Method and device for detecting quality of photovoltaic module

Abstract

The invention provides a photovoltaic module quality detection method and device, relates to the technical field of photovoltaic module detection, and mainly aims to capture defect changes of a photovoltaic module under a dynamic condition and improve the accuracy of quality detection. According to the main technical scheme of the invention, current control parameters are determined according to the detection requirements of the photovoltaic module and the module characteristics of the photovoltaic module, wherein the current control parameters are used for representing parameters for controlling the dynamic current to be injected into the photovoltaic module; performing dynamic current injection on the photovoltaic module according to the current control parameters, and performing image acquisition on the photovoltaic module in the dynamic current injection process to obtain a light-emitting image sequence; performing defect identification and evaluation on the photovoltaic module based on the light-emitting image sequence to obtain defect information corresponding to the photovoltaic module; and generating a quality detection report corresponding to the photovoltaic module according to the defect information. The method and the device are used for realizing quality detection of the photovoltaic module through dynamic current injection.

Description

Technical Field

[0001] The present disclosure relates to the technical field of photovoltaic module detection, and in particular to a photovoltaic module quality detection method and device. Background Art

[0002] As the global demand for renewable energy continues to grow, the quality and performance of photovoltaic modules, as the core components of solar power generation systems, directly affect the efficiency and life of the entire system. Therefore, it is particularly important to conduct efficient and accurate quality inspection of photovoltaic modules. At present, electroluminescence (EL) inspection has become one of the most commonly used inspection methods for the quality inspection of photovoltaic modules due to its high sensitivity and non-destructive characteristics.

[0003] At present, the EL detection of photovoltaic modules in the existing technology usually involves injecting a constant current into the photovoltaic modules, collecting the EL images of the photovoltaic modules under the constant current conditions, and identifying and evaluating defects through image processing technology. However, due to the constant current, this method is difficult to capture the defect changes of photovoltaic modules under dynamic conditions, and cannot fully stimulate some minor or potential defects, resulting in missed detection, thereby reducing the accuracy of quality inspection. Summary of the invention

[0004] A technical problem to be solved by the present disclosure is: how to capture the defect changes of photovoltaic modules under dynamic conditions to improve the accuracy of quality inspection.

[0005] In order to solve the above technical problems, in a first aspect, an embodiment of the present disclosure provides a photovoltaic module quality detection method, comprising:

[0006] Determine a current control parameter according to the detection requirements of the photovoltaic module and the component characteristics of the photovoltaic module, and the current control parameter is used to characterize the parameters for controlling the dynamic current injection into the photovoltaic module;

[0007] Dynamically injecting current into the photovoltaic module according to current control parameters, and collecting images of the photovoltaic module during the dynamic current injection process to obtain a luminous image sequence;

[0008] Defects of photovoltaic modules are identified and evaluated based on the luminous image sequence to obtain defect information corresponding to the photovoltaic modules;

[0009] Generate a quality inspection report corresponding to the photovoltaic module based on the defect information.

[0010] In some embodiments, determining the current control parameter according to the detection requirements of the photovoltaic component and the component characteristics of the photovoltaic component includes:

[0011] According to the detection requirements, the current waveform and waveform frequency are determined in a preset defect-waveform relationship table. The defect-waveform relationship table is determined based on the degree of defect manifestation under different current waveforms;

[0012] Determine the current amplitude range in a preset power-amplitude range relationship table according to the rated power of the photovoltaic module;

[0013] The current waveform, waveform frequency and current amplitude range are used as current control parameters.

[0014] In some embodiments, dynamic current injection is performed on the photovoltaic assembly according to the current control parameter, and images of the photovoltaic assembly are collected during the dynamic current injection process to obtain a sequence of luminous images, including:

[0015] The waveform distortion corresponding to the current waveform is calculated in real time during the dynamic current injection process;

[0016] If the waveform distortion exceeds a preset distortion threshold, the current amplitude range is reduced according to a preset gradient, and the number of distortions corresponding to the current waveform is recorded;

[0017] If the distortion times reach the preset times threshold, the detection stop alarm is triggered.

[0018] In some embodiments, defect identification and evaluation of a photovoltaic module is performed based on a sequence of luminous images to obtain defect information corresponding to the photovoltaic module, including:

[0019] Determine the defect area according to the brightness change corresponding to the luminous image sequence, and extract features in the luminous image sequence according to the defect area to obtain target features, which include static features and dynamic features;

[0020] The static features and dynamic features are input into the trained defect recognition model to obtain defect information, which includes the defect position, defect quantity, defect area, defect type and defect severity corresponding to the defect area.

[0021] In some embodiments, the defect area is determined according to the brightness change corresponding to the luminous image sequence, and features are extracted from the luminous image sequence according to the defect area to obtain target features, including:

[0022] Calculating the brightness value of each pixel in each image included in the luminous image sequence, and forming a brightness-time curve according to the change of the brightness value in the luminous image sequence;

[0023] Based on the brightness-time curve, the luminous image sequence is segmented using the adaptive threshold method to obtain the brightness abnormality area;

[0024] Use edge detection algorithm to identify the edge of the abnormal brightness area and obtain the defect area;

[0025] Dynamic feature extraction is performed on the defect area to obtain brightness change features and area change features. The brightness change features include brightness change rate, curvature and fluctuation amplitude.

[0026] Static features are extracted from the defect area to obtain grayscale features, texture features and shape features, where the shape features at least include position, area and perimeter.

[0027] In some embodiments, the defect recognition model includes a classification sub-model and an evaluation sub-model; static features and dynamic features are input into the trained defect recognition model to obtain the defect location, defect quantity, defect area, defect type and defect severity corresponding to the defect area, including:

[0028] Grayscale features, texture features and shape features are input into the classification sub-model to obtain the defect position, defect quantity, defect area and defect type corresponding to the defect area. The classification sub-model is used to classify and count the defects;

[0029] The brightness change characteristics and area change characteristics are input into the evaluation sub-model to obtain the defect severity corresponding to the defect area. The evaluation sub-model is used to evaluate the severity of the defects.

