A method for detecting the influence of hot spot effect on photovoltaic modules

Abstract

The present invention discloses a method for detecting the influence of hot spot effect on photovoltaic modules, and specifically relates to the technical field of photovoltaic module detection. By acquiring the internal material performance degradation information of the photovoltaic module and calculating the degradation coefficient of the photovoltaic module, the aging degree of the photovoltaic module is evaluated, and the potential aging problem of the photovoltaic module is more accurately identified. The infrared image of the temperature distribution on the surface of the photovoltaic module is acquired, the temperature fluctuation on the surface of the photovoltaic module is analyzed, the temperature fluctuation difference coefficient is calculated, and the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature is evaluated. Slight temperature fluctuations and normal temperature changes during operation are more accurately distinguished, the sensitivity of detection is improved, a fault hot spot fraud model is constructed, the difference between hot spots and fault hot spots is quantified, the recognition ability of pseudo hot spots is significantly improved, and fault warnings are issued for hot spots generated on the surface of photovoltaic modules.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic module detection, and more specifically, to a method for detecting the influence of hot spot effect on photovoltaic modules. Background Art

[0002] At present, photovoltaic power generation technology has been widely used and popularized around the world, and its future development prospects are extremely optimistic. This technology is flexible and diverse, and can adapt to the needs of various scenarios, such as rooftop photovoltaic power generation systems, large photovoltaic power stations, and portable photovoltaic equipment. However, in the production and manufacturing process of solar photovoltaic modules, although strict quality control and testing measures are implemented in all links, due to factors such as equipment, raw materials and production processes, there are still different degrees of quality differences between the produced solar photovoltaic modules. These quality differences will have different degrees of impact on the long-term stability and performance of photovoltaic modules, which will lead to aging problems of photovoltaic modules during actual use. The material performance of the aged photovoltaic modules will deteriorate, such as loose solder joints, oxidation of conductive materials, aging of packaging materials, etc. These degradations will affect the current transmission and heat dissipation performance of the components, resulting in non-uniform fluctuations in surface temperature. Existing detection methods may find it difficult to distinguish such small temperature fluctuations from normal operating temperature changes and may not be able to effectively capture these slight hot spot characteristics, especially in the early aging stages of components, when temperature fluctuations are small and fault manifestations are not obvious. This situation is often mistaken by the detection system as normal temperature fluctuations, which fails to attract sufficient attention and delays the early warning of faults. In addition, aging photovoltaic components may experience periodic hot spot phenomena, that is, temperature fluctuations will recur, which may make it difficult for existing detection methods to determine whether these hot spots are occasional temperature fluctuations or warning signs of long-term faults. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for detecting the influence of hot spot effect on photovoltaic modules to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for detecting the influence of hot spot effect on photovoltaic modules comprises the following steps:

[0006] Step S1, obtaining the internal material performance degradation information of the photovoltaic module, and calculating the photovoltaic module degradation coefficient according to the internal material performance degradation information to evaluate the aging degree of the photovoltaic module;

[0007] Step S2, obtaining an infrared image of the temperature distribution on the surface of the photovoltaic module, analyzing the temperature fluctuation on the surface of the photovoltaic module, calculating the temperature fluctuation difference coefficient, and evaluating the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature;

[0008] Step S3, constructing a fault hot spot anti-fraud model, generating a fault hot spot fraud index, and evaluating the difference between the hot spots generated on the surface of the photovoltaic module and the fault hot spots;

[0009] Step S4, comparing the fault hot spot fraud index with a preset fault hot spot fraud index threshold, and issuing a fault warning for the hot spot generated on the surface of the photovoltaic module.

[0010] In a preferred embodiment, the aging degree of the photovoltaic module is evaluated by obtaining the internal material performance degradation information of the photovoltaic module, analyzing the aging condition of the photovoltaic module, and calculating the degradation coefficient of the photovoltaic module. The logic for obtaining the degradation coefficient of the photovoltaic module is as follows:

[0011] Obtaining the internal material performance degradation information of the photovoltaic module, wherein the internal material performance degradation information includes the solder joint resistance , Conductive material conductivity , Transmittance of packaging materials ;

[0012] Perform data preprocessing on the internal material performance degradation information, including data clarification and standardization;

[0013] According to the fuzzy clustering algorithm, the internal material performance degradation information is subjected to fuzzy clustering analysis to calculate the degradation coefficient of the photovoltaic module, as follows:

[0014] Initialize the membership matrix ,in Indicates the membership degree of the data vector of material performance degradation information within the jth group to the i-th cluster;

