Method for generating photovoltaic array characteristic curve under pollution condition based on image processing

Abstract

The invention provides a method for generating a photovoltaic array characteristic curve under a pollution condition based on image processing. The method comprises the following steps: firstly, acquiring image data of a photovoltaic array; then, carrying out standardization, graying and segmentation processing on the photovoltaic array image; then, calculating an average gray value of the processed image, judging a threshold value and a range, and distinguishing a polluted photovoltaic module, an unpolluted photovoltaic module and a pollution type; then, based on threshold and range judgment, distinguishing the photovoltaic module images and calculating the equivalent illumination intensity of each photovoltaic module; and finally, based on the equivalent illumination intensity calculation result, generating an output characteristic curve of the photovoltaic array by using the mathematical model. According to the method, polluted photovoltaic modules and unpolluted photovoltaic modules in the photovoltaic array can be effectively distinguished, different pollution conditions can be identified, and the output characteristic curve of the photovoltaic array under different pollution conditions can be quickly and accurately generated by calculating the equivalent illumination intensity of each photovoltaic module under different pollution conditions.

Description

Technical Field

[0001] The present invention belongs to the field of photovoltaic system performance evaluation, and in particular relates to a method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing. Background Art

[0002] With the increasing share of renewable energy in power generation, solar photovoltaics (PV) are emerging as a prominent form of energy. According to the International Energy Agency (IEA), global PV installations reached 1,552.3 GW in 2023 and are projected to reach 1,954.6 GW in 2024, with 402.3 GW of new installations. The efficiency of PV power generation is attracting increasing attention from researchers. Unfortunately, the efficiency of these systems can be significantly reduced by various factors, including shading, potential-induced degradation, diode failure, and cell cracking. Contamination is a common factor in this efficiency reduction. Common forms of contamination include dust, shading, and shadows. Contamination prevents PV arrays from effectively absorbing solar radiation, significantly reducing their power generation. It can also create localized hotspots, leading to long-term PV failure. Therefore, effectively identifying and evaluating PV arrays under contaminated conditions is crucial to improving the efficiency of PV systems.

[0003] Previous studies have connected variable loads to the electrical terminals of PV modules to measure the module's voltage and current operating points. The corresponding IV characteristic curves were derived by curve fitting the measured data. The PV module's operating point is measured by adjusting a variable resistor. Furthermore, compared to purely resistive loads, capacitive loads offer simpler and more efficient measurement of the PV module's operating point. However, methods based on passive loads require manual adjustment of the load value, resulting in poor measurement results. As an alternative, active loads can be used instead of these passive loads. Specifically, a DC-DC converter can be used to vary the input resistance as a function of the duty cycle. Electronic loads based on transistors operating in the linear region can be used to measure the PV module's operating point. However, these solutions can only measure a single PV module at a time, resulting in extended measurement times and impractical for large PV arrays.

[0004] Unlike the aforementioned measurement methods, image analysis technology processes and analyzes various types of images of contaminated PV modules to identify and evaluate PV modules under contaminated conditions. Image inspection methods primarily use various types of images for qualitative fault analysis. Common methods include ultraviolet fluorescence, infrared thermography, and electroluminescence imaging. Ultraviolet fluorescence can visually identify aging faults in PV modules, but its image quality is easily affected by environmental factors, making it unsuitable for outdoor PV module inspection. Electroluminescence, based on the luminescence phenomenon produced when electrical energy is converted into light, can capture images of the radiant composite distribution of PV modules. However, this method requires darkroom testing and is also unsuitable for outdoor applications. Infrared thermography is a non-contact inspection technology that captures the uneven temperature distribution of PV modules during operation and uses a thermal imager to precisely locate faulty modules. While image inspection methods can visually identify aging faults in PV modules, their image quality is easily affected by environmental conditions, and the cost of the inspection equipment is generally high. Furthermore, they cannot generate output characteristic curves for PV arrays.

[0005] The present application differs from the prior art in the following ways:

[0006] Technical comparison with patent CN107168049B "Real-time acquisition method of photovoltaic array output characteristic curve"

[0007] First, patent CN107168049B uses a combination of radiation sensors and cameras, comprehensively considering the number of sensors and irradiance detection efficiency to determine sensor installation locations. This approach captures irradiance data for both illuminated and shadowed areas of the photovoltaic array, as well as images of the array's operation. However, our approach only requires capturing images of the photovoltaic array using a camera, eliminating the need for multiple radiation sensors and relying solely on image data for subsequent processing and analysis.

