Photovoltaic module fault positioning method and system

Abstract

The invention relates to photovoltaic fault detection, in particular to a photovoltaic module fault positioning method and system, and the method comprises the steps: collecting an infrared image and a panoramic image of a photovoltaic module, carrying out the three-dimensional reconstruction of the infrared image, and determining the three-dimensional point cloud of the photovoltaic module; performing fault detection on the infrared image, and determining a fault string; determining that a target infrared image of the same fault string is shot, fusing three-dimensional point clouds corresponding to the fault string in the target infrared image, and determining a first geographic position coordinate of the fault string according to the fused three-dimensional point clouds; according to the infrared image, determining first position information of the fault group string in the photovoltaic module and fault positioning of the fault group string in the infrared image; determining second position information of the fault group string in the panoramic image according to the panoramic image and the fault location; according to the technical scheme provided by the invention, the defect that the fault string is difficult to accurately position in the prior art can be effectively overcome.

Description

Technical Field

[0001] The present invention relates to photovoltaic fault detection, and in particular to a photovoltaic component fault locating method and system. Background Art

[0002] As the photovoltaic operation and maintenance business continues to expand, the operation and maintenance personnel need to record the faults they find during inspections in photovoltaic power stations. At present, handheld GPS is generally used to record the location of the fault point and take photos of the environment around the fault point for operation and maintenance. However, the fault point location recorded by the handheld GPS is not accurate enough, and it is not easy to distinguish the location in the photos taken.

[0003] The existing technical solution mainly uses drone-based infrared devices for photovoltaic inspections, but when performing photovoltaic operation and maintenance, operation and maintenance personnel are still required to conduct on-site inspections. However, this method cannot accurately locate the faulty strings. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides a photovoltaic module fault locating method and system, which can effectively overcome the defect of the prior art that it is difficult to accurately locate the faulty group string.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] A photovoltaic module fault location method comprises the following steps:

[0009] S1, collecting infrared images and panoramic images of photovoltaic modules, and performing three-dimensional reconstruction on the infrared images to determine the three-dimensional point cloud of the photovoltaic modules;

[0010] S2, perform fault detection on the infrared image and determine the fault string;

[0011] S3, determining a target infrared image that captures the same faulty string, fusing the three-dimensional point cloud corresponding to the faulty string in the target infrared image, and determining the first geographical location coordinates of the faulty string based on the fused three-dimensional point cloud;

[0012] S4, determining first position information of the faulty string in the photovoltaic assembly and fault location of the faulty string in the infrared image according to the infrared image;

[0013] S5. Determine second position information of the faulty string in the panoramic image according to the panoramic image and the fault location;

[0014] S6, determining the logical location of the faulty string in the panoramic image according to the first location information and the second location information, and determining the second geographical location coordinates of the faulty string according to the logical location;

[0015] S7. Determine the precise geographical location coordinates of the faulty string by combining the first geographical location coordinates and the second geographical location coordinates.

[0016] Preferably, determining in S3 that the target infrared image of the same fault string is captured includes:

[0017] Determine the pending infrared images that capture the faulty cluster, and calculate the similarity between the pending infrared images;

[0018] Among the pending infrared images whose similarity is greater than a preset threshold, a target infrared image that captures the same fault string is determined.

[0019] Preferably, the three-dimensional point cloud corresponding to the fault cluster in the target infrared image is fused in S3, including:

[0020] Extract the feature points of each target infrared image, and track the feature points of the same name in all target infrared images through feature point matching;

[0021] The homography matrix between the target infrared images is calculated using the feature points with the same name, and the three-dimensional point cloud corresponding to the fault cluster in one target infrared image is mapped to another target infrared image according to the homography matrix.

[0022] Preferably, tracking the feature points of the same name in all target infrared images by feature point matching includes:

[0023] Add an additional redundant detection frame of one target infrared image to another target infrared image, and determine the fault type of the fault string to determine the target feature points;

[0024] Feature point matching is performed based on the target feature points, and the feature points with the same name in all target infrared images are tracked.

[0025] Preferably, determining the first position information of the faulty string in the photovoltaic assembly according to the infrared image in S4 includes:

[0026] Segment the infrared image to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image;

[0027] Perform fault detection on the infrared image and determine the pixel coordinates of the fault cluster;

[0028] The first position information of the faulty string in the photovoltaic assembly is determined according to the sub-infrared image and the pixel coordinates corresponding to the faulty string.