[0030] In some embodiments, generating a quality inspection report corresponding to a photovoltaic module according to the defect information includes:

[0031] Generate a defect distribution map corresponding to the photovoltaic module based on the defect position and defect type corresponding to the defect area;

[0032] Generate defect statistics based on the defect quantity, defect area, defect size, and defect severity corresponding to the defect area;

[0033] Use defect distribution diagram and defect statistics as quality inspection report.

[0034] In some embodiments, before using the defect distribution map and the defect statistical information as a quality inspection report, the method further includes:

[0035] Conduct sunlight simulation on defective areas in photovoltaic modules according to standard test conditions;

[0036] Collect the electrical performance parameters of the defective area during the sunlight simulation irradiation process, and determine the loss power corresponding to the defective area based on the electrical performance parameters;

[0037] Determine the degree of impact of the defect area on the performance of the photovoltaic module based on the defect type and its probability, defect location, defect quantity, defect area, defect severity and loss power corresponding to the defect area;

[0038] Generate defect statistics based on the defect quantity, defect area, defect size, and defect severity corresponding to the defect area, including:

[0039] Defect statistics are generated based on the defect quantity, defect area, defect severity, electrical performance parameters, and performance impact corresponding to the defect area.

[0040] In order to solve the above technical problems, in a second aspect, the embodiment of the present disclosure provides a photovoltaic module quality detection device, including:

[0041] A parameter determination unit, used to determine a current control parameter according to a detection requirement for the photovoltaic module and a component characteristic of the photovoltaic module, wherein the current control parameter is used to characterize a parameter for controlling dynamic current injection into the photovoltaic module;

[0042] A detection control unit, used for dynamically injecting current into the photovoltaic module according to the current control parameter obtained by the parameter determination unit, and collecting images of the photovoltaic module during the dynamic current injection process to obtain a luminous image sequence;

[0043] An identification and evaluation unit, used to identify and evaluate defects of the photovoltaic module based on the luminous image sequence obtained by the detection control unit, and obtain defect information corresponding to the photovoltaic module;

[0044] The quality feedback unit is used to generate a quality inspection report corresponding to the photovoltaic module based on the defect information obtained by the identification and evaluation unit.

[0045] To solve the above technical problems, in a third aspect, an embodiment of the present disclosure provides a storage medium, the storage medium includes a stored program, wherein when the program is running, the device where the storage medium is located is controlled to execute the photovoltaic module quality detection method of the first aspect.

[0046] To solve the above technical problems, in a fourth aspect, an embodiment of the present disclosure provides a processor, which is used to run a program, wherein the photovoltaic module quality detection method of the first aspect is executed when the program is running.

[0047] Through the above technical scheme, the present disclosure provides a photovoltaic module quality detection method and device, when it is necessary to perform quality detection on the photovoltaic module, first determine the current control parameter according to the detection requirements of the photovoltaic module and the component characteristics of the photovoltaic module, the current control parameter is used to characterize the parameters of controlling the dynamic current injection into the photovoltaic module, then perform dynamic current injection on the photovoltaic module according to the current control parameter, and collect images of the photovoltaic module during the dynamic current injection process to obtain a luminous image sequence, then perform defect identification and evaluation on the photovoltaic module based on the luminous image sequence to obtain defect information corresponding to the photovoltaic module, and finally generate a quality detection report corresponding to the photovoltaic module based on the defect information. The injected current can be controlled according to the detection requirements and component characteristics, that is, the dynamic current is injected into the photovoltaic module, so that the collected EL image sequence can capture the defect changes under dynamic conditions, thereby discovering dynamic defects that cannot be discovered by constant current detection, significantly improving the detection sensitivity, and discovering more potential defects, and performing defect identification and evaluation through the obtained EL image sequence, so that the defect information is richer and more comprehensive, avoiding the occurrence of missed detection, and improving the accuracy of photovoltaic module quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 is a flow chart of a photovoltaic module quality detection method disclosed in an embodiment of the present disclosure;

[0050] Figure 2 is a flow chart of another photovoltaic module quality detection method disclosed in an embodiment of the present disclosure;

[0051] Figure 3 It is a block diagram of a photovoltaic module quality detection device disclosed in an embodiment of the present disclosure;

[0052] Figure 4 It is a block diagram of another photovoltaic module quality detection device disclosed in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0053] The following is a further detailed description of the embodiments of the present disclosure in conjunction with the accompanying drawings and examples. The detailed description of the following embodiments and the accompanying drawings are used to exemplarily illustrate the principles of the present disclosure, but cannot be used to limit the scope of the present disclosure. The present disclosure can be implemented in many different forms and is not limited to the specific embodiments disclosed herein, but includes all technical solutions that fall within the scope of the claims.

[0054] The present disclosure provides these embodiments to make the present disclosure thorough and complete, and to fully express the scope of the present disclosure to those skilled in the art. It should be noted that unless otherwise specifically stated, the relative arrangement of the parts and steps, the composition of the materials, the numerical expressions and the numerical values ​​set forth in these embodiments should be interpreted as being merely exemplary, and not as limiting.

[0055] It should be noted that, in the description of the present disclosure, unless otherwise specified, the meaning of "multiple" is greater than or equal to two; the terms "upper", "lower", "left", "right", "inner", "outer", etc., indicating the orientation or positional relationship, are only for the convenience of describing the present disclosure and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present disclosure. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0056] In addition, the words "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts. "Vertical" does not mean vertical in the strict sense, but is within the tolerance range. "Parallel" does not mean parallel in the strict sense, but is within the tolerance range. "Include" or "comprising" and similar words mean that the elements before the word include the elements listed after the word, and do not exclude the possibility of including other elements.

[0057] It should also be noted that in the description of the present disclosure, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present disclosure can be understood according to the specific circumstances. When a specific device is described as being located between a first device and a second device, there may or may not be an intermediate device between the specific device and the first device or the second device.

[0058] All terms used in the present disclosure have the same meanings as those understood by those of ordinary skill in the art to which the present disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in general dictionaries, for example, should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0059] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0060] At present, the EL detection of photovoltaic modules in the existing technology usually involves injecting a constant current into the photovoltaic modules, collecting the EL images of the photovoltaic modules under the constant current conditions, and identifying and evaluating defects through image processing technology. However, due to the constant current, this method is difficult to capture the defect changes of photovoltaic modules under dynamic conditions, and cannot fully stimulate some minor or potential defects, resulting in missed detection, thereby reducing the accuracy of quality inspection.

