Photovoltaic module fault diagnosis method based on multi-source information

Abstract

The invention relates to the technical field of photovoltaic module fault diagnosis, in particular to a photovoltaic module fault diagnosis method based on multi-source information, and the method comprises the following steps: S1, collecting a group of images of a photovoltaic module through an infrared camera, and carrying out the preprocessing of the images, so as to obtain an infrared gray image; s2, carrying out feature region segmentation according to the temperature information of the infrared grayscale image, extracting first-level features of each region, and calculating a temperature diagnosis coefficient; s3, establishing sliding windows according to gray level information in the infrared gray level image, calculating a gray level average value of each sliding window, performing gray level consistency judgment according to the gray level average value of each sliding window, and calculating a gray level diagnosis coefficient; s4, calculating a comprehensive diagnosis coefficient according to the temperature diagnosis coefficient and the gray diagnosis coefficient, and judging whether the photovoltaic module breaks down or not; s5, issuing a fault diagnosis result of the photovoltaic module; according to the invention, the accuracy of fault diagnosis is improved through combined diagnosis of the gray value and the temperature.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic module fault diagnosis, and in particular to a photovoltaic module fault diagnosis method based on multi-source information. Background Art

[0002] The current maintenance model, which relies on manual inspections, is costly and labor-intensive. Traditional fault detection methods, however, suffer from low efficiency and inaccuracy due to their inability to fully integrate multi-dimensional information. In this context, overcoming existing technical bottlenecks and developing an efficient, accurate, and multi-source information-based photovoltaic module fault diagnosis method has become a critical technical challenge to improve the reliability and economic efficiency of photovoltaic systems.

[0003] Currently, conventional methods for collecting photovoltaic module images contain a large amount of noise, resulting in large errors in fault diagnosis and high computational complexity. Existing technologies often use a single information source for photovoltaic module fault diagnosis, resulting in insufficient information richness and low diagnostic accuracy. Therefore, in response to the above situation, there is an urgent need to develop a photovoltaic module fault diagnosis method based on multi-source information to overcome the shortcomings of current practical applications. Summary of the Invention

[0004] The object of the present invention is to provide a photovoltaic module fault diagnosis method based on multi-source information to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A photovoltaic module fault diagnosis method based on multi-source information includes the following steps:

[0007] S1, an infrared camera collects a set of images of photovoltaic modules at the photovoltaic site and preprocesses the images to obtain infrared grayscale images;

[0008] S2. Segment the infrared grayscale image into feature regions based on the temperature information in the infrared grayscale image, extract primary features from each feature region, perform temperature consistency judgment, and calculate the temperature diagnosis coefficient;

[0009] S3. Establishing sliding windows based on the grayscale information in the infrared grayscale image, calculating the grayscale average value of each sliding window, and performing grayscale consistency judgment based on the grayscale average value between each sliding window to calculate the grayscale diagnostic coefficient;

[0010] S4. Calculate a comprehensive diagnostic coefficient based on the temperature diagnostic coefficient and the grayscale diagnostic coefficient, and determine whether the photovoltaic module has a fault based on the comprehensive diagnostic coefficient;

[0011] S5. Publish the PV module fault diagnosis results.

[0012] As a further solution of the present invention: in step S1, each pixel in the image contains a temperature value and a color value, and the color value is converted into a grayscale value by graying the color of the image to obtain an infrared grayscale image.

[0013] As a further solution of the present invention: the pretreatment comprises the following steps:

[0014] The color value of each pixel in the infrared image is converted into a grayscale value Q, where the color value includes the red value R, the green value G, and the blue value B. After preprocessing the infrared image, the infrared depth image is obtained. The specific conversion formula is as follows:

[0015] Q=0.236×R+0.3986×G+0.613×B.

[0016] As a further solution of the present invention: in step S2, the temperature diagnostic coefficient c is calculated as follows:

[0017]

[0018] in, Represents the temperature difference between two pixels, t m Represents the temperature value of the mth pixel, t l represents the temperature value of the lth laser point, S refers to the number of pixels in the feature area when calculating the temperature diagnostic coefficient, and T represents the transpose of the vector.

[0019] As a further solution of the present invention: in step S3, the grayscale average value of the sliding window is The calculation formula is as follows:

[0020]

[0021] in, represents the grayscale average of the sliding window, r represents the serial number of the sliding window, and H represents the total number of pixel values ​​in the sliding window.