[0029] Preferably, the segmentation process of the infrared image to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image includes:

[0030] Performing group segmentation on each photovoltaic string in the infrared image to obtain a first group string image corresponding to each photovoltaic string;

[0031] Performing self-correction on the first string images corresponding to each photovoltaic string to obtain a corrected string image corresponding to each photovoltaic string;

[0032] The corrected string images corresponding to each photovoltaic string are segmented and filled to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image.

[0033] Preferably, determining the fault location of the faulty string in the infrared image according to the infrared image in S4 includes:

[0034] The infrared image is reconstructed in three dimensions to determine the three-dimensional point cloud of the photovoltaic module and the fault location of the faulty string in the infrared image.

[0035] Preferably, determining the second position information of the faulty string in the panoramic image according to the panoramic image and the fault location in S5 includes:

[0036] Process the panoramic image to obtain the photovoltaic string segmentation mask;

[0037] The panoramic image is segmented according to the photovoltaic string segmentation mask to obtain a sub-panoramic image corresponding to each photovoltaic string in the panoramic image;

[0038] According to the sub-panoramic image corresponding to the faulty string and the fault location in the infrared image, second position information of the faulty string in the panoramic image is determined.

[0039] Preferably, determining the logical location of the faulty string in the panoramic image according to the first location information and the second location information in S6 includes:

[0040] Determining third position information of the faulty string relative to the photovoltaic assembly according to second position information of the faulty string in the panoramic image;

[0041] Determine whether the first position information is the same as the third position information. If they are the same, use the second position information of the faulty group string in the panoramic image as the logical position of the faulty group string in the panoramic image; otherwise, adjust the third position information according to the first position information, and use the adjusted third position information as the logical position of the faulty group string in the panoramic image.

[0042] A photovoltaic module fault positioning system, comprising an image acquisition module, an infrared image processing module, a fault string detection module, a first geographical location coordinate determination module, a first location information determination module, a fault positioning determination module, a second location information determination module, a second geographical location coordinate determination module and a precise geographical location coordinate determination module;

[0043] An image acquisition module, which acquires infrared images and panoramic images of photovoltaic modules;

[0044] Infrared image processing module, which performs three-dimensional reconstruction of infrared images and determines the three-dimensional point cloud of photovoltaic modules;

[0045] The fault string detection module performs fault detection on the infrared image and determines the fault string;

[0046] A first geographic location coordinate determination module determines a target infrared image of the same faulty string, fuses a three-dimensional point cloud corresponding to the faulty string in the target infrared image, and determines a first geographic location coordinate of the faulty string based on the fused three-dimensional point cloud;

[0047] A first position information determining module, which determines first position information of the faulty string in the photovoltaic assembly according to the infrared image;

[0048] A fault location determination module determines the fault location of the faulty string in the infrared image according to the infrared image;

[0049] A second position information determination module, which determines second position information of the faulty string in the panoramic image according to the panoramic image and the fault location;

[0050] A second geographical location coordinate determination module is configured to determine the logical location of the faulty string in the panoramic image according to the first location information and the second location information, and determine the second geographical location coordinate of the faulty string according to the logical location;

[0051] The precise geographic location coordinate determination module determines the precise geographic location coordinate of the faulty string by combining the first geographic location coordinate and the second geographic location coordinate.

[0052] (III) Beneficial effects

[0053] Compared with the prior art, the photovoltaic module fault location method and system provided by the present invention has the following beneficial effects:

[0054] 1) Collect infrared images and panoramic images of photovoltaic modules, perform three-dimensional reconstruction on the infrared images, determine the three-dimensional point cloud of the photovoltaic modules, perform fault detection on the infrared images, determine the faulty string, determine the target infrared image that captures the same faulty string, fuse the three-dimensional point cloud corresponding to the faulty string in the target infrared image, and effectively determine the first geographical location coordinates of the faulty string based on the fused three-dimensional point cloud;

[0055] 2) Determine the first position information of the faulty string in the PV module and the fault location of the faulty string in the infrared image based on the infrared image, determine the second position information of the faulty string in the panoramic image based on the panoramic image and the fault location, determine the logical location of the faulty string in the panoramic image based on the first position information and the second position information, and determine the second geographical location coordinates of the faulty string based on the logical location; by combining the first geographical location coordinates and the second geographical location coordinates, the precise geographical location coordinates of the faulty string can be effectively determined, thereby enabling accurate location of the faulty string. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0057] Figure 1 It is a schematic diagram of the process of the present invention;

[0058] Figure 2 A schematic diagram of a process for determining the first geographical location coordinates of a faulty string in the present invention;

[0059] Figure 3 It is a schematic diagram of the flow of determining the second geographical location coordinates of the faulty string in the present invention. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] A photovoltaic module fault location method, such as Figure 1 and Figure 2 As shown, ① the infrared image and panoramic image of the photovoltaic module are collected, and the infrared image is reconstructed in three dimensions to determine the three-dimensional point cloud of the photovoltaic module.

