texture density matrix according to the texture density value, taking any point in the texture density matrix as a center point, and taking the ratio of the texture density mean value in a small neighborhood of the center point to the texture density mean value in a large neighborhood as the damage evaluation.

[0010] The beneficial effects are: by constructing a grayscale co-occurrence matrix, the density and directional features of the texture are found. The damage to photovoltaic panels is usually cracks that spread from the center to the surrounding, which is different from dusty textures or reflective textures. This can more accurately find the damaged area and improve the accuracy of damaged area identification.

[0011] Optionally, the initial window is screened to obtain the optimal window, including the steps of: clustering the grayscale values ​​in the initial window to obtain two clustering clusters; verifying the validity of the grayscale image, if valid, calculating the feature resolution index of the initial window, and taking the window corresponding to the minimum feature resolution index as the optimal window, otherwise deleting the grayscale image.

[0012] The beneficial effect is: verifying the validity of the grayscale image is to ensure that the results of the image analysis are of practical significance, that is, to correctly distinguish the cell area and the gate area. If not, it may be due to dust coverage or reflection. At this time, a signal can be sent to the operation and maintenance personnel to prompt them to check the collected image. If there is a lot of dust, the operation and maintenance personnel need to clean the photovoltaic panel. An initial window corresponds to a feature resolution index. The window corresponding to the minimum feature resolution index within a day is taken as the optimal window for that day, which can reduce the impact of reflection on defect identification.

[0013] Optionally, the feature resolution index is calculated as: ,

[0014] ,

[0015] ,in, represents the feature resolution index, Indicates the cell resolution index, represents the gate resolution index, represents the maximum value function; Indicates The exponential function with base , Indicates the first cluster The ratio of pixels to all pixels in the initial window, is the gray value of the first pixel in the first cluster, represents the standard grayscale value of the battery cell, min represents the minimum value function, The first cluster The gray value of each pixel; Indicates the second cluster The ratio of pixels to all pixels in the initial window, is the gray value of the first pixel in the second cluster, Indicates the standard gray value of the gate, The second cluster The gray value of a pixel.

[0016] The beneficial effect is that the smaller the feature resolution index is, the more obvious the feature difference between the battery cell area and the gate area is, thereby improving the image resolution ability, finding the more obvious grayscale images in multiple acquisition drawings within the cycle for analysis, and reducing the misjudgment rate.

[0017] Optionally, the clustering adopts a k-means clustering algorithm, and the number of cluster centers in the cluster is 2.

[0018] Optionally, the standard grayscale value of the cell is: the grayscale value of the cell area in the grayscale image when the photovoltaic panel is installed and used, and the standard grayscale value of the gate is: the grayscale value of the gate area in the grayscale image when the photovoltaic panel is installed and used for the first time.

[0019] The beneficial effect is: using the grayscale value of the cell in the initial grayscale image of the photovoltaic panel as the reference of the cell, and using the grayscale value of the gate in the initial grayscale image of the photovoltaic panel as the reference of the gate, can more accurately reflect the difference between the monitored cell and gate and the initial cell and gate.

[0020] Optionally, in verifying the validity of the grayscale image, the absolute value of the difference between the grayscale value in the grayscale image and the standard grayscale value of the battery cell is used as the first verification value, and the absolute value of the difference between the grayscale value in the grayscale image and the standard grayscale value of the gate is used as the second verification value. When the minimum value of the first verification value and the minimum value of the second verification value are in the same cluster, it is judged to be invalid; otherwise, it is judged to be valid.

[0021] The beneficial effect is: when the minimum value of the first verification value and the minimum value of the second verification value are in the same cluster, it means that the boundary between the battery cell and the grid cannot be identified, and the position may be a reflective position or the boundary may be covered by dust.

[0022] Optionally, the ratio of the mean value of the texture density in a small neighborhood of the center point to the mean value of the texture density in a large neighborhood is used as the damage assessment, the small neighborhood is eight neighborhoods, and the large neighborhood is sixteen neighborhoods or thirty-two neighborhoods.

[0023] The beneficial effect is that each damage will produce a damage center point, and a number of cracks will be generated around the damage center point. In the eight neighborhoods of the center point, the density of cracks is greater than that in the sixteen neighborhoods of the center point. The ratio of the mean texture density in the small neighborhood of the center point to the mean texture density in the large neighborhood is used as damage evaluation, and the damage range and crack density are quantified through damage evaluation.