[0015] The fuzzy clustering iteration includes the following steps:

[0016] Step A1, calculate cluster centers , according to the membership matrix , calculate the cluster center of each cluster i , the expression is as follows ,in represents the data vector of the internal material performance degradation information of the jth group, , j={1,2,3,...,n}, m represents the fuzzy factor;

[0017] Step A2, update membership , according to the calculated cluster center , update each group of internal material performance degradation information data vector Membership , the expression is as follows ,in Represents the data vector and cluster center of the material performance degradation information within the jth group The Euclidean distance of , k is the index number used to traverse all clusters, and c represents the total number of clusters;

[0018] Step A3, calculate the objective function value , the expression is as follows If the objective function value is greater than or equal to the set iteration threshold, repeat steps A1, A2, and A3. Otherwise, stop the fuzzy clustering iteration process and calculate the degradation coefficient of the photovoltaic module. , the expression is as follows .

[0019] In a preferred embodiment, the photovoltaic module degradation coefficient is compared with a photovoltaic module degradation coefficient reference threshold. If the photovoltaic module degradation coefficient is greater than the photovoltaic module degradation coefficient reference threshold, a photovoltaic module degradation abnormal signal is generated; if the photovoltaic module degradation coefficient is less than or equal to the photovoltaic module degradation coefficient reference threshold, no photovoltaic module degradation abnormal signal is generated.

[0020] In a preferred embodiment, the logic for obtaining the temperature fluctuation difference coefficient is as follows:

[0021] Use an infrared thermal imager to obtain the temperature distribution infrared image of the photovoltaic module surface, extract the temperature data of each pixel from the temperature distribution infrared image, and construct a temperature matrix ,in For one The temperature matrix, is the number of pixel rows of the temperature distribution infrared image, is the number of pixel columns of the temperature distribution infrared image, the temperature matrix Each element in Indicated in coordinates The temperature value at; where x={1,2,...,X}, y={1,2,...,Y};

[0022] Temperature Matrix Perform preprocessing, including denoising, smoothing, and standardization;

[0023] The temperature distribution infrared image of the surface of the photovoltaic module is obtained during the period when no abnormal signal of photovoltaic module degradation is generated, and it is used as the reference temperature distribution infrared image. The temperature data of each pixel point is extracted from the reference temperature distribution infrared image to form a reference temperature matrix. , reference temperature matrix Each element in Indicated in coordinates The reference temperature value at

[0024] Temperature Matrix Perform singular value decomposition, the decomposition result is ,in express The left singular vector matrix of is used to measure the temperature matrix The row characteristics of express The diagonal matrix of , where the diagonal elements are singular values, is arranged in descending order and is used to represent the weight of each temperature feature. express The right singular vector matrix of is used to measure the temperature matrix Column features of

[0025] For the reference temperature matrix Perform singular value decomposition, the decomposition result is ,in express The left singular vector matrix of ; express The diagonal matrix of express The right singular vector matrix of ;

[0026] Compare the singular values ​​on the diagonal of the diagonal matrix with a preset singular value threshold, and select h singular values ​​whose singular values ​​are less than or equal to the singular value threshold;

[0027] Reconstruct the temperature matrix based on the selected h singular values Get the low-rank temperature similarity matrix ,Right now ,in represents the left singular vector matrix The submatrix of contains the first h left singular vectors, represents the diagonal matrix of the first h singular values ​​selected, represents the right singular vector matrix The submatrix of contains the first h right singular vectors;

[0028] Reconstruct the reference temperature matrix based on the selected h singular values Get the low-rank benchmark temperature similarity matrix ,Right now ,in represents the left singular vector matrix The submatrix of contains the first h left singular vectors, represents the diagonal matrix of the first h singular values ​​selected, represents the right singular vector matrix The submatrix of contains the first h right singular vectors;

[0029] Calculate the temperature fluctuation difference matrix , the expression is as follows ;

[0030] Calculate the coefficient of variation of temperature fluctuations , the expression is as follows ,in Represents the temperature fluctuation difference matrix The total number of elements in , Represents the temperature fluctuation difference matrix Middle The values ​​of the points are f={1,2,...,F} and g={1,2,...,G}.

[0031] In a preferred embodiment, a fault hot spot anti-fraud model is constructed based on the degradation coefficient of the photovoltaic module and the temperature fluctuation difference coefficient to generate a fault hot spot fraud index. , based on the following formula , where Respectively represent the preset proportional coefficients of the photovoltaic module degradation coefficient and the temperature fluctuation difference coefficient, and Both are greater than 0.

