Photovoltaic panel fault diagnosis method and system based on image processing

Abstract

The invention relates to the field of image processing, in particular to a photovoltaic panel fault diagnosis method and system based on image processing, and the method comprises the steps: obtaining a thermodynamic image of a photovoltaic panel in real time, and carrying out the dichotomy of the thermodynamic image, and obtaining an illumination region and a shadow region; for the illumination area and the shadow area, based on a preset trigger monitoring condition, obtaining a monitoring starting moment of temperature rise of the illumination area and a monitoring starting moment of temperature rise of the shadow area after shielding is removed; starting from a monitoring starting moment, calculating a temperature rise anomaly index of each position point, and traversing each position point in the thermodynamic diagram to obtain a temperature rise anomaly matrix; and finding out non-zero connected domains in the temperature rise abnormity matrix, calculating the abnormity degree of each non-zero connected domain, and when the abnormity degree is smaller than a preset threshold value, determining that the corresponding non-zero connected domain is an abnormal temperature rise area so as to complete fault diagnosis of the photovoltaic panel. According to the invention, the accuracy of photovoltaic panel fault detection is effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing. More specifically, the present invention relates to a method and system for photovoltaic panel fault diagnosis based on image processing. Background Art

[0002] A photovoltaic panel, also known as a solar photovoltaic panel, is a semiconductor device that directly converts solar light energy into electrical energy. Its working principle is based on the photovoltaic effect. When sunlight shines on the surface of the photovoltaic panel, the photon energy is absorbed by the semiconductor material, and finally a potential difference is formed at both ends of the battery. When the external circuit is connected, an electric current is generated, thus realizing the conversion of light energy into electrical energy.

[0003] When using a photovoltaic panel in a high-altitude area, due to the thin air, strong solar radiation during the day, and fast heat dissipation at night, the temperature difference between day and night is large. In addition, in high-altitude areas, the solar radiation intensity is high and the sunshine duration is long, so the solar energy received by the photovoltaic panel is large. Due to the large temperature difference between day and night and strong light, the temperature difference between the shaded area and the unshaded area on the photovoltaic panel is significant. For example, the shaded area may have a lower temperature because there is no direct sunlight, while the illuminated area has a higher temperature.

[0004] However, there are some problems with existing photovoltaic panel fault diagnosis methods. On the one hand, traditional fault diagnosis methods mainly rely on monitoring electrical parameters such as the output current and voltage of the photovoltaic panel. These methods cannot directly reflect the temperature distribution of the photovoltaic panel, and temperature anomalies are often one of the early signs of photovoltaic panel faults. On the other hand, for the situation where the temperature difference between the shaded area and the illuminated area of the photovoltaic panel is significant due to the large temperature difference between day and night and strong light in high-altitude areas, existing methods are difficult to accurately distinguish between normal temperature differences and temperature anomalies caused by faults, which easily leads to misjudgment or missed judgment, thus affecting the accuracy of photovoltaic panel fault diagnosis. Summary of the Invention

[0005] To solve the above technical problem of how to accurately diagnose the faults of photovoltaic panels, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a method for photovoltaic panel fault diagnosis based on image processing includes: Obtain the thermal image of the photovoltaic panel in real time, and perform binary classification on the thermal image to obtain the illuminated area and the shaded area; For the illuminated area and the shaded area, obtain the monitoring start time of the temperature rise of the illuminated area after the occlusion is removed and the monitoring start time of the temperature rise of the shaded area based on a preset trigger monitoring condition; Starting from the monitoring start time, calculate the temperature rise anomaly index for each position point, traverse each position point in the thermal image, and obtain the temperature rise anomaly matrix; Find the non-zero connected regions in the temperature rise anomaly matrix, calculate the anomaly degree of each non-zero connected region, and when the anomaly degree is less than the preset threshold, determine the corresponding non-zero connected region as the abnormal temperature rise area to complete the fault diagnosis of the photovoltaic panel.