[0024] In a second aspect, the present application provides a photovoltaic power generation monitoring and management system, which adopts the following technical solution:

[0025] A photovoltaic power generation monitoring and management system, a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the photovoltaic power generation monitoring and management method mentioned above is implemented.

[0026] The beneficial effect is that the above-mentioned photovoltaic power generation monitoring and management method is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a system based on the memory and the processor for easy use.

[0027] This application has the following technical effects:

[0028] 1. By constructing a grayscale co-occurrence matrix, we can find the density and direction features of the texture. The damage of photovoltaic panels is usually cracks that spread from the center to the surrounding, which is different from the dusty texture or reflective texture. This can help find the damaged area more accurately and improve the accuracy of damaged area identification.

[0029] 2. Each damage will produce a damage center point, and several cracks will be generated around the damage center point. In the eight neighborhoods of the center point, the density of cracks is greater than that in the sixteen neighborhoods of the center point. The ratio of the mean texture density in the small neighborhood of the center point to the mean texture density in the large neighborhood is used as the damage evaluation. The damage range and crack density are quantified through damage evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-restrictive manner, and the same or corresponding numbers represent the same or corresponding parts.

[0031] Figure 1 It is a method flow chart of a photovoltaic power generation monitoring and management method according to an embodiment of the present application.

[0032] Figure 2 This is a structural diagram showing a cell and a grid in a photovoltaic power generation monitoring and management method according to an embodiment of the present application.

[0033] Figure 3 This is a method flow chart of step S2 in a photovoltaic power generation monitoring and management method according to an embodiment of the present application.

[0034] Figure 4 It is a flow chart of a method for damage assessment calculation in a photovoltaic power generation monitoring and management method in an embodiment of the present application.

[0035] Explanation of reference numerals: 1. battery cell; 2. grid. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0037] It should be understood that when the terms "first", "second", etc. are used in the claims, specification and drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0038] The present application embodiment discloses a photovoltaic power generation monitoring and management method, referring to Figure 1 , including steps S1 to S3, which are specifically as follows:

[0039] S1: Mark the cell area and gate area in the grayscale image of the photovoltaic panel to obtain several initial windows.

[0040] Reference Figure 2 The photovoltaic panel includes a cell 1 and a gate 2. The area where the cell 1 is located is the electrode area, and the area where the gate 2 is located is the gate area. The cell 1 is the basic unit of a solar cell and is made of semiconductor material, usually silicon. The cell 1 can be made of monocrystalline silicon, polycrystalline silicon or thin film material. The cell 1 generates current by absorbing sunlight and utilizing the photovoltaic effect. The cells 1 are connected in series and in parallel to form a solar cell module. The gate 2 is a metal wire covering the cell 1, usually made of silver or copper, which is used to collect the current generated by the cell 1 and transmit it to an external circuit. The gate 2 ensures that the current can be effectively extracted from the cell 1.

[0041] When in use, photovoltaic panels are usually composed of multiple photovoltaic panels to form a photovoltaic array. Surface images of the photovoltaic panels in the photovoltaic array are collected and grayscale converted to obtain a grayscale image.

[0042] Since the photovoltaic panel is arranged in a rectangular shape as a whole, the initial window is arranged as a rectangular window, and the rectangular window contains at least one cell area and one grid area. Preferably, the initial window is the area where a photovoltaic panel is located.

[0043] S2: Filter the initial window to obtain the optimal window.

[0044] Due to changes in light intensity and angle, some areas in the surface image will be clearer, while some areas will be blurred due to reflection or severe dust accumulation, so the grayscale image needs to be screened.

[0045] Reference Figure 3 , step S2 includes steps S20 to S22, which are specifically as follows:

[0046] S20: Clustering the grayscale values ​​in the initial window to obtain two clusters.

[0047] The clustering adopts the k-means clustering algorithm, and the number of cluster centers in the cluster is 2. The number of cluster centers refers to the number of categories to be divided into. The k-means clustering algorithm is an existing technology and will not be described in detail here.

[0048] S21: Verify the validity of the grayscale image. If it is valid, calculate the feature resolution index of the initial window, and take the window corresponding to the minimum feature resolution index as the optimal window. Otherwise, delete the grayscale image.