[0032] In a preferred embodiment, the fault hot spot fraud index is compared with a preset fault hot spot fraud index threshold, and a fault warning is performed on the hot spot generated on the surface of the photovoltaic module, as follows:

[0033] If the fault hot spot fraud index is greater than the fault hot spot fraud index threshold, an early warning signal is generated to issue an early warning;

[0034] If the fault hot spot fraud index is less than or equal to the fault hot spot fraud index threshold, then the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is calculated, and the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is added to the ratio time series.

[0035] In a preferred embodiment, the comparison value time series is analyzed to identify the hot spots generated on the surface of the photovoltaic module as occasional temperature fluctuations or long-term fault hot spots, as follows:

[0036] Get ratio time series ,in represents the ratio of the fault hot spot fraud index calculated at the tth time point to the fault hot spot fraud index threshold, t={1,2,...,T};

[0037] The hidden Markov model is used to analyze the time series of comparison values, including the following steps:

[0038] Step B1, define and initialize the core elements of the hidden Markov model, including the state set, state probability distribution, state transition probability matrix, and observation probability matrix;

[0039] Step B2, using a forward-backward algorithm to calculate the state sequence corresponding to the ratio time series;

[0040] Step B3, outputting the state probability that the hot spot generated on the surface of the photovoltaic module is an occasional temperature fluctuation or a long-term fault hot spot.

[0041] Technical effects and advantages of the present invention:

[0042] 1. The present invention obtains the internal material performance degradation information of the photovoltaic module, calculates the photovoltaic module degradation coefficient based on the internal material performance degradation information, evaluates the aging degree of the photovoltaic module, more accurately identifies the potential aging problems of the photovoltaic module, helps to distinguish the temperature fluctuation caused by the aging of the photovoltaic module from the temperature change in normal operation, obtains the infrared image of the temperature distribution on the surface of the photovoltaic module, analyzes the temperature fluctuation on the surface of the photovoltaic module, calculates the temperature fluctuation difference coefficient, evaluates the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature, more accurately distinguishes slight temperature fluctuations from normal temperature changes in operation, avoids misjudgment, improves the sensitivity of detection, reduces missed detection, constructs a fault hot spot fraud model, generates a fault hot spot fraud index for quantifying the difference between hot spots and fault hot spots, significantly improves the recognition ability of pseudo hot spots, and compares the fault hot spot fraud index with a preset fault hot spot fraud index threshold to issue a fault warning for the hot spots generated on the surface of the photovoltaic module.

[0043] 2. According to the present invention, if the fault hot spot fraud index is greater than the fault hot spot fraud index threshold, an early warning signal is generated to issue an early warning, prompting operation and maintenance personnel to pay attention to potential faults of photovoltaic modules. Early warning can help intervene and handle the problem before it expands; if the fault hot spot fraud index is less than or equal to the fault hot spot fraud index threshold, the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is calculated, and the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is added to the ratio time series, and the ratio time series is analyzed to identify the hot spots generated on the surface of the photovoltaic module as occasional temperature fluctuations or long-term fault hot spots. Through accurate hot spot detection and early warning, it can help to promptly deal with potential fault problems of photovoltaic modules, reduce the energy efficiency loss of photovoltaic systems caused by aging and hot spot problems, effectively extend the service life of photovoltaic modules, and help improve the long-term operating efficiency and stability of solar photovoltaic systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0045] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Example: Figure 1 A method for detecting the influence of hot spot effect on photovoltaic modules is provided, comprising the following steps:

[0048] Step S1, obtaining internal material performance degradation information of the photovoltaic module, and calculating the photovoltaic module degradation coefficient according to the internal material performance degradation information to evaluate the aging degree of the photovoltaic module;

[0049] Step S2, obtaining an infrared image of the temperature distribution on the surface of the photovoltaic module, analyzing the temperature fluctuation on the surface of the photovoltaic module, calculating the temperature fluctuation difference coefficient, and evaluating the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature;

[0050] Step S3, constructing a fault hot spot anti-fraud model, generating a fault hot spot fraud index, and evaluating the difference between the hot spots generated on the surface of the photovoltaic module and the fault hot spots;

[0051] Step S4, comparing the fault hot spot fraud index with a preset fault hot spot fraud index threshold, and issuing a fault warning for the hot spot generated on the surface of the photovoltaic module;

[0052] Step S1, obtaining internal material performance degradation information of the photovoltaic module, and calculating the photovoltaic module degradation coefficient according to the internal material performance degradation information to evaluate the aging degree of the photovoltaic module;