[0007] The present invention first differentiates the illumination / shadow regions and sets the monitoring start time, which can exclude the instantaneous interference caused by obstacles (such as fallen leaves and dust accumulation), and only starts the temperature rise monitoring after the occlusion is removed, ensuring that the subsequent analysis focuses on the performance problems of the photovoltaic panel itself; then calculates the temperature rise anomaly index of each pixel point and generates a matrix, realizing the quantitative conversion from the global thermal map to the local abnormal points; then aggregates the discrete abnormal points into region-level features through connected component analysis, and combines the anomaly degree and threshold comparison to focus on the real fault region.

[0008] Preferably, the dividing the thermal image into the illumination region and the shadow region includes: Extract the pixel values of all pixel points in the thermal image, perform binary classification on these pixel values, and further divide the thermal image into the illumination region and the shadow region.

[0009] Preferably, the obtaining the monitoring start time of the temperature rise in the illumination region and the monitoring start time of the temperature rise in the shadow region after the occlusion is removed based on the preset trigger monitoring conditions includes: For the position points in the shadow region collected in real time, calculate the ratio of the gray value of this position point to the maximum gray value of the shadow region as the first parameter, and then calculate the ratio of the average gray value of the shadow region to the maximum gray value of the shadow region as the second parameter. When the difference between the first parameter and the second parameter is greater than the preset threshold, the corresponding sampling time is the monitoring start time; For the position points in the illumination region collected in real time, perform the operation on the shadow region, and then obtain the corresponding monitoring start time.

[0010] By calculating the gray value ratio (the first parameter and the second parameter) and judging whether its difference is greater than the threshold, the start time of the temperature rise in the illumination region and the shadow region after the occlusion is removed can be accurately determined, avoiding the temperature data deviation caused by inaccurate monitoring start time.

[0011] Preferably, the calculation process of the temperature rise anomaly index includes: Calculate the time difference from the monitoring start time to the current time for any position point, and calculate the ratio of the gray value of the position point at the current time to the product of the gray value at the monitoring start time and the time difference as the temperature rise anomaly index of the position point at the current time; traverse each position point in the thermal map to obtain the temperature rise anomaly index corresponding to each position point.

[0012] In a heat map, the temperature change conditions at different position points may vary. By calculating the temperature rise anomaly index for each position point, areas with abnormal temperature rise can be identified. The time difference from the start of monitoring to the current moment is considered during the calculation process, which enables the temperature rise anomaly index to reflect the change of temperature over time. As time goes by, the temperature rise anomaly index will be dynamically updated according to the temperature change conditions at the current moment. This dynamic monitoring method can track the temperature change in real time and detect early signs of temperature anomalies in a timely manner. For example, if the temperature rise anomaly index at a certain position point increases rapidly within a short period of time, it may indicate that a sudden temperature anomaly has occurred at that position point, which requires immediate attention and handling.

[0013] Preferably, the process of obtaining non-zero connected regions in the temperature rise anomaly matrix includes: Combining adjacent non-zero temperature rise anomaly indices in the temperature rise anomaly matrix to form a non-zero connected region.

[0014] In the temperature rise anomaly matrix, a non-zero temperature rise anomaly index indicates that the temperature change at that position point exceeds the normal range, and there may be a temperature anomaly. Combining adjacent non-zero temperature rise anomaly indices into a non-zero connected region means grouping position points that are adjacent in space and all have temperature anomalies into the same region. This can identify the range of the temperature anomaly region, rather than just a single position point.

[0015] Preferably, the calculation of the anomaly degree of each non-zero connected region includes: Taking the maximum value point of the non-zero connected region as the center point, in each set direction, starting from the center point, traversing all adjacent points along that direction, and calculating the decrease amplitude of the temperature rise anomaly index in each direction; Taking the minimum value of the decrease amplitudes in all directions as the anomaly degree of this non-zero connected region.

[0016] Selecting the minimum value of the decrease amplitudes in all directions as the anomaly degree of this non-zero connected region. This minimum value reflects the situation where the temperature rise anomaly index decreases most slowly in all directions starting from the center point, and usually represents the direction where the temperature anomaly is the most stubborn or the most difficult to dissipate.