[0061] After research, the inventor found that a dynamic current injection control parameter can be defined according to the characteristics of the component and the requirements of the detection, that is, dynamic current injection can be used to fully stimulate small or potential defects, and the corresponding EL image sequence is collected during the injection process, so that the EL image sequence can accurately capture the defect changes of the photovoltaic component under dynamic conditions, and then the defect identification and evaluation are performed through the EL image sequence to obtain defect information, and finally a quality inspection report is generated based on this. In this way, the EL image sequence can capture the defect changes under dynamic conditions through the injection of dynamic current, so as to discover dynamic defects that cannot be discovered by constant current detection, significantly improve the sensitivity of detection, and be able to discover more potential defects, and on this basis, defect identification and evaluation are performed, so that the defect information is richer and more comprehensive, avoiding the occurrence of missed detection, and improving the accuracy of photovoltaic component quality inspection.

[0062] Based on the above considerations, the embodiments of the present disclosure provide a photovoltaic module quality detection method and device. The photovoltaic module quality detection method and device provided in the embodiments of the present application are suitable for a darkroom where a programmable power supply and an infrared camera array are deployed. The photovoltaic module EL detection is achieved by controlling the programmable power supply to inject current into the photovoltaic module and controlling the infrared camera array to collect images of the photovoltaic panel.

[0063] like Figure 1 As shown, the embodiment of the present disclosure provides a photovoltaic module quality detection method, through which the defect changes of photovoltaic modules under dynamic conditions can be captured, and the accuracy of quality detection can be improved. The specific execution steps include:

[0064] 101. Determine current control parameters according to detection requirements for photovoltaic modules and module characteristics of photovoltaic modules.

[0065] Among them, the current control parameter is used to characterize the parameters for controlling the dynamic current injection into the photovoltaic module.

[0066] In this step, the detection requirement refers to the requirement of the tester to detect the defect types contained in the photovoltaic module. Specifically, the use scenarios and expected performance of the photovoltaic module can be analyzed to determine which defect types need to be detected, such as microcracks, broken grids, poor welding, hidden cracks, impurity contamination, etc. It should be noted that due to different defect types under different current waveforms, waveform frequencies and other conditions, the degree of manifestation is different. For example, high-frequency square waves may be more likely to excite microcracks, while low-frequency sine waves may be more suitable for detecting hidden cracks, etc. Therefore, different current waveforms and waveform frequencies can be tested in advance for the degree of manifestation of different defect types, and a corresponding relationship table between defect types and waveforms and frequencies can be established and maintained based on the test results. The waveforms include but are not limited to sine waves, square waves, triangle waves, etc., and the waveform frequencies include but are not limited to low frequencies (several hertz to tens of hertz), medium frequencies (hundreds of hertz to several thousand hertz), high frequencies (tens of kilohertz to hundreds of kilohertz), etc. The current waveform and waveform frequency corresponding to the detection requirements can be quickly determined through this relationship table. Similarly, photovoltaic modules of different specifications have different electrical characteristics, including rated power (Pmp), maximum power point current (Imp), maximum power point voltage (Vmp), etc. Therefore, in order to prevent overload and thermal effects during current injection to ensure safety, and to take into account effectiveness, the current amplitude range can be determined according to the rated power in the module characteristics. The rated power determines its maximum current and voltage under normal working conditions, with the maximum current as the upper limit, and a safety margin is set, for example, the current amplitude upper limit is set to 80% to 90% of the maximum current, combined with the above-mentioned tests on current waveform and waveform frequency, a corresponding relationship table between different rated powers and different current amplitude ranges is established, and the current amplitude range corresponding to the module characteristics can be quickly determined through this relationship table. Using the current waveform, waveform frequency and current amplitude range as current control parameters in subsequent steps can achieve dynamic current injection into photovoltaic modules.

[0067] For the determination process of current control parameters, a custom interface can be set up, which contains options for defect type, rated power, etc. Through the tester's selection and confirmation, the corresponding current control parameters are automatically determined according to the above relationship table. In addition, the current injection duration of the programmable power supply and the sampling rate and sampling duration of the infrared camera array can also be set in the custom interface to accurately control the injection of dynamic current and the acquisition of EL images.

[0068] 102. Dynamically inject current into the photovoltaic module according to the current control parameters, and collect images of the photovoltaic module during the dynamic current injection process to obtain a luminous image sequence.

[0069] In actual applications, photovoltaic modules will be placed at the detection points in a darkroom. At the same time, a programmable power supply for current injection and an infrared camera array for EL image acquisition will be deployed in the darkroom. The infrared camera array corresponds to the detection points, and the output end of the programmable power supply is connected to the positive and negative poles of the photovoltaic module. In this step, the current control parameters are for the programmable power supply, and the programmable power supply injects dynamic current into the photovoltaic module according to the current control parameters. At the same time, during the injection process, the infrared camera array shoots the photovoltaic module according to the preset sampling rate and sampling duration to collect the EL image during the dynamic current injection process to form an EL image sequence, that is, a luminous image sequence. The luminous image sequence is stored for subsequent processing and analysis. Through the injection of dynamic current, the EL image sequence captures the defect changes under dynamic conditions, thereby discovering dynamic defects that cannot be discovered by constant current detection, that is, discovering more potential defects.

[0070] During the dynamic current injection process, it is also possible to monitor whether the waveform is distorted, such as by presetting a distortion threshold and comparing the waveform distortion with the distortion threshold, so as to adaptively adjust the current amplitude range or directly stop the test and issue an alarm, etc. This ensures the effect of dynamic current injection and avoids affecting the image quality of the luminous image sequence.

[0071] 103. Defects of photovoltaic modules are identified and evaluated based on the luminous image sequence to obtain defect information corresponding to the photovoltaic modules.

[0072] In this step, EL image sequences with defective areas can be collected as samples in advance, and information such as type, location, etc. can be marked on the defective areas in the samples. A machine learning model can be trained using the samples as a defect recognition model. The machine learning model can be a support vector machine (SVM), a random forest (RF), or a neural network (FCN). This allows the defect recognition model to identify defect information based on the features of the EL image sequence. The defect information includes the defect location, defect quantity, defect area, defect type, and the evaluation of the corresponding defect severity of the defect area. It should be noted that the model can be a fusion model, that is, classification statistics and complexity evaluation are performed simultaneously, or it can be a split model, that is, two sub-models are set, which are used for classification statistics and complexity evaluation respectively.