[0062] ② Perform fault detection on the infrared image and determine the fault string.

[0063] ③ Determine the target infrared image that captures the same faulty string, fuse the three-dimensional point cloud corresponding to the faulty string in the target infrared image, and determine the first geographic location coordinates of the faulty string based on the fused three-dimensional point cloud.

[0064] 1) Determine the target infrared image that captures the same fault string, including:

[0065] Determine the pending infrared images that capture the faulty cluster, and calculate the similarity between the pending infrared images;

[0066] Among the pending infrared images whose similarity is greater than a preset threshold, a target infrared image that captures the same fault string is determined.

[0067] 2) Fusion of the 3D point cloud corresponding to the fault cluster in the target infrared image, including:

[0068] Extract the feature points of each target infrared image, and track the feature points of the same name in all target infrared images through feature point matching;

[0069] The homography matrix between the target infrared images is calculated using the feature points with the same name, and the three-dimensional point cloud corresponding to the fault cluster in one target infrared image is mapped to another target infrared image according to the homography matrix.

[0070] Specifically, all feature points of the same name in the target infrared image are tracked through feature point matching, including:

[0071] Add an additional redundant detection frame of one target infrared image to another target infrared image, and determine the fault type of the fault string to determine the target feature points;

[0072] Feature point matching is performed based on the target feature points, and the feature points with the same name in all target infrared images are tracked.

[0073] The above technical scheme collects infrared images and panoramic images of photovoltaic modules, and reconstructs the infrared images in three dimensions to determine the three-dimensional point cloud of the photovoltaic modules, performs fault detection on the infrared images, determines the faulty string, determines the target infrared image that captures the same faulty string, fuses the three-dimensional point cloud corresponding to the faulty string in the target infrared image, and can effectively determine the first geographical location coordinates of the faulty string based on the fused three-dimensional point cloud.

[0074] like Figure 1 and Figure 3 As shown, ④ determining the first position information of the faulty string in the photovoltaic module and the fault location of the faulty string in the infrared image according to the infrared image.

[0075] 1) Determining the first position information of the faulty string in the photovoltaic module according to the infrared image, including:

[0076] Segment the infrared image to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image;

[0077] Perform fault detection on the infrared image and determine the pixel coordinates of the fault cluster;

[0078] The first position information of the faulty string in the photovoltaic assembly is determined according to the sub-infrared image and the pixel coordinates corresponding to the faulty string.

[0079] Specifically, the infrared image is segmented to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image, including:

[0080] Performing group segmentation on each photovoltaic string in the infrared image to obtain a first group string image corresponding to each photovoltaic string;

[0081] Performing self-correction on the first string images corresponding to each photovoltaic string to obtain a corrected string image corresponding to each photovoltaic string;

[0082] The corrected string images corresponding to each photovoltaic string are segmented and filled to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image.

[0083] 2) Determine the fault location of the faulty string in the infrared image according to the infrared image, including:

[0084] The infrared image is reconstructed in three dimensions to determine the three-dimensional point cloud of the photovoltaic module and the fault location of the faulty string in the infrared image.

[0085] ⑤ Determine the second position information of the faulty string in the panoramic image according to the panoramic image and the fault location, specifically including:

[0086] Process the panoramic image to obtain the photovoltaic string segmentation mask;

[0087] The panoramic image is segmented according to the photovoltaic string segmentation mask to obtain a sub-panoramic image corresponding to each photovoltaic string in the panoramic image;

[0088] According to the sub-panoramic image corresponding to the faulty string and the fault location in the infrared image, second position information of the faulty string in the panoramic image is determined.

[0089] ⑥ Determine the logical location of the faulty string in the panoramic image according to the first location information and the second location information, and determine the second geographical location coordinates of the faulty string according to the logical location.