[0017] Preferably, the decrease amplitude satisfies the relational expression: ; is the decrease amplitude of the temperature rise anomaly index at the center point in the 0-degree direction, is the current point of the temperature rise anomaly index, is the current point the temperature rise anomaly index of the next position point, is the number of steps extended outward from the center point.

[0018] Preferably, the decrease amplitude satisfies the relational expression: ; where is the decrease amplitude of the abnormal temperature rise index in the 0-degree direction of the center point, is the current point of the abnormal temperature rise index, is the current point of the abnormal temperature rise index of the next position point, is the number of steps extending outward from the center point.

[0019] By adopting the above technical solution, Preferably, the calculation process of the abnormal temperature rise index includes: Calculating the total time of any position point from the monitoring start time to the current time, and calculating the difference between the gray value of the position point at the current time and the gray value at the monitoring start time; Taking the ratio of the calculated difference in gray value and the total time as the temperature rise trend of the position point at the current time; taking the ratio of the temperature rise trend to the average value of the normal temperature rise trend as the abnormal temperature rise index; Traversing each position point in the thermal image, so as to obtain the abnormal temperature rise index corresponding to each position point.

[0020] In a second aspect, a photovoltaic panel fault diagnosis system based on image processing includes: a processor and a memory, and the memory stores computer program instructions, which implement any one of the photovoltaic panel fault diagnosis methods based on image processing when the computer program instructions are executed by the processor.

[0021] The beneficial effects of the present invention are: By dividing the thermal image into a light area and a shadow area, and respectively calculating the abnormal temperature rise index of each area, it is possible to accurately locate the specific area where a fault occurs on the photovoltaic panel, avoid the error caused by large-area detection of the entire photovoltaic panel, and improve the accuracy of fault detection; Regardless of whether the photovoltaic panel is in the light area or the shadow area, this method can obtain the corresponding monitoring start time based on the preset trigger monitoring conditions and calculate the abnormal temperature rise index, so as to adapt to the fault diagnosis requirements under different lighting conditions, and improve the versatility and adaptability of the method; by calculating the abnormal temperature rise index and combining the abnormal degree of the non-zero connected domain to determine the fault area, when the abnormal degree is greater than or equal to the preset threshold, it is determined that the corresponding non-zero connected domain is an abnormal temperature rise area, thus effectively reducing the false alarm rate and avoiding misjudgment caused by normal temperature rise changes. Description of the Drawings

[0022] Figure 1It is the flowchart of the method from step S1 to step S4 in a photovoltaic panel fault diagnosis method based on image processing according to an embodiment of the present invention. Detailed implementation manners

[0023] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments.

[0024] Refer to Figure 1 , a photovoltaic panel fault diagnosis method based on image processing includes steps S1 - S4, specifically as follows: S1: Obtain the thermal image of the photovoltaic panel in real time, and divide the thermal image into a light region and a shadow region.

[0025] In one embodiment, an infrared thermal imager or other device capable of capturing the temperature distribution is used to collect the thermal image of the photovoltaic panel in real time. These images reflect the temperature distribution of different regions on the surface of the photovoltaic panel.

[0026] The thermal map used in the photovoltaic panel fault diagnosis is usually a grayscale image, where the grayscale value of each pixel represents the temperature at that position.

[0027] Further extract the pixel values of all pixel points in the thermal image, and perform binary classification processing on these pixel values. The commonly used binary classification method is the Otsu threshold method. By automatically determining a threshold, the pixel values are divided into two categories: one category represents a higher temperature (i.e., the light region), and the other category represents a lower temperature (i.e., the shadow region).

[0028] Through the above binary classification, the thermal image is divided into a light region and a shadow region.

[0029] S2: For the light region and the shadow region, obtain the monitoring start time of the temperature rise in the light region after the occlusion is removed and the monitoring start time of the temperature rise in the shadow region based on the preset trigger monitoring conditions.