[0073] Since the luminous image sequence contains multiple continuous frames of images, no matter which defect type is used, it will affect the brightness of the image. Therefore, the luminous image sequence can be segmented according to the brightness change of the luminous image sequence to obtain the defect area contained therein, which can be one or more. Since the defect area runs through the entire luminous image sequence, the static features and dynamic features of the defect area can be extracted, and the static features include but are not limited to grayscale features, texture features and shape features, etc., wherein the shape features include but are not limited to position, area, perimeter, circularity, etc., and the dynamic features include but are not limited to brightness change features, area change features, etc., wherein the brightness change features include but are not limited to brightness change rate, curvature, fluctuation amplitude, etc. The extracted features are output to the above-mentioned defect recognition model, so that it outputs the defect position, defect number, defect area, defect type and defect severity corresponding to the defect area. Among them, the defect position refers to the center coordinates or edge coordinates of each defect area, the defect number refers to the total number of all independent defect areas divided according to the defect type, the defect area refers to the area calculated for each defect area and the total area of ​​the same defect type, the defect type refers to the defect type of each defect area, and the defect severity refers to the severity of each defect area.

[0074] It should be noted that for the identification of defect types, the similarity probability between the defect area and each defect type can be additionally output, and the defect type with the highest similarity probability can be selected as the final output. Alternatively, a similarity probability threshold can be set and the defect type above the threshold can be used as the final output.

[0075] 104. Generate a quality inspection report corresponding to the photovoltaic module based on the defect information.

[0076] In this step, since the defect information includes defect location, defect number, defect area, defect type and defect severity. Therefore, in order to make the quality inspection report more detailed and comprehensive. A report format can be pre-set, such as including a defect distribution map and defect statistical information. That is, the defect distribution map of the photovoltaic module is generated by the defect location and defect type corresponding to the defect area, and the defect statistical information of the photovoltaic module is generated according to the defect number, defect area, defect area and defect severity corresponding to the defect area. When generating the defect distribution map, the defect location and defect type can be added as labels in the defect area, and the defect area can be highlighted, or it can be highlighted in different colors according to the different defect types. This embodiment does not limit this. In addition, the quality inspection report can also include IV testing of the defect area to obtain the degree of its impact on the performance of the photovoltaic module.

[0077] Based on the above Figure 1It can be seen from the implementation method that the photovoltaic module quality detection method provided by the present disclosure is that when the photovoltaic module needs to be quality-tested, the current control parameter is first determined according to the detection requirements of the photovoltaic module and the component characteristics of the photovoltaic module, and the current control parameter is used to characterize the parameters of controlling the dynamic current injection into the photovoltaic module, and then the photovoltaic module is dynamically injected with current according to the current control parameter, and the photovoltaic module is imaged during the dynamic current injection process to obtain a luminous image sequence, and then the photovoltaic module is defect-identified and evaluated based on the luminous image sequence to obtain the defect information corresponding to the photovoltaic module, and the defect information includes the defect position, defect number, defect area, defect type and defect severity corresponding to the defect area, and finally the quality detection report corresponding to the photovoltaic module is generated according to the defect information. The injected current can be controlled according to the detection requirements and component characteristics, that is, the dynamic current is injected into the photovoltaic module, so that the collected EL image sequence can capture the defect changes under dynamic conditions, thereby discovering dynamic defects that cannot be discovered by constant current detection, significantly improving the detection sensitivity, and discovering more potential defects, and defect identification and evaluation through the obtained EL image sequence, so that the defect information is richer and more comprehensive, avoiding the occurrence of missed detection, and improving the accuracy of photovoltaic module quality detection.

[0078] like Figure 2 As shown, the preferred embodiment of the present disclosure is in the above Figure 1 Based on this, a detailed description of the process of photovoltaic module quality inspection is given, and the specific steps include:

[0079] 201. Determine a current waveform and a waveform frequency in a preset defect-waveform relationship table according to detection requirements.

[0080] The defect-waveform relationship table is determined based on the test of the degree of defect appearance under different current waveforms.

[0081] In this step, a defect-waveform relationship table is established in advance based on the degree of manifestation of different defect types under different current waveforms and waveform frequencies. The power-amplitude range relationship table is used to characterize the correspondence between different defect types and waveforms and frequencies. Specifically, a set of standard photovoltaic modules can be selected to perform EL detection under different current waveforms (such as sine waves, square waves, triangle waves, etc.) and frequencies (low frequency, medium frequency and high frequency), and record the degree of manifestation of various defects (such as brightness changes, area changes, edge clarity, etc.) under different current waveforms and frequencies. By analyzing the test data, the best manifestation degree of each defect type under different current waveforms and frequencies is determined, and the test results are organized into a defect-waveform relationship table to clarify the best detection waveform and frequency for each defect type.

[0082] Since the detection requirement refers to the requirement of the tester to detect the defect type contained in the photovoltaic module, the defect type to be detected can be directly determined by the detection requirement, matched in the defect-waveform relationship table, and the appropriate current waveform and waveform frequency can be determined from the defect-waveform relationship table as the parameters of dynamic current injection.

[0083] 202. Determine a current amplitude range in a preset power-amplitude range relationship table according to the rated power of the photovoltaic module.

[0084] In this step, a plurality of photovoltaic modules of different specifications are obtained in advance, and the corresponding rated power and other relevant technical parameters are obtained. Specifically, the technical specification information of the photovoltaic module can be collected to obtain the rated power (Pmp), the maximum power point current (Imp) and the maximum power point voltage (Vmp). According to the rated power of the photovoltaic module, a power-amplitude range relationship table is established. According to the maximum power point current (Imp), the current amplitude is set not to exceed the maximum power point current. In order to ensure safety, a certain safety margin is usually set, for example, the current amplitude is set to 80% to 90% of the maximum power point current. According to the detection requirements, a suitable current amplitude range is selected, for example, dynamic current injection is performed between 0.5Imp and 0.9Imp, and photovoltaic modules with different rated powers are established with the corresponding current amplitude range to form a table, that is, a power-amplitude range relationship table. According to the rated power of the photovoltaic module in this step, the current amplitude range is determined from the power-amplitude range relationship table as a parameter for dynamic current injection.