[0090] Specifically, determining the logical location of the faulty string in the panoramic image according to the first location information and the second location information includes:

[0091] Determining third position information of the faulty string relative to the photovoltaic assembly according to second position information of the faulty string in the panoramic image;

[0092] Determine whether the first position information is the same as the third position information. If they are the same, use the second position information of the faulty group string in the panoramic image as the logical position of the faulty group string in the panoramic image; otherwise, adjust the third position information according to the first position information, and use the adjusted third position information as the logical position of the faulty group string in the panoramic image.

[0093] ⑦ Determine the precise geographic location coordinates of the faulty string by combining the first geographic location coordinates and the second geographic location coordinates.

[0094] The above technical scheme determines the first position information of the faulty string in the photovoltaic component and the fault location of the faulty string in the infrared image based on the infrared image, determines the second position information of the faulty string in the panoramic image based on the panoramic image and the fault location, determines the logical location of the faulty string in the panoramic image based on the first position information and the second position information, and determines the second geographical location coordinates of the faulty string based on the logical location, and by combining the first geographical location coordinates and the second geographical location coordinates, the precise geographical location coordinates of the faulty string can be effectively determined, thereby achieving accurate location of the faulty string.

[0095] On the basis of the photovoltaic module fault locating method disclosed above, the technical solution of the present application further discloses a photovoltaic module fault locating system, including an image acquisition module, an infrared image processing module, a fault string detection module, a first geographical location coordinate determination module, a first location information determination module, a fault location determination module, a second location information determination module, a second geographical location coordinate determination module and a precise geographical location coordinate determination module;

[0096] An image acquisition module, which acquires infrared images and panoramic images of photovoltaic modules;

[0097] Infrared image processing module, which performs three-dimensional reconstruction of infrared images and determines the three-dimensional point cloud of photovoltaic modules;

[0098] The fault string detection module performs fault detection on the infrared image and determines the fault string;

[0099] A first geographic location coordinate determination module determines a target infrared image of the same faulty string, fuses a three-dimensional point cloud corresponding to the faulty string in the target infrared image, and determines a first geographic location coordinate of the faulty string based on the fused three-dimensional point cloud;

[0100] A first position information determining module, which determines first position information of the faulty string in the photovoltaic assembly according to the infrared image;

[0101] A fault location determination module determines the fault location of the faulty string in the infrared image according to the infrared image;

[0102] A second position information determination module, which determines second position information of the faulty string in the panoramic image according to the panoramic image and the fault location;

[0103] A second geographical location coordinate determination module is configured to determine the logical location of the faulty string in the panoramic image according to the first location information and the second location information, and determine the second geographical location coordinate of the faulty string according to the logical location;

[0104] The precise geographic location coordinate determination module determines the precise geographic location coordinate of the faulty string by combining the first geographic location coordinate and the second geographic location coordinate.

[0105] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic module fault location method, characterized in that: The following steps are involved: S1, collecting infrared images and panoramic images of photovoltaic modules, and performing three-dimensional reconstruction on the infrared images to determine the three-dimensional point cloud of the photovoltaic modules; S2, perform fault detection on the infrared image and determine the fault string; S3, determining a target infrared image that captures the same faulty string, fusing the three-dimensional point cloud corresponding to the faulty string in the target infrared image, and determining the first geographical location coordinates of the faulty string according to the fused three-dimensional point cloud; S4, determining first position information of the faulty string in the photovoltaic assembly and fault location of the faulty string in the infrared image according to the infrared image; S5. Determine second position information of the faulty string in the panoramic image according to the panoramic image and the fault location; S6, determining the logical location of the faulty string in the panoramic image according to the first location information and the second location information, and determining the second geographical location coordinates of the faulty string according to the logical location; S7. Determine the precise geographical location coordinates of the faulty string by combining the first geographical location coordinates and the second geographical location coordinates.

2. The photovoltaic module fault location method according to claim 1, characterized in that: In S3, it is determined that the target infrared image of the same fault string is captured, including: Determine the pending infrared images that capture the faulty cluster, and calculate the similarity between the pending infrared images; Among the pending infrared images whose similarity is greater than a preset threshold, a target infrared image that captures the same fault string is determined.