[0053] With the long-term use of photovoltaic modules, the material properties inside the photovoltaic modules will deteriorate, such as loose solder joints, oxidation of conductive materials, aging of packaging materials, etc. These degradations will affect the current transmission and heat dissipation performance of photovoltaic modules, resulting in non-uniform fluctuations in the surface temperature of photovoltaic modules. These non-uniform fluctuations may cause the detection system to mistakenly regard them as normal temperature fluctuations, resulting in misjudgment. Therefore, by obtaining the internal material performance degradation information of photovoltaic modules, analyzing the aging of photovoltaic modules, and calculating the degradation coefficient of photovoltaic modules, the aging degree of photovoltaic modules is evaluated. The logic for obtaining the degradation coefficient of photovoltaic modules is as follows:

[0054] Obtaining the internal material performance degradation information of the photovoltaic module, wherein the internal material performance degradation information includes the solder joint resistance , Conductive material conductivity , Transmittance of packaging materials ;

[0055] The resistance of the solder joint reflects the quality of the solder connection. A higher resistance may mean that the solder joint is deteriorating or loose. The conductivity of the conductive material indicates the material's ability to conduct current. The lower the conductivity, the worse the performance of the conductive material. The light transmittance of the packaging material reflects the ability of the photovoltaic module to receive sunlight. A decrease in light transmittance means that the material is aging and the ability of light to pass through is weakened.

[0056] Perform data preprocessing on the internal material performance degradation information, including data clarification and standardization;

[0057] It should be noted that data clarity includes removing missing values ​​and outliers, and standardization includes normalizing the data. Commonly used normalization methods include Min-Max normalization and Z-Score normalization.

[0058] According to the fuzzy clustering algorithm, the internal material performance degradation information is subjected to fuzzy clustering analysis to calculate the degradation coefficient of the photovoltaic module, as follows:

[0059] Initialize the membership matrix ,in Indicates the membership degree of the data vector of material performance degradation information within the jth group to the i-th cluster;

[0060] It should be noted that when initializing the membership matrix When ,and , where c represents the preset total number of clusters, which is set according to different degradation states of the PV modules, for example, the degradation states of the PV modules are classified into slight, moderate, severe, etc.;

[0061] The fuzzy clustering iteration includes the following steps:

[0062] Step A1, calculate cluster centers , according to the membership matrix , calculate the cluster center of each cluster i , the expression is as follows ,in represents the data vector of the internal material performance degradation information of the jth group, , j={1,2,3,...,n}, m represents the fuzzy factor;

[0063] It should be noted that the fuzzy factor is usually a preset real number that satisfies m>1. The specific value can be adjusted according to the characteristics of the data set and the clustering requirements. The default value is usually set to m=2. The fuzzy factor m determines the degree of influence of the distance from the data vector to the cluster center on the membership degree.

[0064] Step A2, update membership , according to the calculated cluster center , update each group of internal material performance degradation information data vector Membership , the expression is as follows ,in Represents the data vector and cluster center of the material performance degradation information within the jth group The Euclidean distance of , k is the index number used to traverse all clusters, and c represents the total number of clusters;

[0065] It should be noted that Euclidean distance is a commonly used distance measurement method, which is used to measure the similarity between the internal material performance degradation information data vector and the cluster center, and will not be described in detail here;

[0066] Step A3, calculate the objective function value , the expression is as follows If the objective function value is greater than or equal to the set iteration threshold, repeat steps A1, A2, and A3. Otherwise, stop the fuzzy clustering iteration process and calculate the degradation coefficient of the photovoltaic module. , the expression is as follows

[0067] The degradation coefficient of photovoltaic modules is used to measure the degree of degradation or aging of photovoltaic modules. Since the temperature fluctuations on the surface of photovoltaic modules during the aging process are usually slight and irregular, they are easily mistaken for temperature changes caused by environmental factors or normal operation, resulting in the inability of existing detection methods to effectively capture real fault signals, thus causing misjudgment. The larger the degradation coefficient of photovoltaic modules, the greater the probability that the existing detection methods may misjudge it as normal operating temperature, that is, the greater the degree to which the hot spots on the surface of photovoltaic modules are true fault hot spots;

[0068] The degradation coefficient of the photovoltaic module is compared with the reference threshold of the degradation coefficient of the photovoltaic module. If the degradation coefficient of the photovoltaic module is greater than the reference threshold of the degradation coefficient of the photovoltaic module, a photovoltaic module degradation abnormal signal is generated; if the degradation coefficient of the photovoltaic module is less than or equal to the reference threshold of the degradation coefficient of the photovoltaic module, no photovoltaic module degradation abnormal signal is generated;