[0022] As a further solution of the present invention: in step S3, the grayscale diagnostic coefficient is calculated as follows:

[0023] Calculate the grayscale residual:

[0024]

[0025] Among them, d and i are not equal;

[0026] Calculate the grayscale diagnostic coefficient:

[0027] As a further solution of the present invention: Step S4 specifically includes:

[0028] (1) Calculate the comprehensive diagnostic coefficient from the temperature diagnostic coefficient and the grayscale diagnostic coefficient;

[0029] (2) Initializing the comprehensive diagnostic threshold;

[0030] (3) If the comprehensive diagnostic coefficient is greater than or equal to the comprehensive diagnostic threshold, it indicates that the PV module has failed and the fault point is located at the position where the grayscale diagnostic coefficient is the largest;

[0031] If the comprehensive diagnosis coefficient is less than the comprehensive diagnosis threshold, it indicates that the PV module has not failed.

[0032] As a further solution of the present invention: the feature region segmentation in step S2 is to divide the image into M n×n equally divided regions, each region containing the same number of pixels.

[0033] As a further solution of the present invention: the sliding window size in step S3 is The sliding windows move along the track direction, and each sliding window does not overlap.

[0034] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. In the present invention, an infrared camera is used to collect a set of images of photovoltaic modules at a photovoltaic site. Through grayscale processing, the computational efficiency of the scheme is improved. Through the joint diagnosis of grayscale value and temperature, the accuracy of fault diagnosis is improved. The infrared grayscale image is segmented into feature areas based on the temperature information in the infrared grayscale image, and the temperature diagnosis coefficient is calculated.

[0037] 2. Establish a sliding window based on the grayscale information in the infrared grayscale image and calculate the grayscale diagnostic coefficient. Calculate the comprehensive diagnostic coefficient based on the temperature diagnostic coefficient and the grayscale diagnostic coefficient, and determine whether the photovoltaic module is faulty based on the comprehensive diagnostic coefficient.

[0038] 3. This solution solves the error problem of a single information source in fault diagnosis by using multi-source information. When calculating the grayscale diagnostic coefficient, a sliding window is used to scan the image in blocks to achieve precise positioning of the fault point. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Flowchart of a photovoltaic module fault diagnosis method based on multi-source information in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will provide a clear and complete description of 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 them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0041] The specific implementation of the present invention is described in detail below with reference to specific embodiments.

[0042] See also Figure 1 , an embodiment of the present invention provides a photovoltaic module fault diagnosis method based on multi-source information, comprising the following steps:

[0043] Step S1: an infrared camera collects a set of images of photovoltaic modules at a photovoltaic site, and pre-processes the images to obtain infrared grayscale images;

[0044] Step S2: Segment the infrared grayscale image into feature regions based on the temperature information in the infrared grayscale image, extract primary features from each feature region, perform temperature consistency judgment, and calculate the temperature diagnosis coefficient;

[0045] Step S3: establishing sliding windows based on the grayscale information in the infrared grayscale image, calculating the grayscale average of each sliding window, and performing grayscale consistency judgment based on the grayscale average between each sliding window to calculate the grayscale diagnostic coefficient;

[0046] Step S4: Calculate a comprehensive diagnostic coefficient based on the temperature diagnostic coefficient and the grayscale diagnostic coefficient, and determine whether the photovoltaic module has a fault based on the comprehensive diagnostic coefficient;

[0047] Step S5: Publish the photovoltaic module fault diagnosis result.

[0048] Among them, step S1: the infrared camera collects a group of images of photovoltaic modules at the photovoltaic site, and pre-processes the images to obtain infrared grayscale images;

[0049] Use an infrared camera to collect a set of images of photovoltaic panels at the photovoltaic site. Each pixel in the image contains a temperature value and a color value.

[0050] In order to improve the efficiency of color processing and preprocess the image, the preprocessing is mainly the process of color grayscale, which converts the color value into grayscale value to obtain the infrared grayscale image;

[0051] The preprocessing process is mainly as follows:

[0052] The infrared camera acquires a set of infrared images. Each pixel in the infrared image contains a color value: red value R, green value G, and blue value B;

[0053] Grayscale the infrared image and convert the color value of each pixel in the infrared image into a grayscale value. The specific conversion formula is as follows:

[0054] Q=0.236xR+0.3986xG+0.613xB;

[0055] Where Q represents the grayscale value of the pixel in the infrared image;

[0056] After preprocessing, the infrared image is converted into an infrared depth image; the infrared camera is used to monitor the photovoltaic modules in real time to achieve accurate monitoring of all photovoltaic modules.