[0008] Second, patent CN107168049B utilizes digital image processing techniques such as iterative threshold segmentation and logarithmic operator edge detection to identify shaded modules. This involves a series of complex operations, including image preprocessing, grid line removal, filtering, boundary detection, and removal of irrelevant bright objects. Our approach, however, normalizes, grayscales, and segments the acquired photovoltaic array images. We then calculate the average grayscale value and apply threshold and range determination to distinguish contaminated and uncontaminated modules, as well as the type of contamination. This results in a more streamlined and efficient image processing process.

[0009] Third, patent CN107168049B primarily focuses on obtaining the output characteristic curve of a photovoltaic array under partial shading conditions. This takes into account the irradiance differences between different components in the photovoltaic array caused by partial shading, but its detection is too simplistic. Our patent, on the other hand, focuses on generating characteristic curves for photovoltaic arrays under polluted conditions, specifically addressing the impact of pollution factors such as dust, obstruction, and shadows on the power generation efficiency of photovoltaic arrays. This allows for the detection of multiple pollution types.

[0010] Technical comparison with patent CN111625048A "Global Maximum Power Point Tracking Method and Device"

[0011] First, patent CN111625048A uses image processing to identify shadow conditions in various areas of the photovoltaic array, classifying shadow coverage conditions and matching them to a preset power-voltage (PV) characteristic curve model. However, PV curve models must be built offline and rely on historical data, requiring manual updates for newly emerging shadow or pollution conditions. Our approach, however, generates characteristic curves through real-time image analysis, eliminating the need for offline data and automatically adapting to varying pollution conditions, offering greater generalization capabilities.

[0012] Second, the processing method in patent CN111625048A relies on pre-defined area divisions, which is not adaptable to irregularly shaped or unevenly distributed contaminated areas, and may result in some contaminated areas being missed or misidentified. Using image processing technology to standardize, grayscale, and segment photovoltaic array images, this method can accurately identify the contamination status of each photovoltaic module, regardless of module shape or contamination distribution, and has strong adaptability and accuracy. Summary of the Invention

[0013] This paper proposes a method for generating characteristic curves for photovoltaic arrays under polluted conditions based on image processing. This method overcomes the difficulties faced by traditional methods in effectively identifying polluted photovoltaic arrays and simultaneously generates their output characteristic curves. Contaminated photovoltaic modules are detected by processing and analyzing images of the photovoltaic array. The equivalent light intensity of each photovoltaic module under polluted conditions is then defined and calculated. By integrating these calculations into a mathematical model, the output characteristic curve of the photovoltaic array under polluted conditions is accurately generated.

[0014] To achieve the above object, the technical solution adopted by the present invention is:

[0015] The method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing is characterized by comprising the following steps:

[0016] Step S1: Acquire image data of the photovoltaic array;

[0017] Step S2: normalizing, graying and segmenting the photovoltaic array image;

[0018] Step S3: Calculate the average grayscale value of the processed image and perform threshold and range determination to distinguish between contaminated photovoltaic modules, uncontaminated photovoltaic modules, and pollution types;

[0019] Step S4: Based on the set threshold and range determination, the photovoltaic module images are distinguished and the equivalent light intensity of each photovoltaic module is calculated;

[0020] Step S5: Based on the calculation result of the equivalent light intensity, a characteristic curve of the photovoltaic array is generated using a mathematical model.

[0021] As a preferred technical solution of the present invention: in step S1, image data of the photovoltaic array is obtained by shooting with a camera.

[0022] As a preferred technical solution of the present invention: Step S2 is specifically as follows:

[0023] First, the photovoltaic array image that is an irregular convex quadrilateral in the original photovoltaic array image is converted into a rectangular image using perspective mapping, and then the color image is converted into a grayscale image using formula (1).

[0024] (1)

[0025] Finally, the grayscale image is segmented to obtain small images reflecting the pollution status of each photovoltaic module.

