Photovoltaic power generation diagnosis method and system based on multi-dimensional influence quantity

Abstract

The invention discloses a photovoltaic power generation diagnosis method and system based on multi-dimensional influence quantity, and belongs to the technical field of photovoltaic power generation, and the method comprises the steps: 1, collecting photovoltaic power generation grid-connected point data and multi-dimensional influence quantity data at high frequency, and carrying out the preprocessing of the data; step 2, based on the preprocessed data, forming a mapping relation between a multi-dimensional influence quantity and photovoltaic output through clustering statistics, performing performance evaluation on a real-time state of the photovoltaic output based on the mapping relation, and diagnosing problems existing in a photovoltaic system; and step 3, displaying the final diagnosis result, and performing corresponding system optimization and maintenance by the user according to the diagnosis result. The photovoltaic power generation diagnosis system can comprehensively consider various influence quantities, provides an accurate diagnosis result, and helps a user to optimize and maintain the photovoltaic power generation system, thereby improving the overall efficiency and stability of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and in particular to a photovoltaic power generation diagnosis method and system based on multi-dimensional influence quantities. Background Art

[0002] Nowadays, with the increasing global energy crisis and the worsening global warming problem, the development and utilization of clean energy has received widespread attention. Photovoltaic power generation, as a clean, efficient and sustainable way to obtain energy, has developed rapidly in recent years. It converts sunlight energy directly into electrical energy through solar panels, and has the advantages of no pollution, no noise, and no emissions, making it very suitable for application in remote areas and urban buildings.

[0003] However, the efficiency of photovoltaic power generation systems is not static. It is affected by a variety of environmental factors and the system's own conditions. For example, changes in light intensity directly affect the power generation of solar panels; temperature increases will cause the efficiency of panels to decrease; and changes in humidity will also affect the stability and safety of the system. In addition, system component aging, surface dust accumulation, line loss and other issues will also have a negative impact on power generation efficiency.

[0004] When conducting traditional analysis on power generation efficiency, if traditional linear models are used to evaluate various influencing quantities and the interactions between influencing quantities are ignored, errors may occur. For example, the power generation P can be expressed as a linear combination of influencing quantities such as light intensity, temperature, and humidity, that is, P = (light intensity) * K1 + (temperature) * K2 + (humidity) * K3 + ... This method simplifies the complexity to a certain extent, but fails to fully consider the interactions and nonlinear relationships between influencing quantities. If the system performance is judged only based on the change in light intensity, the problem of decreased efficiency caused by increased temperature may be ignored.

[0005] Prior art CN117993741A discloses an abnormal diagnosis method, device, computer equipment and storage medium for photovoltaic power generation. The method includes: obtaining the operation data of photovoltaic stations, training a preset machine learning model based on the historical operation data of photovoltaic stations, and generating a photovoltaic power generation prediction model; using the photovoltaic power generation prediction model to obtain the power generation that should be generated on the day, and performing daily power generation fluctuation assessment and power generation efficiency assessment on multiple photovoltaic stations to determine photovoltaic stations with abnormal power generation; comparing the operation data corresponding to the photovoltaic stations with abnormal power generation with the diagnosis judgment conditions, and generating abnormal diagnosis results for photovoltaic power generation. The shortcomings of the prior art documents are: 1. The model operation of the prior art needs to obtain the historical data of the photovoltaic station to determine the model operation parameters; 2. The frequency of photovoltaic station operation data collection in the prior art cannot be improved; 3. The power generation prediction model mentioned in the prior art does not consider comprehensive factors. Summary of the invention

[0006] The purpose of the present invention is to provide a photovoltaic power generation diagnosis method based on multi-dimensional influencing quantities, which can comprehensively consider multiple influencing quantities such as light intensity, ambient temperature, ambient humidity, etc., and accurately diagnose the performance of a photovoltaic power generation system.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A first aspect of the present invention provides a photovoltaic power generation diagnosis method based on multi-dimensional influence quantities, comprising the following steps:

[0009] High-frequency collection of photovoltaic power generation grid-connected point data and its multi-dimensional impact data, and pre-processing of the data;

[0010] Based on the preprocessed data, a mapping relationship between multi-dimensional influencing variables and photovoltaic output is formed through clustering statistics. Based on this mapping relationship, the real-time status of photovoltaic output is evaluated to diagnose problems in the photovoltaic system.