[0057] Step S2: Segment the infrared grayscale image into feature regions based on the temperature information in the infrared grayscale image, extract primary features from each feature region, perform temperature consistency judgment, and calculate the temperature diagnosis coefficient;

[0058] Based on the temperature information in the infrared grayscale image, the infrared grayscale image is segmented into M n×n feature regions. Each feature region contains the same number of pixels. To improve the detection efficiency of photovoltaic modules, the first-level features are extracted from each feature region. The specific location of the fault point is found by comparing the values ​​of the first-level features.

[0059] Extract the first-level features from each feature area, perform temperature consistency judgment, and calculate the temperature diagnostic coefficient. The specific formula is as follows:

[0060]

[0061] in, Represents the temperature difference between two pixels, t m Represents the temperature value of the mth pixel, t l represents the temperature value of the lth laser point, S refers to the number of pixels in the feature area when calculating the temperature diagnostic coefficient, and T represents the transpose of the vector.

[0062] Step S3: establishing sliding windows based on the grayscale information in the infrared grayscale image, calculating the grayscale average value of each sliding window, and performing grayscale consistency judgment based on the grayscale average value between each sliding window to calculate the grayscale diagnostic coefficient;

[0063] Establish a sliding window based on the grayscale information in the infrared grayscale image, and set the sliding window size to The sliding windows move along the trajectory direction, and each sliding window does not overlap.

[0064] The grayscale average of the sliding window is calculated as follows:

[0065]

[0066] in, represents the grayscale average of the sliding window, r represents the serial number of the sliding window, H represents the total number of pixel values ​​in the sliding window, Q i Represents the grayscale value of the i-th sliding window.

[0067] Grayscale consistency judgment, calculation of grayscale diagnostic coefficient, the specific formula is as follows:

[0068] Calculate the grayscale residual:

[0069]

[0070] Where d and i are not equal, represents the grayscale average of the d-th sliding window, Represents the grayscale average value of the i-th sliding window;

[0071] Calculate the grayscale diagnostic coefficient:

[0072]

[0073] Among them, E i Represents the grayscale residual of the i-th sliding window.

[0074] Step S4: Calculate a comprehensive diagnostic coefficient based on the temperature diagnostic coefficient and the grayscale diagnostic coefficient, and determine whether the photovoltaic module has a fault based on the comprehensive diagnostic coefficient;

[0075] Calculate the comprehensive diagnostic coefficient from the temperature diagnostic coefficient and the grayscale diagnostic coefficient;

[0076] Initialize comprehensive diagnostic threshold;

[0077] If the comprehensive diagnostic coefficient is greater than or equal to the comprehensive diagnostic threshold, it indicates that the photovoltaic module has failed and the fault point is located at the position where the grayscale diagnostic coefficient is the largest. If the comprehensive diagnostic coefficient is less than the comprehensive diagnostic threshold, it indicates that the photovoltaic module has not failed.

[0078] Step S5: Publish the photovoltaic module fault diagnosis result.

[0079] Therefore, in the present invention, an infrared camera is used to collect a group of images of photovoltaic modules at a photovoltaic site, and grayscale processing is used to improve the computational efficiency of the scheme. The accuracy of fault diagnosis is improved through joint diagnosis of grayscale value and temperature. The infrared grayscale image is segmented into feature areas according to the temperature information in the infrared grayscale image, and the temperature diagnosis coefficient is calculated. A sliding window is established according to the grayscale information in the infrared grayscale image, and the grayscale diagnosis coefficient is calculated. A comprehensive diagnostic coefficient is calculated from the temperature diagnosis coefficient and the grayscale diagnosis coefficient, and whether the photovoltaic module has a fault is determined based on the comprehensive diagnostic coefficient. This scheme solves the error problem of a single information source in fault diagnosis by using multi-source information. When calculating the grayscale diagnosis coefficient, a sliding window is used to scan the image in blocks, thereby achieving fine positioning of the fault point.