[0069] Step S2, obtaining an infrared image of the temperature distribution on the surface of the photovoltaic module, analyzing the temperature fluctuation on the surface of the photovoltaic module, calculating the temperature fluctuation difference coefficient, and evaluating the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature;

[0070] The temperature fluctuation difference coefficient is used to measure the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature. When the temperature fluctuation difference coefficient is smaller, it means that the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature is smaller, and the existing detection method may misjudge it as the normal operating temperature. The probability is greater, that is, the hot spot on the surface of the photovoltaic module is more likely to be a fault hot spot;

[0071] The logic for obtaining the temperature fluctuation difference coefficient is as follows:

[0072] Use an infrared thermal imager to obtain the temperature distribution infrared image of the photovoltaic module surface, extract the temperature data of each pixel from the temperature distribution infrared image, and construct a temperature matrix ,in For one The temperature matrix, is the number of pixel rows of the temperature distribution infrared image, is the number of pixel columns of the temperature distribution infrared image, the temperature matrix Each element in Indicated in coordinates The temperature value at; where x={1,2,...,X}, y={1,2,...,Y};

[0073] The temperature matrix It can be expressed as ;

[0074] Temperature Matrix Perform preprocessing, including denoising, smoothing, and standardization;

[0075] The temperature distribution infrared image of the surface of the photovoltaic module is obtained during the period when no abnormal signal of photovoltaic module degradation is generated, and it is used as the reference temperature distribution infrared image. The temperature data of each pixel point is extracted from the reference temperature distribution infrared image to form a reference temperature matrix. , reference temperature matrix Each element in Indicated in coordinates The reference temperature value at

[0076] The reference temperature matrix It can be expressed as ;

[0077] Temperature Matrix Perform singular value decomposition, the decomposition result is ,in express The left singular vector matrix of is used to measure the temperature matrix The row characteristics of express The diagonal matrix of , where the diagonal elements are singular values, is arranged in descending order and is used to represent the weight of each temperature feature. express The right singular vector matrix of is used to measure the temperature matrix Column features of

[0078] It should be noted that for the temperature matrix Singular value decomposition can be performed using a linear algebra library such as NumPy, which is an existing mature technology and will not be described in detail here;

[0079] For the reference temperature matrix Perform singular value decomposition, the decomposition result is ,in express The left singular vector matrix of ; express The diagonal matrix of express The right singular vector matrix of ;

[0080] Compare the singular values ​​on the diagonal of the diagonal matrix with a preset singular value threshold, and select h singular values ​​whose singular values ​​are less than or equal to the singular value threshold;

[0081] Reconstruct the temperature matrix based on the selected h singular values Get the low-rank temperature similarity matrix ,Right now ,in represents the left singular vector matrix The submatrix of contains the first h left singular vectors, represents the diagonal matrix of the first h singular values ​​selected, represents the right singular vector matrix The submatrix of contains the first h right singular vectors;

[0082] Reconstruct the reference temperature matrix based on the selected h singular values Get the low-rank benchmark temperature similarity matrix ,Right now ,in represents the left singular vector matrix The submatrix of contains the first h left singular vectors, represents the diagonal matrix of the first h singular values ​​selected, represents the right singular vector matrix The submatrix of contains the first h right singular vectors;

[0083] Calculate the temperature fluctuation difference matrix , the expression is as follows ;

[0084] Calculate the coefficient of variation of temperature fluctuations , the expression is as follows ,in Represents the temperature fluctuation difference matrix The total number of elements in , Represents the temperature fluctuation difference matrix Middle The value of the point, f={1,2,...,F}, g={1,2,...,G};

[0085] Step S3, constructing a fault hot spot anti-fraud model, generating a fault hot spot fraud index, and evaluating the difference between the hot spots generated on the surface of the photovoltaic module and the fault hot spots;

[0086] Constructing the anti-fraud index of hot spot failure based on the degradation coefficient of photovoltaic modules and the temperature fluctuation difference coefficient , based on the following formula , where Respectively represent the preset proportional coefficients of the photovoltaic module degradation coefficient and the temperature fluctuation difference coefficient, and All are greater than 0;

[0087] It should be noted that before building the hot spot anti-fraud model, the degradation coefficient of the photovoltaic module and the temperature fluctuation difference coefficient need to be normalized. Commonly used normalization methods include Min-Max normalization and Z-Score normalization. Set it according to the actual situation. For example, adopt the expert empowerment method, that is, invite experts in relevant fields to determine the preset proportion coefficients of various indicators through professional opinion surveys and comprehensive evaluations;