[0030] When the photovoltaic panel is occluded, the temperature distribution in the occluded region and the light region will change. After the occlusion is removed, the light region will start to have a temperature rise, and the shadow region may also have a temperature rise due to the change of the ambient temperature. However, only the data obtained after the start time of the temperature rise can truly reflect the temperature rise situation of the photovoltaic panel after the occlusion is removed. If the monitoring start time is set too early or too late, it may lead to inaccurate temperature rise data, thus affecting the result of the fault diagnosis.

[0031] Therefore, by setting the preset trigger monitoring conditions, the start time of the temperature rise in the light region and the shadow region after the occlusion is removed, that is, the monitoring start time, can be accurately captured.

[0032] In one embodiment, for the position points in the shadow area collected in real time, calculate the ratio of the gray value of the position point to the maximum gray value of the shadow area as the first parameter, and then calculate the ratio of the average gray value of the shadow area to the maximum gray value of the shadow area as the second parameter. When the difference between the first parameter and the second parameter is greater than a preset threshold (initially set to 0.1 in the embodiments of the present invention and can be adjusted appropriately according to actual situations), the corresponding sampling moment is the starting moment of monitoring.

[0033] Similarly, for the position points in the illuminated area, when the ratio of the gray value of the position point to the gray value of the illuminated area minus the ratio of the minimum value to the maximum value of the gray value of the illuminated area is greater than 0.1, the corresponding sampling moment is the starting moment of monitoring.

[0034] It should be noted that the gray value in the heat map reflects the temperature of each position point on the photovoltaic panel. The higher the gray value, the higher the temperature; the lower the gray value, the lower the temperature.

[0035] S3: Starting from the starting moment of monitoring, calculate the temperature rise anomaly index of each position point, traverse each position point in the heat map, and obtain the temperature rise anomaly matrix.

[0036] The starting moment of monitoring finally obtained in the above step S2 is to find a reasonable starting point. Starting from this starting point for analysis, the temperature rise anomaly of the photovoltaic panel can be more accurately identified and quantified.

[0037] In one embodiment, calculate the time difference from the starting moment of monitoring to the current moment for any position point, and calculate the ratio of the gray value of the position point at the current moment to the product of the gray value at the starting moment of monitoring and the time difference, to obtain a dimensionless relative value, which is the temperature rise anomaly index of the position point at the current moment.

[0038] If the temperature rise anomaly index of a certain position point is significantly higher than that of other position points, it may indicate that there is a fault at that position point.

[0039] Furthermore, traverse each position point in the heat map, calculate its temperature rise anomaly index, record the temperature rise anomaly index of each position point, and form a matrix. The value of the position point without the starting time of monitoring is 0, and the finally obtained matrix is the temperature rise anomaly matrix.

[0040] Through the above operations, the temperature rise anomaly of the photovoltaic panel can be more accurately identified and quantified, providing a solid foundation for subsequent fault diagnosis.

[0041] In another embodiment, considering the influence of different environmental conditions, give the temperature rise anomaly index of the position point at the current moment based on the temperature rise trend, that is, the following relationship is satisfied:

[0042]

[0043] Wherein, is the temperature rise anomaly index of the position point at the current moment is the temperature rise trend of the position point at the current moment is the temperature rise trend of the position point at the current moment is the temperature rise trend of the position point at the current moment is the gray value of the position point at the current moment is the gray value of the position point at the current moment is the gray value of the position point at the current moment is the gray value of the position point at the current moment is the total time from the start time of monitoring to the current moment is the mean value of the normal temperature rise trend. Among them, temperature rise monitoring can be carried out by collecting the temperature rise data of a group of normal photovoltaic panels, and the temperature rise trend of each position point is recorded.

[0044] S4: Find the non-zero connected regions in the temperature rise anomaly matrix, calculate the anomaly degree of each non-zero connected region. When the anomaly degree is less than the preset threshold, determine the corresponding non-zero connected region as the abnormal temperature rise area to complete the fault diagnosis of the photovoltaic panel.

[0045] It should be noted that if there is too much cloud cover or unstable light, it may cause inaccurate temperature data, that is, inaccurate gray values of the position points in the thermal image, resulting in misjudgment as the start time of monitoring. This will cause error points to appear in the temperature rise anomaly matrix.