[0011] The final diagnosis results are displayed, and users can perform corresponding system optimization and maintenance based on the diagnosis results.

[0012] Optionally, the photovoltaic power generation system grid connection point data and its multi-dimensional influencing quantity data include power generation parameters and environmental parameters of the photovoltaic power generation system, the environmental parameters include light intensity, temperature and humidity, and the power generation parameters include power generation, voltage and current.

[0013] Optionally, forming a mapping relationship between a multi-dimensional influence quantity and photovoltaic output by clustering statistics includes:

[0014] The mapping relationship between the multi-dimensional influencing quantity and the photovoltaic output is statistically analyzed by the DBSCAN clustering algorithm, and the mapping relationship is formed by clustering clusters automatically determined by the DBSCAN clustering algorithm.

[0015] Optionally, the preprocessing of data includes format conversion and data screening.

[0016] Optionally, the format conversion includes converting data formats of multiple influencing quantities into a unified data format.

[0017] Optionally, the data screening includes eliminating periodic fluctuations, excluding short-term random disturbances, measuring noise and decomposing low-frequency components in the extracted data.

[0018] Optionally, the decomposing and extracting low-frequency components in the data includes:

[0019] The Savitzky-Golay filter is based on local polynomial fitting. It uses the data in the sliding window to construct a polynomial function to fit the local data, and then uses the fitting result to replace the data point in the center of the window. The formula of the Savitzky-Golay filter is:

[0020]

[0021] in, is the smoothed signal, N is the sliding window size, which indicates the number of data points included in each fitting, and c k is the filter coefficient calculated by polynomial fitting, y(t) represents the original signal, and y(t+k) represents the signal at time t+k.

[0022] Optionally, the mapping relationship between the multi-dimensional influence quantity and the photovoltaic output is formed by clustering statistics, and the real-time status of the photovoltaic output is evaluated based on the mapping relationship. The diagnosis of the problems existing in the photovoltaic system includes:

[0023] Taking multidimensional influencing variables as independent variables and photovoltaic output as dependent variables, the mapping relationship between multidimensional influencing variables and photovoltaic output can be formed based on high-frequency data collection. Based on this mapping relationship, the theoretical output value of the photovoltaic system under different environmental conditions can be calculated.

[0024] Compare and analyze the calculated theoretical output value with the actual measured value to identify the deviation of system performance through comparison;

[0025] These deviations are further analyzed, and when the system deviation is greater than the set threshold, it is diagnosed that there may be problems with the photovoltaic system, and the abnormal results of the photovoltaic system are output.

[0026] Optionally, the causes of the deviation include equipment failure and excessive temperature.

[0027] The second aspect of the present invention provides a photovoltaic power generation diagnosis system based on multidimensional influence quantities, which is used to execute the photovoltaic power generation diagnosis method based on multidimensional influence quantities described in the first aspect of the present invention, and the system includes a data acquisition module, a data processing module, an influence factor analysis module, a diagnosis evaluation module and a user interaction interface;

[0028] The system includes a data acquisition module, a data processing module, a multi-dimensional influence quantity analysis module, a diagnosis and evaluation module and a user interaction interface;

[0029] The data acquisition module includes a system operation parameter acquisition unit and an environmental parameter acquisition unit, which are used to collect relevant data of the operation of the photovoltaic power generation system in real time, including the power generation parameters and environmental parameter data of the photovoltaic power generation system, and transmit the data to the data processing module; the data processing module pre-processes the received data of different types, and transmits the processed data to the multi-dimensional influence quantity analysis module; the multi-dimensional influence quantity analysis module is used to establish a mapping relationship diagram based on real-time data, which is used to reflect the relationship between the performance of the photovoltaic system and various influence quantities; the diagnosis and evaluation module evaluates the performance of the photovoltaic power generation system according to the multi-dimensional influence quantity mapping relationship diagram, and outputs the diagnosis results; the user interaction interface is used to display the diagnosis results and provide a user operation interface, so that the user can perform corresponding system optimization and maintenance according to the diagnosis results.