[0085] 203. The current waveform, waveform frequency and current amplitude range are used as current control parameters.

[0086] By integrating the parameters obtained in steps 201 and 202, current control parameters for controlling the dynamic current injection into the photovoltaic module can be obtained.

[0087] According to steps 201-203, dynamic current injection can be subsequently implemented on the photovoltaic module, thereby fully stimulating certain tiny or potential defects, so that the subsequent EL image sequence can capture the defect changes of the photovoltaic module under dynamic conditions.

[0088] 204. Dynamically inject current into the photovoltaic module according to the current control parameter, and collect images of the photovoltaic module during the dynamic current injection process to obtain a luminous image sequence.

[0089] This step is combined with the description of step 102 in the above method, and the same content is not repeated here. It should be noted that the dynamic current injection is performed on the photovoltaic module according to the current control parameter, and in the process of capturing the image of the photovoltaic module during the dynamic current injection process, the waveform distortion corresponding to the current waveform is calculated in real time during the dynamic current injection process; if the waveform distortion exceeds the preset distortion threshold, the current amplitude range is reduced according to the preset gradient, and the distortion times corresponding to the current waveform are recorded; if the distortion times reach the preset times threshold, the detection stop alarm is triggered.

[0090] In this step, a distortion calculation method can be predefined to accurately reflect the distortion of the current waveform. The distortion calculation method includes but is not limited to total harmonic distortion, peak distortion, root mean square distortion, etc. During the dynamic current injection process, a high-precision current sensor can be used to collect current waveform data in real time, and the distortion of the current waveform can be calculated in real time according to the selected distortion calculation method. At the same time, a preset distortion threshold is determined in advance according to the characteristics and detection requirements of the photovoltaic module. The distortion threshold can be determined by experimental testing of the performance changes of the photovoltaic module under different distortions. A certain safety margin can also be set on this basis to ensure that the distortion threshold can ensure the detection effect while not causing damage to the photovoltaic module. For example, the preset distortion threshold is 10%. The calculated waveform distortion is compared with the distortion threshold. If it is greater than, it means that the upper limit of the current amplitude range at this time is too large, so the upper limit of the current amplitude range can be reduced to achieve the reduction of the current amplitude range, thereby avoiding the current waveform distortion during the dynamic current injection process and ensuring the test effect. Specifically, an adjustment gradient can be pre-set. For example, if the current amplitude range is 0.1Isc to 1.2Isc (Isc is the short-circuit current), the adjustment gradient is 0.2Isc. Since the distortion of the current waveform can occur multiple times, in addition to gradually narrowing the current amplitude range according to the preset gradient. The number of distortions can also be recorded. When the number of distortions is large, it means that continuing the dynamic current injection process poses a risk of damage to the photovoltaic components. Therefore, a number threshold can be set to determine whether to stop the detection by judging whether the number of distortions reaches the number threshold, that is, triggering the detection stop alarm. The detection stop alarm can notify the tester through sound and light alarms, SMS notifications, and other methods.

[0091] 205. Determine a defective area according to a brightness change corresponding to the luminous image sequence, and extract features from the luminous image sequence according to the defective area to obtain target features.

[0092] In this step, the luminous image sequence can be preprocessed in advance to improve the image quality, including using Gaussian filtering or other denoising algorithms to remove noise in the image, using histogram equalization or other image enhancement techniques to improve the image contrast, and normalizing the image brightness value to between 0 and 1 for subsequent processing.

[0093] It should be noted that the specific execution process of determining the defect area according to the brightness change corresponding to the luminous image sequence and extracting features in the luminous image sequence according to the defect area to obtain the target features is as follows: calculating the brightness value of each pixel in each image contained in the luminous image sequence, and forming a brightness-time curve according to the change of the brightness value in the luminous image sequence; based on the brightness-time curve, segmenting the luminous image sequence using an adaptive threshold method to obtain the brightness abnormality area; using an edge detection algorithm to perform edge recognition on the brightness abnormality area to obtain the defect area; performing dynamic feature extraction on the defect area to obtain brightness change features and area change features, and the brightness change features include brightness change rate, curvature and fluctuation amplitude; performing static feature extraction on the defect area to obtain grayscale features, texture features and shape features, and the shape features include at least position, area and perimeter.

[0094] Among them, the target features include static features and dynamic features. The dynamic features specifically include: brightness change features and area change features, and the brightness change features include brightness change rate, curvature and fluctuation amplitude, while the static features include: grayscale features, texture features and shape features, and the shape features at least include position, area and circumference.

[0095] In this step, the image is read frame by frame from the luminous image sequence, and the brightness value of each pixel in each image is calculated. The brightness value can be a weighted average of a grayscale value or an RGB value, for example, the brightness value = 0.299R + 0.587G + 0.114B. Since the luminous image sequence is collected according to a preset sampling rate, that is, the entire sequence is a time series, therefore, the brightness value of each pixel at different time points is arranged in chronological order to form a brightness-time curve. Using the adaptive threshold method, the adaptive threshold is calculated according to the statistical characteristics of the brightness-time curve (such as mean, standard deviation), and the part of the brightness-time curve that exceeds the adaptive threshold is marked as a brightness abnormal area. In this process, the connected region analysis method is used to merge adjacent brightness abnormal areas into one to obtain an accurate brightness abnormal area. Exemplarily, the formula can be used: adaptive threshold = μ + kσ, where μ is the mean of the brightness value, σ is the standard deviation of the brightness value, and k is a constant (such as 1.5). Use the Canny edge detection algorithm or other edge detection algorithms to perform edge recognition on the brightness abnormal area, that is, extract its edge pixels to form the outline of the defect area and obtain the complete defect area.