[0088] It can be seen from the above calculation expression that the larger the degradation coefficient of the photovoltaic module and the smaller the temperature fluctuation difference coefficient, the larger the fault hot spot fraud index is, indicating that the smaller the difference between the hot spot generated on the surface of the photovoltaic module and the fault hot spot is, the greater the probability of the fault hot spot is; conversely, the smaller the degradation coefficient of the photovoltaic module and the larger the temperature fluctuation difference coefficient is, the smaller the fault hot spot fraud index is, indicating that the greater the difference between the hot spot generated on the surface of the photovoltaic module and the fault hot spot is, the smaller the probability of the fault hot spot is;

[0089] Step S4, comparing the fault hot spot fraud index with a preset fault hot spot fraud index threshold, and providing a fault warning for the hot spot generated on the surface of the photovoltaic module, as follows:

[0090] If the fault hot spot fraud index is greater than the fault hot spot fraud index threshold, it means that the smaller the difference between the hot spot generated on the surface of the photovoltaic module and the fault hot spot, the greater the probability of the fault hot spot, and an early warning signal is generated to alert the operation and maintenance personnel to the potential failure of the photovoltaic module. Early warning can help intervene and deal with the problem before it expands;

[0091] If the fault hot spot fraud index is less than or equal to the fault hot spot fraud index threshold, the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is calculated, and the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is added to the ratio time series. The ratio time series is analyzed to identify the hot spots generated on the surface of the photovoltaic module as occasional temperature fluctuations or long-term fault hot spots, as follows:

[0092] Get ratio time series ,in represents the ratio of the fault hot spot fraud index calculated at the tth time point to the fault hot spot fraud index threshold, t={1,2,...,T};

[0093] The hidden Markov model is used to analyze the time series of comparison values, including the following steps:

[0094] Step B1, define and initialize the core elements of the hidden Markov model, including the state set , represents the implicit state of hot spots generated on the surface of photovoltaic modules, Indicates occasional temperature fluctuations. Indicates the state of long-term fault hot spot; initialization state probability distribution , represents the probability of occasional temperature fluctuations in the initial stage, represents the probability of a long-term hot spot failure state in the initial stage, which can be set by technicians in this field based on historical data or prior information of photovoltaic modules; initialization state transition probability matrix , indicating that from the implicit state Transfer to hidden state The probability of ,and ; Initialize the observation probability matrix , indicating that in the implicit state The observed ratio The probability of ;

[0095] Step B2, using a forward-backward algorithm to calculate the state sequence corresponding to the ratio time series;

[0096] Forward calculation: Initialize forward variables , the forward variable is used to represent the state at time point t And the observed ratio time series The probability of; forward recursive calculation based on the initialized forward vector ;

[0097] Backward calculation: Initialize backward variables , the backward variable is used to represent the state at time point t And observe the ratio time series from t+1 to T The probability of; perform backward recursive calculation based on the initialized backward variables ;

[0098] Step B3, outputting the state probability that the hot spot generated on the surface of the photovoltaic module is an occasional temperature fluctuation or a long-term fault hot spot;

[0099] The output expression is as follows , where represents the forward variable, represents the backward variable;

[0100] The present invention obtains the internal material performance degradation information of the photovoltaic module, calculates the photovoltaic module degradation coefficient based on the internal material performance degradation information, evaluates the aging degree of the photovoltaic module, more accurately identifies the potential aging problems of the photovoltaic module, helps to distinguish the temperature fluctuation caused by the aging of the photovoltaic module from the temperature change in normal operation, obtains the infrared image of the temperature distribution on the surface of the photovoltaic module, analyzes the temperature fluctuation on the surface of the photovoltaic module, calculates the temperature fluctuation difference coefficient, evaluates the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature, more accurately distinguishes between slight temperature fluctuations and normal temperature changes in operation, avoids misjudgment, improves the sensitivity of detection, reduces missed detection, constructs a fault hot spot fraud model, generates a fault hot spot fraud index for quantifying the difference between the hot spot and the fault hot spot, significantly improves the recognition ability of pseudo hot spots, and compares the fault hot spot fraud index with a preset fault hot spot fraud index threshold to issue a fault warning for the hot spot generated on the surface of the photovoltaic module.