[0030] The beneficial effect of the present invention is that, compared with the prior art, the photovoltaic power generation diagnostic system of the present invention can comprehensively consider multiple influencing quantities, provide accurate diagnostic results, and help users optimize and maintain the photovoltaic power generation system, thereby improving the overall efficiency and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Attached Figure 1 It is a three-dimensional influence quantity mapping relationship generated by taking light intensity, temperature and humidity as influence quantities;

[0032] Attached Figure 2 It is a schematic diagram of the system structure of this method. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0034] When conducting traditional analysis on power generation efficiency, if traditional linear models are used to evaluate various influencing factors and the interactions between influencing factors are ignored, errors may occur. In order to more comprehensively understand the impact of these influencing factors on power generation efficiency, it is recommended to use a multidimensional analysis method. Taking light intensity, temperature and humidity as an example, through long-term data collection and analysis, a surface graph can be drawn in three-dimensional space, such as Figure 1As shown in the figure. This surface graph can show the theoretical power generation under different environmental conditions. By comparing the actual power with the theoretical power, we can quickly find some abnormal conditions in the power generation system, locate potential problems, and ensure the stable operation of the power generation system. In addition, we can observe the impact of light intensity on power generation under specific temperature and humidity conditions. This detailed analysis helps to optimize the operating parameters of power generation equipment and improve the overall power generation efficiency.

[0035] Therefore, in order to improve the overall efficiency and reliability of photovoltaic power generation systems, a photovoltaic power generation diagnostic method that can comprehensively consider multiple influencing quantities is needed. Such a system should be able to monitor and analyze multiple environmental parameters such as light intensity, temperature, humidity, etc. in real time, as well as evaluate the operating status and aging degree of the internal components of the system. By comprehensively analyzing these data, the diagnostic method can more accurately identify the reasons for the decline in performance of the photovoltaic power generation system, provide a scientific basis for the optimization and maintenance of the system, and thus improve the overall efficiency and stability of the photovoltaic power generation system.

[0036] In Example 1, the present invention provides a photovoltaic power generation diagnosis method based on multi-dimensional influence quantities, comprising the following steps:

[0037] Step 1, installing the electronic device of the present invention near the photovoltaic power generation system grid connection point, realizing high-frequency collection of photovoltaic power generation grid connection point data and multi-dimensional influencing quantity data, and pre-processing the data;

[0038] In a further preferred but non-restrictive embodiment, in step 1, high-frequency collection of photovoltaic power generation grid-connected point data and its multi-dimensional influence quantity data is achieved. By installing the device at the photovoltaic power generation system grid-connected point, millisecond-level high-speed sampling of grid-connected point data and multi-dimensional influence quantity data can be achieved, providing a high-precision and high-reliability data source for subsequent multi-dimensional influence quantity mapping relationships.

[0039] In a further preferred but non-limiting embodiment, in step 1, the photovoltaic power generation grid connection point data and its multi-dimensional influencing quantity data include environmental parameters and power generation parameters of the photovoltaic power generation system, wherein the environmental parameters include light intensity, temperature and humidity, etc., and the power generation parameters include power generation, voltage and current, active power, etc.

[0040] In a further preferred but non-limiting embodiment, in step 1, preprocessing the data includes data screening, format conversion, etc. to ensure the accuracy and consistency of the data; format conversion refers to converting the data formats of multiple influencing quantities into a unified data format; data screening includes eliminating periodic fluctuations, excluding short-term random disturbances, measuring noise, and decomposing and extracting low-frequency components in the data.

[0041] Specifically, the format conversion uses multiple vectors to achieve unified expression of multi-dimensional influence quantities. Multiple vectors can achieve unified expression of geometric objects of different dimensions and types. After expanding their characteristics, they can achieve integrated and formalized expression of complex objects and spatial relationships.