[0096] After the defect area is obtained, the features can be extracted according to the above static features and dynamic features. For dynamic features, the brightness change features include brightness change rate, curvature and fluctuation amplitude. The brightness change rate can be obtained by taking the derivative of the brightness-time curve, the curvature can be obtained by taking the second-order derivative of the brightness-time curve, and the fluctuation amplitude can be obtained by calculating the difference between the maximum and minimum values ​​of the brightness-time curve. The area change feature specifically refers to the area change rate. The area-time curve can be constructed first, and the area-time curve can be numerically differentiated to obtain the area change rate. For static features, the grayscale features include average grayscale and standard deviation. The average grayscale can be calculated by calculating the average grayscale value of the defect area, and the standard deviation can be calculated by calculating the standard deviation of the grayscale value of the defect area. Texture features can include contrast, correlation, energy, entropy, etc., which can be extracted using the grayscale co-occurrence matrix. Shape features include position, area and perimeter. The position can select the center coordinates of the defect area, the area can be calculated by counting the pixels of the defect area and the unit area corresponding to each pixel, and the perimeter can be approximately determined by counting the number of pixels on the contour or calculated using the arc length formula.

[0097] 206. Input the static features and dynamic features into the trained defect recognition model to obtain defect information.

[0098] The defect information includes the defect location, defect quantity, defect area, defect type and defect severity corresponding to the defect area.

[0099] In this step, the defect recognition model includes a classification sub-model and an evaluation sub-model, wherein the classification sub-model is used to classify and count defects, and the evaluation sub-model is used to evaluate the severity of defects. Specifically, the EL image sequence of defective areas is collected in advance as a sample, the static features and dynamic features of the defective areas are extracted from the samples, and labels are generated for each defective area, including the defect location, defect number, defect area, defect type and defect severity. Select any one of the support vector machine (SVM), random forest (Random Forest), convolutional neural network (CNN), deep learning model (such as ResNet, Inception, etc.) as the classification sub-model for training, so that it can accurately identify the location, number, area and type of defects based on static features. In addition, select any one of the support vector regression (SVR), random forest regression (RandomForest Regression), deep learning model (LSTM, GRU, etc.) as the evaluation sub-model for training, so that it can accurately identify the severity of defects based on dynamic features.

[0100] Specific training process: Normalize the feature values ​​to ensure that the features are compared under the same dimension; use the training set to train the model and use the validation set to tune the hyperparameters; use the test set to evaluate the performance of the model and calculate indicators such as accuracy, recall, and F1 score.

[0101] It should be noted that the specific execution process of inputting static features and dynamic features into the trained defect recognition model to obtain the defect position, defect quantity, defect area, defect type and defect severity corresponding to the defect area is: inputting grayscale features, texture features and shape features into the classification sub-model to obtain the defect position, defect quantity, defect area and defect type corresponding to the defect area; inputting brightness change features and area change features into the evaluation sub-model to obtain the defect severity corresponding to the defect area.

[0102] Since the classification sub-model is applicable to static features, and the evaluation sub-model is applicable to dynamic features, the grayscale features, texture features, and shape features of the defect area are input into the classification sub-model to obtain the defect location, defect quantity, defect area, and defect type. The brightness change features and area change features of the defect area are input into the evaluation sub-model to obtain the defect severity corresponding to the defect area. By integrating the defect location, defect quantity, defect area, defect type, and defect severity, complete defect information can be obtained.

[0103] 207. Generate a quality inspection report corresponding to the photovoltaic module based on the defect information.

[0104] This step is combined with the description of step 104 in the above method, and the same content will not be repeated here. It should be noted that the specific execution process of generating the quality inspection report corresponding to the photovoltaic module based on the defect information is: generating the defect distribution map corresponding to the photovoltaic module based on the defect location and defect type corresponding to the defect area; generating defect statistical information according to the defect number, defect area, defect area and defect severity corresponding to the defect area; using the defect distribution map and defect statistical information as the quality inspection report.

[0105] In this step, use a drawing library (such as Matplotlib, OpenCV) to draw the defect location on the image of the photovoltaic module and mark the defect type. Specifically, different colors or marks can be used to distinguish different types of defects to obtain a defect distribution map. The number of defects, defect area, and defect severity are summarized to generate defect statistics. Specifically, the number of defects, total defect area, average defect area, and average defect severity can be calculated. Format the statistical information into an easy-to-read table or text form. Create a template for the quality inspection report in advance, including the title, defect distribution map, defect statistical information, etc. Fill the defect distribution map and defect statistical information into the report template to generate the quality inspection report corresponding to the photovoltaic module. By generating a defect distribution map and defect statistical information, the defect situation of the photovoltaic module can be comprehensively evaluated, thereby providing intuitive and detailed quality inspection results.

[0106] In order to further improve the richness of the quality inspection report and evaluate the defect distribution map and defect statistical information before using it as the quality inspection report, it also includes: simulating sunlight irradiation on the defective area in the photovoltaic module according to standard test conditions; collecting electrical performance parameters of the defective area during the sunlight simulation irradiation, and determining the loss power corresponding to the defective area based on the electrical performance parameters; determining the degree of influence of the defective area on the performance of the photovoltaic module based on the defect type, defect location, defect number, defect area, defect severity and loss power corresponding to the defective area; generating defect statistical information according to the defect number, defect area, defect area and defect severity corresponding to the defective area, including: generating defect statistical information according to the defect number, defect area, defect area, defect severity, electrical performance parameters and performance influence corresponding to the defective area.

[0107] In this step, a sunlight simulator that complies with international standards (such as IEC 61215) can be installed in the darkroom in advance, and the parameters of the sunlight simulator such as light intensity, spectral distribution and temperature can be set to meet the standard test conditions, i.e., AM1.5G spectrum, 1000W / m 2 Light intensity, 25°C ambient temperature. Sunlight irradiation is used to simulate the lighting environment under actual working conditions to evaluate the electrical performance of the defective area. That is, each defective area can be exposed to simulated sunlight under the same conditions, and the electrical performance parameters of each defective area during the simulated sunlight irradiation process are recorded, including short-circuit current (Isc), open-circuit voltage (Voc), maximum power point (Pmp), fill factor (FF) and other parameters. The power loss of each defective area is calculated based on the electrical performance parameters. Since the power loss is equal to the difference between the expected power under standard conditions and the actual measured power, the actual measured power can be obtained through the electrical performance parameters, and the power loss of the defective area can be determined by combining the difference calculation.