[0026] As a preferred technical solution of the present invention: Step S3 is as follows:

[0027] The average grayscale value of the PV module images under different pollution conditions is calculated and compared with the average grayscale value of the PV module images under normal conditions. The threshold for distinguishing contaminated PV modules from uncontaminated PV modules is set. In addition, the average grayscale value of the PV module images under different pollution conditions is used to divide the range of the average grayscale value of the PV module images under different pollution conditions to distinguish different pollution types.

[0028] As a preferred technical solution of the present invention: Step S4 is specifically as follows:

[0029] According to the selected threshold and range, when the average grayscale value of the photovoltaic module image is higher or lower than the threshold, the photovoltaic module is judged to be a contaminated photovoltaic module; when the average grayscale value of the photovoltaic module image is within a certain range, the photovoltaic module is judged to be polluted accordingly. Based on the above judgment results, the equivalent light intensity of each photovoltaic module is calculated according to the different degrees of shading caused by different pollution to the photovoltaic modules.

[0030] As a preferred technical solution of the present invention: Step S5 is specifically as follows:

[0031] Based on the calculation result of the equivalent light intensity obtained in step S4, the characteristic curves of the photovoltaic array under different pollution conditions are generated using the mathematical model of the photovoltaic module.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] This method uses perspective mapping to normalize irregular photovoltaic array images into rectangular structures. Combining grayscale conversion with a module segmentation algorithm, and using a mean grayscale threshold determination strategy, it can rapidly identify shadowed modules and classify their shadow types. Furthermore, by incorporating a mathematical model of photovoltaic modules, it can simultaneously identify contaminated modules and generate their output characteristic curves, overcoming the limitations of traditional single-image analysis and measurement methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of the method of the present invention;

[0035] Figure 2 yes Figure 1 Photovoltaic array images taken in;

[0036] Figure 3 yes Figure 1 Characteristic curve of the photovoltaic array in;

[0037] Figure 4 is a schematic diagram for photovoltaic array image processing in the present invention;

[0038] Figure 5 is the range of average grayscale values of the photovoltaic module images under different pollution conditions in the present invention;

[0039] Figure 6 It is a comparison between the photovoltaic array output characteristic curve generated in the embodiment of the present invention and the actual curve. DETAILED DESCRIPTION

[0040] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0041] like Figure 1-3 As shown, the method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing technology proposed in the present invention includes the following steps:

[0042] Step S1: Acquire image data of the photovoltaic array;

[0043] Step S2: Standardize, grayscale and segment the photovoltaic array image:

[0044] In this embodiment, the perspective mapping technology is first used to convert the original photovoltaic array image, which is an irregular convex quadrilateral, into a rectangular image. Figure 4(2) As shown. Secondly, use formula (1) to convert the color image into a grayscale image, as shown in Figure 4 (3). Then, the grayscale image is segmented as follows: Figure 4 As shown in (4), a small image reflecting the pollution status of each photovoltaic module is obtained.

[0045] (1)

[0046] Step S3: Calculate the average grayscale value of the processed image and perform threshold and range determination to distinguish between contaminated photovoltaic modules, uncontaminated photovoltaic modules, and pollution types:

[0047] In this embodiment, the shadowed photovoltaic modules are detected and located by examining the segmented images and identifying images with abnormal grayscale gradients. The average grayscale value of each photovoltaic module image is calculated as:

[0048]

[0049] like <𝜀 or >𝜀 (𝜀 is determined by the average grayscale value of all modules), then the photovoltaic module is determined to be a shadow module. The judgment result is as follows Figure 4 (5) shows that the shadow types are distinguished according to the average grayscale value range corresponding to different shadow types. The range diagram of the average grayscale value of photovoltaic module images under different pollution conditions is shown in Figure 5 shown.