[0049] Set the standard grayscale value of the cell: The standard grayscale value of the cell is: the grayscale value of the cell area in the grayscale image when the photovoltaic panel is installed and used, and the standard grayscale value of the gate is: the grayscale value of the gate area in the grayscale image when the photovoltaic panel is first installed and used. The absolute value of the difference between the grayscale value in the grayscale image and the standard grayscale value of the cell is used as the first verification value, and the absolute value of the difference between the grayscale value in the grayscale image and the standard grayscale value of the gate is used as the second verification value. When the minimum value of the first verification value and the minimum value of the second verification value are in the same cluster, it is judged as invalid, otherwise it is judged as valid.

[0050] When the minimum value of the first verification value and the minimum value of the second verification value are in the same cluster, it means that the boundary between the cell 1 and the grid 2 cannot be identified, and the position may be a reflective position or the boundary may be covered by dust.

[0051] Verifying the validity of the grayscale image is to ensure that the results of the image analysis are of practical significance, that is, they can correctly distinguish between the cell area and the gate area. If not, it may be due to dust coverage or reflection. At this time, a signal can be sent to the operation and maintenance personnel to prompt them to check the collected images. If there is a lot of dust, the operation and maintenance personnel need to clean the photovoltaic panels.

[0052] When the verification is effective, the feature resolution index is calculated. In one embodiment, the feature resolution index is calculated as follows: ,

[0053] ,

[0054] ,

[0055] in, represents the feature resolution index, Indicates the cell resolution index, represents the gate resolution index, represents the maximum value function; Indicates The exponential function with base , Indicates the first cluster The ratio of pixels to all pixels in the initial window, is the gray value of the first pixel in the first cluster, represents the standard grayscale value of the battery cell, min represents the minimum function, The first cluster The gray value of each pixel; Indicates the second cluster The ratio of pixels to all pixels in the initial window, is the gray value of the first pixel in the second cluster, Indicates the standard gray value of the gate, The second cluster The gray value of a pixel.

[0056] Taking an initial window as an example, an initial window corresponds to a feature resolution index. The window corresponding to the minimum feature resolution index within a day is taken as the optimal window for that day, which can reduce the impact of reflection on defect recognition. The smaller the feature resolution index, the more obvious the feature difference between the cell area and the gate area, thereby improving the image resolution ability. Finding the more obvious grayscale images in multiple acquisition drawings within the cycle for analysis can reduce the misjudgment rate.

[0057] S3: In the optimal window, the damage evaluation is calculated, and when the damage evaluation is greater than 0, an alarm signal indicating the presence of damage is generated.

[0058] Reference Figure 4 The calculation method of damage evaluation includes steps S30 and S31, which are as follows:

[0059] S30: In the optimal window, construct entropy values ​​of the gray level co-occurrence matrix in multiple directions, take the maximum entropy value as the texture density value, the direction corresponding to the maximum entropy value is the texture direction of the crack, and each optimal window corresponds to a texture density value.

[0060] Constructing a gray-level co-occurrence matrix and calculating entropy values ​​are both existing technologies and will not be described in detail here.

[0061] S31: construct a texture density matrix according to the texture density value, take any point in the texture density matrix as the center point, and take the ratio of the mean texture density in the small neighborhood of the center point to the mean texture density in the large neighborhood as the damage evaluation.

[0062] In the present application, the small neighborhood is eight neighborhoods, and the large neighborhood is sixteen neighborhoods or thirty-two neighborhoods.

[0063] The eight-neighborhood refers to the neighborhood consisting of eight pixels around a pixel. Specifically, it includes the neighboring pixels above, below, left, right and four diagonal directions of the pixel, excluding the central pixel, for a total of eight pixels.

[0064] The sixteen-neighborhood is based on the eight-neighborhood, and also includes the four pixels above, below, left, and right of each eight-neighborhood pixel, thus forming a The grid has 16 pixels in total, excluding the central pixel.

[0065] The thirty-two neighborhoods include all the pixels in the sixteen neighborhoods, and are further extended to the pixels above, below, left, right and in four diagonal directions of each of the sixteen neighborhood pixels. In addition to the central pixel, there are 32 pixels in total.