[0046] In one embodiment, in order to distinguish the error points and real abnormal points in the temperature difference anomaly matrix, the error points and real abnormal points can be distinguished by analyzing the gradient change of the temperature rise anomaly index.

[0047] First, take any non-zero connected region (i.e., the region composed of adjacent non-zero temperature rise anomaly indexes) in the temperature rise anomaly matrix as an example. If the non-zero connected region is composed of error points, there is no gradient descent in the temperature rise anomaly index of the position points from the center to the surrounding of the non-zero connected region, that is, the surrounding position points will not be affected by heat and follow the temperature rise. On the contrary, if the non-zero connected region is an abnormal point, there is no gradient rise in the temperature rise anomaly index of the position points from the center to the surrounding of the non-zero connected region, and the surrounding position points will be affected by heat and follow the temperature rise of this position point.

[0048] Furthermore, in order to ensure that there is at least one direction with gradient descent, eight directions with an included angle of 45 degrees are set here, namely 0 degrees, 45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, 270 degrees, and 315 degrees.

[0049] Next, taking any non-zero connected domain in the temperature rise anomaly matrix as an example, the maximum value point of this non-zero connected domain is used as the center point. In each direction, starting from the center point, all adjacent points are traversed along this direction, the ratio of the temperature rise anomaly indices of each pair of adjacent points is calculated, and then the average value of these ratios is obtained to get the decrease amplitude of the temperature rise anomaly index in this direction.

[0050] Among them, taking any one direction as an example, the above decrease amplitude satisfies the relational expression:

[0051] In the formula, is the decrease amplitude of the temperature rise anomaly index of the center point in the 0-degree direction, is the temperature rise anomaly index of the current point , is the temperature rise anomaly index of the next position point of the current point (that is, the next point of the center point along the 0-degree direction), is the number of steps extended outward from the center point.

[0052] In another embodiment, another calculation method is also provided:

[0053] In the formula, is the decrease amplitude of the temperature rise anomaly index of the center point in the 0-degree direction, is the temperature rise anomaly index of the current point , is the temperature rise anomaly index of the next position point of the current point (that is, the next point of the center point along the 0-degree direction), is the number of steps extended outward from the center point.

[0054] The above-mentioned equal distribution of the total change rate to each step can reflect the average effect of the overall attenuation trend.

[0055] According to the above operations, the decrease amplitudes of the temperature rise anomaly indices of the center points in all directions can be calculated, and then the minimum value of the decrease amplitudes is retained as the anomaly degree of this non-zero connected domain.

[0056] Furthermore, the anomaly degrees of all non-zero connected domains are obtained. The calculated anomaly degree is compared with a preset threshold (set to 1 in the embodiment of the present invention). If the anomaly degree is less than the preset threshold, it indicates that the temperature rise of this non-zero connected domain is within the normal range, and it can be determined as a normal area; if the anomaly degree is greater than or equal to the preset threshold, it indicates that the temperature rise of this non-zero connected domain is abnormal, and then it is determined as an abnormal temperature rise area.

[0057] The system includes a processor and a memory, and the memory stores computer program instructions which, when executed by the processor, implement the photovoltaic panel fault diagnosis method based on image processing according to the first aspect of the present invention.

[0058] The system also includes other components well-known to those skilled in the art such as a communication bus and a communication interface, and their settings and functions are known in the art, so they will not be elaborated herein.

[0059] It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent shall be subject to the appended claims.

Claims

1. A photovoltaic panel fault diagnosis method based on image processing, characterized in that, Including: Obtain the thermal image of the photovoltaic panel in real time, and divide the thermal image into a light area and a shadow area; For the light area and the shadow area, obtain the monitoring start time of the temperature rise in the light area after the occlusion is removed and the monitoring start time of the temperature rise in the shadow area based on the preset trigger monitoring conditions; Starting from the monitoring start time, calculate the temperature rise anomaly index at each position point, traverse each position point in the thermal image, and obtain the temperature rise anomaly matrix; Find the non-zero connected domains in the temperature rise anomaly matrix, calculate the anomaly degree of each non-zero connected domain, and when the anomaly degree is greater than or equal to the preset threshold, determine the corresponding non-zero connected domain as the abnormal temperature rise area to complete the fault diagnosis of the photovoltaic panel.