[0080] Example: First, an infrared camera is used to capture a set of images of photovoltaic modules at a photovoltaic site, and the images are preprocessed to obtain infrared grayscale images. It should be noted that the photovoltaic site range scanned by the infrared camera has no additional obstructions to avoid interference with fault diagnosis caused by obstructions;

[0081] The infrared camera collects images of photovoltaic panels at the photovoltaic site. Each pixel in the image contains temperature and color values.

[0082] In order to improve the efficiency of color processing and preprocess the image, the preprocessing is mainly the process of color grayscale, which converts the color value into grayscale value to obtain the infrared grayscale image;

[0083] The preprocessing process is mainly as follows:

[0084] The infrared camera acquires a set of infrared images. Each pixel in the infrared image contains a color value: red value R, green value G, and blue value B;

[0085] Grayscale the infrared image and convert the color value of each pixel in the infrared image into a grayscale value. The specific conversion formula is as follows:

[0086] Q=0.236xR+0.3986xG+0.613xB;

[0087] Where Q represents the grayscale value of the pixel in the infrared image;

[0088] After pre-processing, the infrared image is converted into an infrared depth image. The infrared camera is used to monitor the photovoltaic modules in real time, thus achieving accurate monitoring of all photovoltaic modules.

[0089] After preprocessing, the infrared depth image has a smaller image size, so it requires less computing power for the storage medium and has certain economic value.

[0090] After obtaining the infrared depth image, it is necessary to process the infrared depth image and perform feature region segmentation on the infrared grayscale image according to the temperature information in the infrared grayscale image, and divide the infrared grayscale image into M n×n feature regions. For example, in this scheme, the size of the infrared depth image is 1280×1280, so the infrared grayscale image can be divided into 100 128×128 feature regions.

[0091] Each feature region contains the same number of pixels. To improve the detection efficiency of photovoltaic modules, a primary feature is extracted from each feature region. The specific location of the fault point is found by comparing the values ​​of the primary features.

[0092] Extract the first-level features from each feature area, perform temperature consistency judgment, and calculate the temperature diagnostic coefficient. The specific formula is as follows:

[0093]

[0094] in, Represents the temperature difference between two pixels, t m Represents the temperature value of the mth pixel, t l represents the temperature value of the lth laser point, S refers to the number of pixels in the feature area when calculating the temperature diagnostic coefficient, and T represents the transpose of the vector.

[0095] A sliding window is established based on the grayscale information in the infrared grayscale image, and the grayscale average value of each sliding window is calculated. The grayscale consistency is judged by the grayscale average value between each sliding window, and the grayscale diagnostic coefficient is calculated.

[0096] A sliding window is established according to the grayscale information in the infrared grayscale image. The size of the sliding window is set to m×m. The sliding window moves along the trajectory direction, and each sliding window does not overlap.

[0097] In this embodiment, the sliding window size is 64×64;

[0098] Calculate the grayscale average of the sliding window:

[0099]

[0100] in, represents the grayscale average value of the sliding window, r represents the sequence number of the sliding window, and H represents the total number of pixel values ​​in the sliding window. In this embodiment, H=4096;

[0101] Grayscale consistency judgment, calculation of grayscale diagnostic coefficient, the specific formula is as follows:

[0102] Calculate the grayscale residual:

[0103]

[0104] Among them, d and i are not equal;

[0105] Calculate the grayscale diagnostic coefficient:

[0106] Then, the comprehensive diagnostic coefficient is calculated by the temperature diagnostic coefficient and the grayscale diagnostic coefficient, and whether the photovoltaic module has a fault is determined according to the comprehensive diagnostic coefficient;

[0107] The comprehensive diagnostic coefficient is calculated from the temperature diagnostic coefficient and the grayscale diagnostic coefficient. It should be noted that the comprehensive diagnostic coefficient is the sum of the temperature diagnostic coefficient and the grayscale diagnostic coefficient. By calculating the temperature diagnostic coefficient and the grayscale diagnostic coefficient, the combined diagnosis of multi-source information is achieved, which improves the information richness and the accuracy of the fault.

[0108] Initialize comprehensive diagnostic threshold;

[0109] If the comprehensive diagnostic coefficient is greater than or equal to the comprehensive diagnostic threshold, it indicates that the PV module has failed and the fault point is located at the position with the largest grayscale diagnostic coefficient. If the comprehensive diagnostic coefficient is less than the comprehensive diagnostic threshold, it indicates that the PV module has not failed.