[0108] After obtaining the lost power, the defect type, defect location, defect number, defect area, defect severity and lost power can be comprehensively considered to evaluate the degree of impact of the defect area on the performance of the photovoltaic module. Specifically, weights are assigned to some or all of the defect type, defect location, defect number, defect area, defect severity and lost power. For example, the defect type weight is 0.3, the defect location weight is 0.2, the defect number weight is 0.1, the defect area weight is 0.2, the defect severity weight is 0.1, and the loss power weight is 0.1. The weighted summation method is used to calculate the comprehensive score of each defect area. The higher the score, the greater the performance impact. The electrical performance parameters and the degree of performance impact are added to the defect statistics in the quality inspection report, so that the generated quality inspection report is richer and more comprehensive, and a more accurate evaluation of the inspection results is achieved.

[0109] like Figure 3 As shown above, Figure 1-2 The embodiment of the method shown in the figure is implemented. The embodiment of the present disclosure provides a photovoltaic module quality detection device, which is used to capture the defect changes of photovoltaic modules under dynamic conditions and improve the accuracy of quality detection. The embodiment of the device corresponds to the aforementioned method embodiment. For ease of reading, this embodiment will no longer repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in this embodiment can correspond to all the contents of the aforementioned method embodiment. The device includes:

[0110] A parameter determination unit 31, used to determine a current control parameter according to a detection requirement for the photovoltaic component and a component characteristic of the photovoltaic component, wherein the current control parameter is used to characterize a parameter for controlling dynamic current injection into the photovoltaic component;

[0111] A detection control unit 32, used to perform dynamic current injection on the photovoltaic module according to the current control parameters obtained by the parameter determination unit 31, and to collect images of the photovoltaic module during the dynamic current injection process to obtain a light-emitting image sequence;

[0112] An identification and evaluation unit 33 is used to identify and evaluate defects of the photovoltaic module based on the luminous image sequence obtained by the detection control unit 32, and obtain defect information corresponding to the photovoltaic module;

[0113] The quality feedback unit 34 is used to generate a quality inspection report corresponding to the photovoltaic module according to the defect information obtained by the identification and evaluation unit 33.

[0114] In some embodiments, Figure 4 As shown, the parameter determination unit 31 includes:

[0115] The waveform frequency determination module 311 is used to determine the current waveform and the waveform frequency in a preset defect-waveform relationship table according to the detection requirements. The defect-waveform relationship table is determined based on the test of the degree of appearance of the defect under different current waveforms;

[0116] An amplitude range determination module 312, for determining a current amplitude range in a preset power-amplitude range relationship table according to the rated power of the photovoltaic module;

[0117] The parameter summarizing module 313 is used to use the current waveform and waveform frequency obtained by the waveform frequency determining module 311 and the current amplitude range obtained by the amplitude range determining module 312 as current control parameters.

[0118] In some embodiments, Figure 4 As shown, the detection control unit 32 includes:

[0119] The distortion calculation module 321 is used to calculate the waveform distortion corresponding to the current waveform in real time during the dynamic current injection process;

[0120] The adjustment processing module 322 is used to reduce the current amplitude range according to a preset gradient and record the distortion times corresponding to the current waveform if the waveform distortion obtained by the distortion calculation module 321 exceeds a preset distortion threshold;

[0121] The alarm feedback module 323 is used to trigger a detection stop alarm if the distortion times recorded by the adjustment processing module 322 reaches a preset times threshold.

[0122] In some embodiments, Figure 4 As shown, the identification and evaluation unit 33 includes:

[0123] The defect feature extraction module 331 is used to determine the defect area according to the brightness change corresponding to the luminous image sequence, and extract features from the luminous image sequence according to the defect area to obtain target features, which include static features and dynamic features;

[0124] The identification and evaluation module 332 is used to input the static features and dynamic features obtained by the defect feature extraction module 331 into the trained defect recognition model to obtain defect information, which includes the defect position, defect number, defect area, defect type and defect severity corresponding to the defect area.

[0125] In some embodiments, Figure 4 As shown, the defect feature extraction module 331 is specifically used for:

[0126] Calculating the brightness value of each pixel in each image included in the luminous image sequence, and forming a brightness-time curve according to the change of the brightness value in the luminous image sequence;

[0127] Based on the brightness-time curve, the luminous image sequence is segmented using the adaptive threshold method to obtain the brightness abnormality area;

[0128] Use edge detection algorithm to identify the edge of the abnormal brightness area and obtain the defect area;

[0129] Dynamic feature extraction is performed on the defect area to obtain brightness change features and area change features. The brightness change features include brightness change rate, curvature and fluctuation amplitude.

[0130] Static features are extracted from the defect area to obtain grayscale features, texture features and shape features, where the shape features at least include position, area and perimeter.

[0131] In some embodiments, Figure 4 As shown, the defect recognition model includes a classification sub-model and an evaluation sub-model; the recognition evaluation module 332 is specifically used to:

[0132] Grayscale features, texture features and shape features are input into the classification sub-model to obtain the defect position, defect quantity, defect area and defect type corresponding to the defect area. The classification sub-model is used to classify and count the defects;

[0133] The brightness change characteristics and area change characteristics are input into the evaluation sub-model to obtain the defect severity corresponding to the defect area. The evaluation sub-model is used to evaluate the severity of the defects.

[0134] In some embodiments, Figure 4 As shown, the quality feedback unit 34 includes:

[0135] A defect distribution determination module 341 is used to generate a defect distribution map corresponding to the photovoltaic module based on the defect position and defect type corresponding to the defect area;

[0136] A statistical information determination module 342 is used to generate defect statistical information according to the defect quantity, defect area, defect area and defect severity corresponding to the defect area;

[0137] The report determination module 343 is used to use the defect distribution map obtained by the defect distribution determination module 341 and the defect statistical information obtained by the statistical information determination module 342 as a quality inspection report.