[0042] In a further preferred but non-limiting embodiment, the decomposing and extracting low-frequency components in the data specifically includes:

[0043] Compared with the high-frequency collection of grid-connected point data, the change of grid-connected point output characteristics is a relatively slow process. Therefore, the low-frequency components in the data can be extracted by adding a low-pass filter. This method uses the Savitzky-Golay filter, which is widely used in signal processing tasks because it can better preserve the local characteristics of the original signal while smoothing the signal.

[0044] The Savitzky-Golay filter (SG filter for short) is a smoothing method based on local polynomial fitting. Its core idea is to use the data in the sliding window to construct a polynomial function to fit the local data, and then use the fitting result to replace the data points in the center of the window to achieve the goal of smoothing the signal.

[0045] The mathematical formula of the SG filter can be expressed as:

[0046]

[0047] in, is the smoothed signal; N is the sliding window size, which indicates the number of data points included in each fitting, and needs to be an odd number to ensure window symmetry; c k It is the filter coefficient calculated by polynomial fitting, which weighs the contribution of each point in the window to the smooth value of the center point; y(t) represents the original signal, and y(t+k) represents the signal at time t+k.

[0048] Step 2: Based on the preprocessed data, a mapping relationship between multi-dimensional influencing variables and photovoltaic output is formed through clustering statistics. Based on this mapping relationship, the real-time status of photovoltaic output is evaluated to diagnose problems existing in the photovoltaic system.

[0049] In a further preferred but non-limiting embodiment, in step 2, the mapping relationship between the multidimensional influence quantity and the photovoltaic output is statistically calculated by the DBSCAN clustering algorithm. DBSCAN is a density-based clustering algorithm that divides areas with sufficiently high density into clusters and can find clusters of any shape in a noisy background. Abnormal points (noise points) can be found while clustering, and are insensitive to abnormal points in the data set. There is no need to preset the number of clusters, and the algorithm automatically determines the number of cluster clusters. For the cluster clusters automatically determined by the algorithm, that is, the mapping relationship between the multidimensional influence quantity and the photovoltaic output.

[0050] Specifically, taking light intensity as an example, when the influencing factors of other dimensions are constant, the light intensity within a certain range and the photovoltaic output value collected at high frequency are taken, and the photovoltaic output value can be statistically analyzed according to the DBSCAN clustering algorithm. Thus, a series of core objects are obtained, and the output value of this core object is the mapping relationship between light intensity and photovoltaic output under the current conditions. If the core object is not density-reachable, each core object can be made reachable by adjusting the neighborhood distance, thereby realizing the continuous mapping of the relationship between light intensity and photovoltaic output. Similarly, the same method can be used to obtain the continuous mapping relationship between other dimensional influencing factors and photovoltaic output.

[0051] In a further preferred but non-limiting embodiment, the step 2 specifically comprises:

[0052] Step 2.1, take the multidimensional influencing variable as the independent variable and the photovoltaic output as the dependent variable, and draw in the same space. Based on the high-frequency data collection in step 1, the mapping relationship between the multidimensional influencing variable and the photovoltaic output can be quickly formed. Here, taking the ambient temperature, humidity and light intensity as an example, after interpolation and smoothing, the following can be formed. Figure 1 The mapping relationship diagram is shown. Based on this mapping relationship, the theoretical output value of the photovoltaic system under different environmental conditions can be calculated. This theoretical output value is the result of comprehensive analysis of various influencing factors in reality, and can reflect the theoretical system performance under certain environmental conditions;

[0053] Step 2.2, compare and analyze the calculated theoretical output value with the actual measured value, through which the deviation of system performance can be identified;

[0054] Step 2.3, further analyzing these deviations can determine the possible causes of the deviations. Through this analysis, possible problems in the photovoltaic system can be diagnosed more accurately and the corresponding diagnostic results can be output.

[0055] Specifically, compared with the high-frequency collection of grid-connected point data, the change in the grid-connected point output characteristics is a relatively slow process. When the calculated theoretical output is compared with the actual measured value, and the deviation is more than 1 times the theoretical value, it can be considered that there is a problem with the system.

[0056] In a further preferred but non-limiting embodiment, in step 2.3, possible causes of the deviation include equipment failure, inverter efficiency reduction, excessive temperature, etc.