[0066] Each damage will produce a damage center point, and several cracks will be generated around the damage center point. The density of cracks in the eight neighborhoods of the center point is greater than that in the sixteen neighborhoods of the center point. The ratio of the mean texture density in the small neighborhood of the center point to the mean texture density in the large neighborhood is used as the damage evaluation, which quantifies the damage range and crack density.

[0067] When the damage evaluation is greater than 0, an alarm signal indicating the presence of damage is generated. The alarm signal may be a text message such as "damaged area identified" sent to the monitoring platform.

[0068] An embodiment of the present application also discloses a photovoltaic power generation monitoring and management system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a photovoltaic power generation monitoring and management method according to the present application is implemented.

[0069] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0070] In the present application, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high bandwidth memory HBM (High Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0071] Although this specification has shown and described a plurality of embodiments of the present application, it is obvious to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, modifications and alternatives without departing from the thought and spirit of the present application. It should be understood that in the process of practicing the present application, various alternatives to the embodiments of the present application described herein may be adopted.

[0072] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A photovoltaic power generation monitoring and management method, characterized in that: Includes steps: Mark the cell area and the gate area in the grayscale image of the photovoltaic panel to obtain a number of initial windows; screen the initial windows to obtain the optimal window, including the steps of: Cluster the grayscale values ​​in the initial window to obtain two clusters; Verify the validity of the grayscale image. If it is valid, calculate the feature resolution index of the initial window, and take the window corresponding to the minimum feature resolution index as the optimal window. Otherwise, delete the grayscale image. In the optimal window, calculate the damage evaluation. When the damage evaluation is greater than 0, generate an alarm signal of damage. Among them, the calculation method of damage evaluation is: construct the entropy value of the grayscale co-occurrence matrix in multiple directions, take the maximum entropy value as the texture density value, the direction corresponding to the maximum entropy value is the texture direction of the crack, and each optimal window corresponds to a texture density value; construct a texture density matrix according to the texture density value, take any point in the texture density matrix as the center point, and take the ratio of the mean texture density in the small neighborhood of the center point to the mean texture density in the large neighborhood as the damage evaluation.

2. The photovoltaic power generation monitoring and management method according to claim 1, characterized in that: The calculation formula of the feature resolution index is: , ， ， in, represents the feature resolution index, Indicates the cell resolution index, represents the gate resolution index, represents the maximum value function; Indicates The exponential function with base , Indicates the first cluster The ratio of pixels to all pixels in the initial window, is the gray value of the first pixel in the first cluster, represents the standard grayscale value of the battery cell, min represents the minimum value function, The first cluster The gray value of each pixel; Indicates the second cluster The ratio of pixels to all pixels in the initial window, is the gray value of the first pixel in the second cluster, Indicates the standard gray value of the gate, The second cluster The gray value of a pixel.

3. The photovoltaic power generation monitoring and management method according to claim 1, characterized in that: The clustering adopts the k-means clustering algorithm, and the number of cluster centers in the cluster is 2.

4. The photovoltaic power generation monitoring and management method according to claim 1, characterized in that: The standard grayscale value of the cell is: the grayscale value of the cell area in the grayscale image when the photovoltaic panel is installed and used, and the standard grayscale value of the gate is: the grayscale value of the gate area in the grayscale image when the photovoltaic panel is installed and used for the first time.

5. The photovoltaic power generation monitoring and management method according to claim 1, characterized in that: In verifying the validity of the grayscale image, the absolute value of the difference between the grayscale value in the grayscale image and the standard grayscale value of the battery cell is used as the first verification value, and the absolute value of the difference between the grayscale value in the grayscale image and the standard grayscale value of the gate is used as the second verification value. When the minimum value of the first verification value and the minimum value of the second verification value are in the same cluster, it is judged to be invalid, otherwise it is judged to be valid.

6. The photovoltaic power generation monitoring and management method according to claim 1, characterized in that: The ratio of the mean texture density in the small neighborhood of the center point to the mean texture density in the large neighborhood is used as the damage evaluation. The small neighborhood is eight neighbors, and the large neighborhood is sixteen neighbors or thirty-two neighbors.

7. A photovoltaic power generation monitoring and management system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the photovoltaic power generation monitoring and management method according to any one of claims 1 to 6 is implemented.

Images