[0138] In some embodiments, Figure 4 As shown, the device also includes:

[0139] A simulation test module 344 is used to simulate sunlight irradiation on defective areas in the photovoltaic module according to standard test conditions before the statistical information determination module 342;

[0140] The loss calculation module 345 is used to collect the electrical performance parameters of the defective area during the sunlight simulation irradiation process of the simulation test module 344, and determine the loss power corresponding to the defective area according to the electrical performance parameters;

[0141] The performance impact determination module 346 determines the degree of impact of the defect area on the performance of the photovoltaic module based on the defect type, defect location, defect number, defect area, defect severity corresponding to the defect area and the loss power obtained by the loss calculation module 345;

[0142] The statistical information determination module 342 is specifically used to:

[0143] Defect statistics are generated based on the defect quantity, defect area, defect severity, electrical performance parameters, and performance impact corresponding to the defect area.

[0144] The embodiment of the present disclosure also provides a storage medium, the storage medium is used to store a computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute the above Figure 1-2 Photovoltaic module quality inspection method in.

[0145] The present disclosure also provides a processor for running a program, wherein the program executes the above Figure 1-2 Photovoltaic module quality inspection method in.

[0146] So far, various embodiments of the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0147] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. In particular, the various technical features mentioned in the various embodiments may be combined in any manner as long as there is no structural conflict.

Claims

1. A photovoltaic module quality detection method, characterized in that: The method comprises: Determining a current control parameter according to a detection requirement for a photovoltaic component and a component characteristic of the photovoltaic component, wherein the current control parameter is used to characterize a parameter for controlling dynamic current injection into the photovoltaic component; Performing dynamic current injection on the photovoltaic assembly according to the current control parameter, and performing image acquisition on the photovoltaic assembly during the dynamic current injection process to obtain a luminous image sequence; Based on the luminous image sequence, the photovoltaic module is identified and evaluated for defects to obtain defect information corresponding to the photovoltaic module; A quality inspection report corresponding to the photovoltaic component is generated according to the defect information.

2. The photovoltaic module quality detection method according to claim 1, characterized in that: Determining current control parameters according to detection requirements for photovoltaic components and component characteristics of the photovoltaic components includes: Determine the current waveform and waveform frequency in a preset defect-waveform relationship table according to the detection requirements, wherein the defect-waveform relationship table is determined based on a test of the degree of appearance of the defect under different current waveforms; Determining a current amplitude range in a preset power-amplitude range relationship table according to the rated power of the photovoltaic module; The current waveform, the waveform frequency and the current amplitude range are used as the current control parameters.

3. The photovoltaic module quality detection method according to claim 2, characterized in that: Dynamically injecting current into the photovoltaic component according to the current control parameter, and collecting images of the photovoltaic component during the dynamic current injection process to obtain a light-emitting image sequence, including: Calculating in real time during the dynamic current injection process the waveform distortion corresponding to the current waveform; If the waveform distortion exceeds a preset distortion threshold, the current amplitude range is reduced according to a preset gradient, and the number of distortions corresponding to the current waveform is recorded; If the distortion times reach a preset times threshold, a detection stop alarm is triggered.

4. The photovoltaic module quality detection method according to claim 1, characterized in that: Defect identification and evaluation of the photovoltaic module based on the light-emitting image sequence to obtain defect information corresponding to the photovoltaic module includes: Determine the defect area according to the brightness change corresponding to the luminous image sequence, and extract features in the luminous image sequence according to the defect area to obtain target features, wherein the target features include static features and dynamic features; The static features and the dynamic features are input into a trained defect recognition model to obtain the defect information, wherein the defect information includes the defect position, defect quantity, defect area, defect type and defect severity corresponding to the defect area.

5. The photovoltaic module quality detection method according to claim 4, characterized in that: Determining a defective area according to a brightness change corresponding to the luminous image sequence, and extracting features from the luminous image sequence according to the defective area to obtain target features, including: Calculating the brightness value of each pixel in each image included in the luminous image sequence, and forming a brightness-time curve according to the change of the brightness value in the luminous image sequence; Based on the brightness-time curve, segmenting the luminous image sequence using an adaptive threshold method to obtain brightness abnormality areas; Using an edge detection algorithm to perform edge recognition on the abnormal brightness area to obtain the defect area; Extracting dynamic features of the defect area to obtain brightness change features and area change features, wherein the brightness change features include brightness change rate, curvature and fluctuation amplitude; Static feature extraction is performed on the defect area to obtain grayscale features, texture features and shape features, and the shape features at least include position, area and perimeter.

6. The photovoltaic module quality detection method according to claim 4, characterized in that: The defect recognition model includes a classification sub-model and an evaluation sub-model; the static features and the dynamic features are input into the trained defect recognition model to obtain the defect information, including: Inputting the grayscale feature, the texture feature and the shape feature into the classification sub-model to obtain the defect position, the defect quantity, the defect area and the defect type corresponding to the defect area, the classification sub-model is used to classify and count the defects; The brightness change feature and the area change feature are input into the evaluation sub-model to obtain the defect severity corresponding to the defect area. The evaluation sub-model is used to evaluate the severity of the defect.

7. The photovoltaic module quality detection method according to any one of claims 4 to 6, characterized in that: Generating a quality inspection report corresponding to the photovoltaic module according to the defect information, including: generating a defect distribution map corresponding to the photovoltaic module based on the defect position and the defect type corresponding to the defect area; Generate the defect statistical information according to the defect quantity, the defect area, the defect size and the defect severity corresponding to the defect area; The defect distribution map and the defect statistical information are used as the quality inspection report.

8. A photovoltaic module quality detection device, characterized in that: The device comprises: A parameter determination unit, used to determine a current control parameter according to a detection requirement for a photovoltaic component and a component characteristic of the photovoltaic component, wherein the current control parameter is used to characterize a parameter for controlling dynamic current injection into the photovoltaic component; A detection control unit, configured to perform dynamic current injection on the photovoltaic assembly according to the current control parameter obtained by the parameter determination unit, and to perform image acquisition on the photovoltaic assembly during the dynamic current injection process to obtain a luminous image sequence; an identification and evaluation unit, configured to identify and evaluate defects of the photovoltaic component based on the light-emitting image sequence obtained by the detection control unit, and obtain defect information corresponding to the photovoltaic component; A quality feedback unit is used to generate a quality inspection report corresponding to the photovoltaic component according to the defect information obtained by the identification and evaluation unit.

9. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is executed, the device where the storage medium is located is controlled to execute the photovoltaic module quality detection method according to any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the photovoltaic module quality detection method according to any one of claims 1 to claim 7.

Images