[0057] In a further preferred but non-limiting embodiment, in step 2.3, the output of the corresponding diagnostic results may include the current power generation efficiency of the system, potential causes of performance loss, and recommended optimization measures.

[0058] In a further preferred but non-limiting embodiment, in step 2, the multi-dimensional influence quantity mapping relationship specifically includes:

[0059] A multidimensional analysis method, such as three-dimensional vector analysis, is used to establish a mapping relationship that can comprehensively consider the effects of light intensity, ambient temperature and ambient humidity on photovoltaic power generation efficiency.

[0060] In a specific embodiment, light intensity, ambient temperature, and ambient humidity may be used as three dimensions of a three-dimensional vector, and the overall performance of the system may be evaluated by analyzing the impact of changes in these three dimensions on power generation efficiency.

[0061] In a further preferred but non-limiting embodiment, in step 2, the multi-dimensional influencing quantity mapping relationship can fully reflect the complex relationship between the photovoltaic system performance and various influencing quantities. This mapping relationship can be a mathematical mapping relationship based on physical principles, which describes the relationship between system performance and various influencing factors through theoretical derivation; or it can be a machine learning mapping relationship obtained through real-time data training, which learns the relationship between system performance and influencing factors in a data-driven manner.

[0062] Step 3: Display the final diagnosis results so that users can perform corresponding system optimization and maintenance based on the diagnosis results.

[0063] In a further preferred but non-limiting embodiment, in step 3, the user may choose to shut down or reduce the power generation to ensure the safety of the photovoltaic power generation system.

[0064] In Embodiment 2 of the present invention, the present invention provides a photovoltaic power generation diagnosis system based on multi-dimensional influence quantities, which is used to execute the photovoltaic power generation diagnosis method based on multi-dimensional influence quantities described in Embodiment 1. The system structure is as follows: Figure 2As shown, the system includes a data acquisition module, a data processing module, a multi-dimensional influence quantity analysis module, a diagnosis and evaluation module and a user interaction interface; the data acquisition module includes a system operation parameter acquisition unit and an environmental parameter acquisition unit, which collects power generation parameters and environmental parameter data of the photovoltaic power generation system in real time through sensors, the environmental parameters include light intensity, temperature and humidity, etc., and the power generation parameters include power generation, voltage and current, etc.;

[0065] The data acquisition module is connected to the data processing module through a circuit line, and the collected data is transmitted to the data processing module, which pre-processes the data, including data screening, format conversion, etc., to ensure the accuracy and consistency of the data;

[0066] The data processing module is connected to the multidimensional influence quantity analysis module through a circuit line, and the processed data is sent to the multidimensional influence quantity analysis module, which uses a multidimensional analysis method, such as a three-dimensional vector analysis, to establish a multidimensional influence quantity mapping relationship, which can comprehensively consider the influence of light intensity, ambient temperature and ambient humidity on photovoltaic power generation efficiency. For example, light intensity, ambient temperature and ambient humidity can be used as the three dimensions of the three-dimensional vector, and the overall performance of the system can be evaluated by analyzing the influence of changes in these three dimensions on power generation efficiency.

[0067] The multi-dimensional influence quantity analysis module is connected to the diagnosis and evaluation module through a circuit line. The diagnosis and evaluation module evaluates the performance of the photovoltaic power generation system according to the mapping relationship provided by the influence factor analysis module and outputs a diagnosis result. The diagnosis result may include the current power generation efficiency of the system, potential causes of performance loss, and recommended optimization measures.

[0068] The diagnostic evaluation module is connected to a user interaction interface via a circuit line or wirelessly, and the user interaction interface is used to display the diagnostic results and provide a user operation interface so that the user can perform corresponding system optimization and maintenance according to the diagnostic results.

[0069] Embodiment 3 of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. The electronic device is installed at the end of a photovoltaic power generation system. When the computer program is loaded into the processor, a photovoltaic power generation diagnostic method based on multidimensional influencing quantities as described in Embodiment 2 is implemented.

[0070] Embodiment 4 of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the photovoltaic power generation diagnosis method based on multi-dimensional influence quantities according to embodiment 1 is implemented.