2. The photovoltaic panel fault diagnosis method based on image processing according to claim 1, wherein The dividing the thermal image into a light area and a shadow area includes: Extract the pixel values of all pixel points in the thermal image, perform binary classification processing on these pixel values, and then divide the thermal image into a light area and a shadow area.

3. The photovoltaic panel fault diagnosis method based on image processing according to claim 2, wherein, The obtaining the monitoring start time of the temperature rise in the light area after the occlusion is removed and the monitoring start time of the temperature rise in the shadow area based on the preset trigger monitoring conditions includes: For the position points in the shadow area collected in real time, calculate the ratio of the gray value of this position point to the maximum gray value of the shadow area as the first parameter, and then calculate the ratio of the average gray value of the shadow area to the maximum gray value of the shadow area as the second parameter. When the difference between the first parameter and the second parameter is greater than the preset threshold, the corresponding sampling time is the monitoring start time; For the position points in the light area collected in real time, perform operations on the shadow area, and then obtain the corresponding monitoring start time.

4. A photovoltaic panel fault diagnosis method based on image processing according to claim 3, characterized in that, The calculation process of the temperature rise anomaly index includes: Calculate the time difference from the monitoring start time to the current time for any position point, calculate the ratio of the gray value of the position point at the current time to the product of the gray value at the monitoring start time and the time difference, and use it as the temperature rise anomaly index of this position point at the current time; traverse each position point in the thermal image to obtain the temperature rise anomaly index corresponding to each position point.

5. A photovoltaic panel fault diagnosis method based on image processing according to claim 4, characterized in that, The obtaining process of the non-zero connected domains in the temperature rise anomaly matrix includes: Form a non-zero connected domain by combining adjacent non-zero temperature rise anomaly indexes in the temperature rise anomaly matrix.

6. The photovoltaic panel fault diagnosis method based on image processing according to claim 5, wherein, The calculating the anomaly degree of each non-zero connected domain includes: Take the maximum value point of the non-zero connected domain as the center point, and in each set direction, starting from the center point, traverse all adjacent points along this direction, and calculate the decrease amplitude of the temperature rise anomaly index in each direction; Take the minimum value of the decrease amplitudes in all directions as the anomaly degree of this non-zero connected domain.

7. A photovoltaic panel fault diagnosis method based on image processing according to claim 6, characterized in that, The decrease amplitude satisfies the relational expression: ; is the decreasing amplitude of the temperature rise anomaly index in the 0-degree direction with the center point, is the current point of the temperature rise anomaly index, is the current point of the temperature rise anomaly index at the next position point, is the number of steps extending outward from the center point.

8. A method for diagnosing photovoltaic panel faults based on image processing according to claim 6, characterized in that, The decrease amplitude satisfies the relational expression: ; wherein, is the decreasing amplitude of the temperature rise anomaly index at the center point in the 0-degree direction, is the current point of the temperature rise anomaly index, is the current point the temperature rise anomaly index of the next position point, is the number of steps extending outward from the center point.

9. A photovoltaic panel fault diagnosis method based on image processing according to claim 2, characterized in that, The calculation process of the temperature rise anomaly index includes: Calculate the total time from the monitoring start time to the current time for any position point, and calculate the difference between the gray value of the position point at the current time and the gray value at the monitoring start time; Take the ratio of the calculated difference in gray value and the total time as the temperature rise trend at the current time of this position point; take the ratio of the temperature rise trend to the average value of the normal temperature rise trend as the temperature rise anomaly index; Traverse each position point in the thermal image to obtain the temperature rise anomaly index corresponding to each position point.

10. A photovoltaic panel fault diagnosis system based on image processing, characterized in that, Comprising: A processor and a memory, the memory storing computer program instructions which, when executed by the processor, implement the method for diagnosing faults of a photovoltaic panel based on image processing according to any one of claims 1-9.

Images