[0101] According to the present invention, if the fault hot spot fraud index is greater than the fault hot spot fraud index threshold, an early warning signal is generated to issue an early warning, prompting operation and maintenance personnel to pay attention to potential faults of photovoltaic components. Early warning can help intervene and handle the problem before it expands; if the fault hot spot fraud index is less than or equal to the fault hot spot fraud index threshold, the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is calculated, and the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is added to the ratio time series, and the ratio time series is analyzed to identify the hot spots generated on the surface of the photovoltaic components as occasional temperature fluctuations or long-term fault hot spots. Through accurate hot spot detection and early warning, it can help to promptly deal with potential fault problems of photovoltaic components, reduce the energy efficiency loss of the photovoltaic system caused by aging and hot spot problems, effectively extend the service life of photovoltaic components, and help improve the long-term operating efficiency and stability of solar photovoltaic systems.

[0102] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0103] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0104] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for detecting the influence of hot spot effect on photovoltaic modules, characterized in that: The steps include: Step S1, obtaining internal material performance degradation information of the photovoltaic module, and calculating the photovoltaic module degradation coefficient according to the internal material performance degradation information to evaluate the aging degree of the photovoltaic module; Step S2, obtaining an infrared image of the temperature distribution on the surface of the photovoltaic module, analyzing the temperature fluctuation on the surface of the photovoltaic module, calculating the temperature fluctuation difference coefficient, and evaluating the difference between the temperature fluctuation on the surface of the photovoltaic module and the normal operating temperature; Step S3, constructing a fault hot spot anti-fraud model, generating a fault hot spot fraud index, and evaluating the difference between the hot spots generated on the surface of the photovoltaic module and the fault hot spots; Step S4, comparing the fault hot spot fraud index with a preset fault hot spot fraud index threshold, and issuing a fault warning for the hot spot generated on the surface of the photovoltaic module; The aging degree of photovoltaic modules is evaluated by obtaining the internal material performance degradation information of photovoltaic modules, analyzing the aging of photovoltaic modules, and calculating the degradation coefficient of photovoltaic modules. The logic for obtaining the degradation coefficient of photovoltaic modules is as follows: Obtaining internal material performance degradation information of the photovoltaic module, wherein the internal material performance degradation information includes solder joint resistance hdz, conductive material conductivity ddl, and packaging material transmittance fzl; Perform data preprocessing on the internal material performance degradation information, including data clarification and standardization; According to the fuzzy clustering algorithm, the internal material performance degradation information is subjected to fuzzy clustering analysis to calculate the degradation coefficient of the photovoltaic module, as follows: Initialize the membership matrix UL = [ul ij ], where ul ij Indicates the membership degree of the data vector of material performance degradation information within the jth group to the i-th cluster; The fuzzy clustering iteration includes the following steps: Step A1, calculate the cluster center vz i , according to the membership matrix UL, calculate the cluster center vz of each cluster i i , the expression is as follows Among them, sl j represents the data vector of the internal material performance degradation information of the jth group, j = {1, 2, 3, ..., n}, m represents the fuzzy factor; Step A2, update the membership degree ul ij , according to the calculated cluster center vz i , update each group of internal material performance degradation information data vector sl j The membership degree ul ij , the expression is as follows where ||sl j -vz i || represents the data vector of the material performance degradation information of the jth group and the cluster center vz i The Euclidean distance of , k is the index number used to traverse all clusters, and c represents the total number of clusters; Step A3, calculate the objective function value Juv, the expression is as follows If the objective function value is greater than or equal to the set iteration threshold, repeat steps A1, A2, and A3. Otherwise, stop the fuzzy clustering iteration process and calculate the degradation coefficient of the photovoltaic module Gflh. The expression is as follows: The degradation coefficient of the photovoltaic module is compared with the reference threshold of the degradation coefficient of the photovoltaic module. If the degradation coefficient of the photovoltaic module is greater than the reference threshold of the degradation coefficient of the photovoltaic module, a photovoltaic module degradation abnormal signal is generated; if the degradation coefficient of the photovoltaic module is less than or equal to the reference threshold of the degradation coefficient of the photovoltaic module, no photovoltaic module degradation abnormal signal is generated; The logic for obtaining the temperature fluctuation difference coefficient is as follows: Use an infrared thermal imager to obtain an infrared image of the temperature distribution on the surface of the photovoltaic module, extract the temperature data of each pixel point from the temperature distribution infrared image, and construct a temperature matrix AJ, where AJ is an X*Y temperature matrix, X is the number of pixel rows in the temperature distribution infrared image, Y is the number of pixel columns in the temperature distribution infrared image, and each element T in the temperature matrix AJ is x,y represents the temperature value at the coordinate (x, y); wherein x = {1, 2, ..., X}, y = {1, 2, ..., Y}; Preprocess the temperature matrix AJ, including denoising, smoothing and standardization; The temperature distribution infrared image of the surface of the photovoltaic module is obtained during the period when no abnormal signal of photovoltaic module degradation is generated, and it is used as the reference temperature distribution infrared image. The temperature data of each pixel point is extracted from the reference temperature distribution infrared image to form a reference temperature matrix AJ. base , reference temperature matrix AJ base Each element Tbase x,y Represents the reference temperature value at coordinate (x, y); Perform singular value decomposition on the temperature matrix AJ, and the decomposition result is AJ=UΣV T , where U represents the left singular vector matrix of X*X, which is used to measure the row characteristics of the temperature matrix AJ; Σ represents the diagonal matrix of X*Y, and the diagonal elements are singular values, arranged in descending order, which are used to represent the weight of each temperature feature, V T Represents the right singular vector matrix of Y*Y, which is used to measure the column characteristics of the temperature matrix AJ; For the reference temperature matrix AJ base Perform singular value decomposition, the decomposition result is AJ base =UbaseΣbaseVbase T , where Ubase represents the left singular vector matrix of X*X; Σbase represents the diagonal matrix of X*Y, Vbase T represents the right singular vector matrix of Y*Y; Compare the singular values ​​on the diagonal of the diagonal matrix with a preset singular value threshold, and select h singular values ​​whose singular values ​​are less than or equal to the singular value threshold; Reconstruct the temperature matrix AJ according to the selected h singular values ​​to obtain the low-rank temperature similarity matrix AJ h ,Right now Among them U h Represents the submatrix of the left singular vector matrix U, containing the first h left singular vectors, Σ h represents the diagonal matrix of the first h singular values ​​selected, represents the right singular vector matrix V T The submatrix of contains the first h right singular vectors; Reconstruct the reference temperature matrix AJ based on the selected h singular values base Get the low-rank benchmark temperature similarity matrix AJ baseh ,Right now Ubase h Represents the submatrix of the left singular vector matrix Ubase, containing the first h left singular vectors, Σbase h represents the diagonal matrix of the first h singular values ​​selected, Represents the right singular vector matrix Vbase T The submatrix of contains the first h right singular vectors; Calculate the temperature fluctuation difference matrix ΔAJ, the expression is as follows ΔAJ=AJ h -AJ baseh ; Calculate the temperature fluctuation difference coefficient Wdbd, the expression is as follows Where YS represents the total number of elements in the temperature fluctuation difference matrix ΔAJ, ΔAJ f,g represents the value of the (f,g)th point in the temperature fluctuation difference matrix ΔAJ, f={1,2,...,F}, g={1,2,...,G}; According to the degradation coefficient of photovoltaic modules and the temperature fluctuation difference coefficient, a fault hot spot anti-fraud model is constructed to generate the fault hot spot fraud index Gzqz. The formula is as follows: Wherein α and β represent the preset proportional coefficients of the photovoltaic module degradation coefficient and the temperature fluctuation difference coefficient, respectively, and α and β are both greater than 0.