[0110] It should be noted that when a photovoltaic module fails, the grayscale or temperature value of its infrared depth image will change significantly compared to the surrounding pixels. However, a single information value is difficult to achieve accurate judgment. By jointly monitoring temperature and grayscale, single information errors can be avoided, thereby improving the accuracy of photovoltaic module fault diagnosis. Compared with traditional solutions, the present invention requires less computing power and has higher calculation accuracy.

[0111] After PV module fault diagnosis based on multiple information sources, the PV module fault diagnosis results are released and reported to the PV module duty or inspection personnel in a timely manner to eliminate the impact of the faulty PV module on the entire PV module site.

[0112] It should be noted that, in the present invention, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A photovoltaic module fault diagnosis method based on multi-source information, characterized in that: The following steps are involved: S1, an infrared camera collects a set of images of photovoltaic modules at the photovoltaic site and preprocesses the images to obtain infrared grayscale images; S2. Segment the infrared grayscale image into feature regions based on the temperature information in the infrared grayscale image, extract primary features from each feature region, perform temperature consistency judgment, and calculate the temperature diagnosis coefficient; S3. Establishing sliding windows based on the grayscale information in the infrared grayscale image, calculating the grayscale average value of each sliding window, and performing grayscale consistency judgment based on the grayscale average value between each sliding window to calculate the grayscale diagnostic coefficient; S4. Calculate a comprehensive diagnostic coefficient based on the temperature diagnostic coefficient and the grayscale diagnostic coefficient, and determine whether the photovoltaic module has a fault based on the comprehensive diagnostic coefficient; S5. Publish the PV module fault diagnosis results.

2. The photovoltaic module fault diagnosis method based on multi-source information according to claim 1, characterized in that: In step S1, each pixel in the image contains a temperature value and a color value. By graying the color of the image, the color value is converted into a gray value to obtain an infrared gray image.

3. The photovoltaic module fault diagnosis method based on multi-source information according to claim 2, characterized in that: The pretreatment comprises the following steps: The color value of each pixel in the infrared image is converted into a grayscale value Q, where the color value includes the red value R, the green value G, and the blue value B. After preprocessing the infrared image, the infrared depth image is obtained. The specific conversion formula is as follows: Q=0.236×R+0.3986×G+0.613×B.

4. The photovoltaic module fault diagnosis method based on multi-source information according to claim 1, characterized in that: In step S2, the temperature diagnostic coefficient c is calculated as follows: in, Represents the temperature difference between two pixels, t m Represents the temperature value of the mth pixel, t l represents the temperature value of the lth laser point, S refers to the number of pixels in the feature area when calculating the temperature diagnostic coefficient, and T represents the transpose of the vector.

5. The photovoltaic module fault diagnosis method based on multi-source information according to claim 1, characterized in that: In step S3, the grayscale average value of the sliding window The calculation formula is as follows: in, represents the grayscale average of the sliding window, r represents the serial number of the sliding window, and H represents the total number of pixel values ​​in the sliding window.

6. The photovoltaic module fault diagnosis method based on multi-source information according to claim 5, characterized in that: In step S3, the grayscale diagnostic coefficient is calculated as follows: Calculate the grayscale residual: Among them, d and i are not equal; Calculate the grayscale diagnostic coefficient:

7. The photovoltaic module fault diagnosis method based on multi-source information according to claim 1, characterized in that: The step S4 specifically includes: (1) Calculate the comprehensive diagnostic coefficient from the temperature diagnostic coefficient and the grayscale diagnostic coefficient; (2) Initializing the comprehensive diagnostic threshold; (3) If the comprehensive diagnostic coefficient is greater than or equal to the comprehensive diagnostic threshold, it indicates that the PV module has failed and the fault point is located at the position where the grayscale diagnostic coefficient is the largest; If the comprehensive diagnosis coefficient is less than the comprehensive diagnosis threshold, it indicates that the PV module has not failed.

8. The photovoltaic module fault diagnosis method based on multi-source information according to claim 1, characterized in that: The feature region segmentation in step S2 is to divide the image into M n×n equally divided regions, each region containing the same number of pixels.

9. The photovoltaic module fault diagnosis method based on multi-source information according to claim 1, characterized in that: The sliding window size in step S3 is m×m, the sliding window moves along the track direction, and each sliding window does not overlap.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images