3. The photovoltaic module fault location method according to claim 2, characterized in that: The 3D point cloud corresponding to the fault cluster in the target infrared image fused in S3 includes: Extract the feature points of each target infrared image, and track the feature points of the same name in all target infrared images through feature point matching; The homography matrix between the target infrared images is calculated using the feature points with the same name, and the three-dimensional point cloud corresponding to the fault cluster in one target infrared image is mapped to another target infrared image according to the homography matrix.

4. The photovoltaic module fault location method according to claim 3, characterized in that: The tracking of all feature points of the same name in the target infrared images by feature point matching includes: Add an additional redundant detection frame of one target infrared image to another target infrared image, and determine the fault type of the fault string to determine the target feature points; Feature point matching is performed based on the target feature points, and the feature points with the same name in all target infrared images are tracked.

5. The photovoltaic module fault location method according to claim 1, characterized in that: In S4, first position information of the faulty string in the photovoltaic assembly is determined according to the infrared image, including: Segment the infrared image to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image; Perform fault detection on the infrared image and determine the pixel coordinates of the fault cluster; The first position information of the faulty string in the photovoltaic assembly is determined according to the sub-infrared image and the pixel coordinates corresponding to the faulty string.

6. The photovoltaic module fault location method according to claim 5, characterized in that: The infrared image is segmented to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image, including: Performing group segmentation on each photovoltaic string in the infrared image to obtain a first group string image corresponding to each photovoltaic string; Performing self-correction on the first string images corresponding to each photovoltaic string to obtain a corrected string image corresponding to each photovoltaic string; The corrected string images corresponding to each photovoltaic string are segmented and filled to obtain a sub-infrared image corresponding to each photovoltaic string in the infrared image.

7. The photovoltaic module fault location method according to claim 5, characterized in that: In S4, the fault location of the faulty string in the infrared image is determined according to the infrared image, including: The infrared image is reconstructed in three dimensions to determine the three-dimensional point cloud of the photovoltaic module and the fault location of the faulty string in the infrared image.

8. The photovoltaic module fault location method according to claim 7, characterized in that: In S5, second position information of the faulty string in the panoramic image is determined according to the panoramic image and the fault location, including: Process the panoramic image to obtain the photovoltaic string segmentation mask; The panoramic image is segmented according to the photovoltaic string segmentation mask to obtain a sub-panoramic image corresponding to each photovoltaic string in the panoramic image; According to the sub-panoramic image corresponding to the faulty string and the fault location in the infrared image, second position information of the faulty string in the panoramic image is determined.

9. The photovoltaic module fault location method according to claim 8, characterized in that: In S6, the logical location of the faulty string in the panoramic image is determined according to the first location information and the second location information, including: Determining third position information of the faulty string relative to the photovoltaic assembly according to second position information of the faulty string in the panoramic image; Determine whether the first position information is the same as the third position information. If they are the same, use the second position information of the faulty group string in the panoramic image as the logical position of the faulty group string in the panoramic image; otherwise, adjust the third position information according to the first position information, and use the adjusted third position information as the logical position of the faulty group string in the panoramic image.

10. A photovoltaic module fault locating system, used to execute the photovoltaic module fault locating method according to claim 1, characterized in that: It includes an image acquisition module, an infrared image processing module, a fault string detection module, a first geographical location coordinate determination module, a first location information determination module, a fault location determination module, a second location information determination module, a second geographical location coordinate determination module and a precise geographical location coordinate determination module; An image acquisition module, which acquires infrared images and panoramic images of photovoltaic modules; Infrared image processing module, which performs three-dimensional reconstruction of infrared images and determines the three-dimensional point cloud of photovoltaic modules; The fault string detection module performs fault detection on the infrared image and determines the fault string; A first geographic location coordinate determination module determines a target infrared image of the same faulty string, fuses a three-dimensional point cloud corresponding to the faulty string in the target infrared image, and determines a first geographic location coordinate of the faulty string based on the fused three-dimensional point cloud; A first position information determining module, which determines first position information of the faulty string in the photovoltaic assembly according to the infrared image; A fault location determination module determines the fault location of the faulty string in the infrared image according to the infrared image; A second position information determination module, which determines second position information of the faulty string in the panoramic image according to the panoramic image and the fault location; A second geographical location coordinate determination module is configured to determine the logical location of the faulty string in the panoramic image according to the first location information and the second location information, and determine the second geographical location coordinate of the faulty string according to the logical location; The precise geographic location coordinate determination module determines the precise geographic location coordinate of the faulty string by combining the first geographic location coordinate and the second geographic location coordinate.

Images