2. A method for detecting the influence of hot spot effect on photovoltaic modules according to claim 1, characterized in that: The fault hot spot fraud index is compared with the preset fault hot spot fraud index threshold, and a fault warning is issued for the hot spot generated on the surface of the photovoltaic module, as follows: If the fault hot spot fraud index is greater than the fault hot spot fraud index threshold, an early warning signal is generated to issue an early warning; If the fault hot spot fraud index is less than or equal to the fault hot spot fraud index threshold, then the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is calculated, and the ratio of the fault hot spot fraud index to the fault hot spot fraud index threshold is added to the ratio time series.

3. A method for detecting the influence of hot spot effect on photovoltaic modules according to claim 2, characterized in that: By analyzing the time series of the contrast values, the hot spots generated on the surface of the photovoltaic modules are identified as occasional temperature fluctuations or long-term fault hot spots, as follows: Get the ratio time series BZ = {bz t }={bz1,bz2,...,bz T }, where bz t represents the ratio of the fault hot spot fraud index calculated at the t-th time point to the fault hot spot fraud index threshold, t = {1, 2, ..., T}; The hidden Markov model is used to analyze the time series of comparison values, including the following steps: Step B1, define and initialize the core elements of the hidden Markov model, including the state set, state probability distribution, state transition probability matrix, and observation probability matrix; Step B2, using a forward-backward algorithm to calculate the state sequence corresponding to the ratio time series; Step B3, outputting the state probability that the hot spot generated on the surface of the photovoltaic module is an occasional temperature fluctuation or a long-term fault hot spot.

Images