[0050] Step S4: Based on the set threshold and range determination, the photovoltaic module images are distinguished and the equivalent light intensity of each photovoltaic module is calculated:

[0051] In this embodiment, based on the threshold and range determination, contaminated PV modules are distinguished from uncontaminated PV modules, and the contamination type is determined. Taking into account the losses caused by contamination, the equivalent solar irradiance incident on the surface of the PV module can be mathematically modeled as follows (the transmission coefficient 𝜏 represents the ratio of transmitted light to incident light):

[0052]

[0053] Step S5: Based on the calculation result of equivalent light intensity, a characteristic curve of the photovoltaic array is generated using a mathematical model:

[0054] In this embodiment, based on the single diode model, the open circuit voltage Voc and short circuit current Isc under actual operating conditions are calculated according to the equivalent light intensity calculated above and the measured temperature. The implicit transcendental equation is solved using the piecewise linear method (PLM) function, and the voltage and current in the photovoltaic cell expression are decoupled to obtain the explicit equation for the photovoltaic module output current. Considering the effect of the bypass diode on the IV characteristics, the bypass diode is connected in parallel to the photovoltaic cell string and turns on at negative voltage, affecting the IV characteristics of the photovoltaic cell string. By solving = Get the short-circuit current of the photovoltaic cell string , using piecewise functions of Vcs to accurately model submodule voltages.

[0055] For a photovoltaic array consisting of m×n photovoltaic modules, the voltage and current are calculated using the following formula to obtain the output characteristic curve of the photovoltaic array. The output result is as follows: Figure 6 As shown, the generated output characteristic curve has a high consistency with the actually measured output characteristic curve, which illustrates the effectiveness of the method of the present invention.

[0056]

[0057] .

[0058] Based on the above method, the present invention uses perspective mapping to normalize irregular photovoltaic array images into rectangular structures. Combining grayscale conversion with a module segmentation algorithm, and using a mean grayscale threshold determination strategy, it can rapidly identify shadowed modules and classify their shadow types. Furthermore, by incorporating a mathematical model of photovoltaic modules, the proposed method can simultaneously identify contaminated modules and generate their output characteristic curves, overcoming the limitations of traditional single-image analysis and measurement methods.

[0059] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A method for generating characteristic curves of photovoltaic arrays under pollution conditions based on image processing, characterized in that: The steps include: Step S1: Acquire image data of the photovoltaic array; Step S2: normalizing, graying and segmenting the photovoltaic array image; Step S3: Calculate the average grayscale value of the processed image and perform threshold and range determination to distinguish between contaminated photovoltaic modules, uncontaminated photovoltaic modules, and pollution types; Step S4: Based on the set threshold and range determination, the photovoltaic module images are distinguished and the equivalent light intensity of each photovoltaic module is calculated; Step S5: Based on the calculation result of the equivalent light intensity, a characteristic curve of the photovoltaic array is generated using a mathematical model.

2. The method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing according to claim 1, characterized in that: In step S1 , image data of the photovoltaic array is obtained by photographing with a camera.

3. The method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing according to claim 1, characterized in that: Step S2 is specifically as follows: First, the photovoltaic array image that is an irregular convex quadrilateral in the original photovoltaic array image is converted into a rectangular image using perspective mapping, and then the color image is converted into a grayscale image using formula (1). （1） Finally, the grayscale image is segmented to obtain small images reflecting the pollution status of each photovoltaic module.

4. The method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing according to claim 1, characterized in that: Step S3 is as follows: The average grayscale value of the PV module images under different pollution conditions is calculated and compared with the average grayscale value of the PV module images under normal conditions. The threshold for distinguishing contaminated PV modules from uncontaminated PV modules is set. In addition, the average grayscale value of the PV module images under different pollution conditions is used to divide the range of the average grayscale value of the PV module images under different pollution conditions to distinguish different pollution types.

5. The method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing according to claim 1, characterized in that: Step S4 is specifically as follows: According to the selected threshold and range, when the average grayscale value of the photovoltaic module image is higher or lower than the threshold, the photovoltaic module is judged to be a contaminated photovoltaic module; when the average grayscale value of the photovoltaic module image is within a certain range, the photovoltaic module is judged to be polluted accordingly. Based on the above judgment results, the equivalent light intensity of each photovoltaic module is calculated according to the different degrees of shading caused by different pollution to the photovoltaic modules.

6. The method for generating a photovoltaic array characteristic curve under pollution conditions based on image processing according to claim 1, characterized in that: Step S5 is specifically as follows: Based on the calculation result of the equivalent light intensity obtained in step S4, the characteristic curves of the photovoltaic array under different pollution conditions are generated using the mathematical model of the photovoltaic module.

Images