[0071] Through the above implementation, the photovoltaic power generation diagnostic system of the present invention can comprehensively consider a variety of influencing quantities, provide accurate diagnostic results, and help users optimize and maintain photovoltaic power generation systems, thereby improving the overall efficiency and stability of the system. The technical effects of the present invention include: 1. After the mapping diagram of the preset parameters of the present invention is loaded, it is directly optimized through real-time data to improve the accuracy of the mapping diagram and ensure timeliness; 2. The present invention can realize high-frequency collection of relevant data through electronic equipment installed at the end of the photovoltaic power generation system, and realize efficient and accurate data processing through edge computing; 3. The present invention is mainly applicable to distributed photovoltaic power generation systems.

[0072] In summary, the present invention can ensure high-frequency collection and calculation of multi-dimensional influencing factors by deploying electronic devices on the edge side, improve the accuracy of photovoltaic output prediction, and further improve the accuracy of photovoltaic power generation system diagnosis. At the same time, the power generation diagnosis method mentioned in the present invention can realize accurate diagnosis and prediction of each single photovoltaic power generation system through distributed installation, and on this basis, realize accurate prediction of photovoltaic power generation in the entire area or even in a group composed of multiple areas.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A photovoltaic power generation diagnosis method based on multi-dimensional influence quantity, characterized in that: The steps include: High-frequency collection of photovoltaic power generation grid-connected point data and its multi-dimensional impact data, and pre-processing of the data; Based on the preprocessed data, a mapping relationship between multi-dimensional influencing variables and photovoltaic output is formed through clustering statistics. Based on this mapping relationship, the real-time status of photovoltaic output is evaluated to diagnose problems in the photovoltaic system. The final diagnosis results are displayed, and users can perform corresponding system optimization and maintenance based on the diagnosis results.

2. A photovoltaic power generation diagnosis method based on multi-dimensional influence quantities according to claim 1, characterized in that: The photovoltaic power generation system grid connection point data and its multi-dimensional influencing quantity data include power generation parameters and environmental parameters of the photovoltaic power generation system. The environmental parameters include light intensity, temperature and humidity, and the power generation parameters include power generation, voltage and current.

3. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 1, characterized in that: The mapping relationship between multi-dimensional influence quantity and photovoltaic output formed by clustering statistics includes: The mapping relationship between the multi-dimensional influencing quantity and the photovoltaic output is statistically analyzed by the DBSCAN clustering algorithm, and the mapping relationship is formed by clustering clusters automatically determined by the DBSCAN clustering algorithm.

4. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 1, characterized in that: The data preprocessing includes format conversion and data screening.

5. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 4 is characterized by: The format conversion includes converting the data formats of multiple influencing quantities into a unified data format.

6. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 4, characterized in that: The data screening includes removing periodic fluctuations, eliminating short-term random disturbances, measuring noise and decomposing low-frequency components in the extracted data.

7. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 6 is characterized by: The decomposition and extraction of low-frequency components in the data includes: The Savitzky-Golay filter is based on local polynomial fitting. It uses the data in the sliding window to construct a polynomial function to fit the local data, and then uses the fitting result to replace the data point in the center of the window. The formula of the Savitzky-Golay filter is: in, is the smoothed signal, N is the sliding window size, indicating the number of data points included in each fitting, c k is the filter coefficient calculated by polynomial fitting, y(t) represents the original signal, and y(t+k) represents the signal at time t+k.

8. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 1, characterized in that: The mapping relationship between multi-dimensional influence quantity and photovoltaic output is formed by clustering statistics, and the real-time status of photovoltaic output is evaluated based on the mapping relationship. The problems existing in the photovoltaic system are diagnosed as follows: Taking multidimensional influencing variables as independent variables and photovoltaic output as dependent variables, the mapping relationship between multidimensional influencing variables and photovoltaic output can be formed based on high-frequency data collection. Based on this mapping relationship, the theoretical output value of the photovoltaic system under different environmental conditions can be calculated. Compare and analyze the calculated theoretical output value with the actual measured value to identify the deviation of system performance through comparison; These deviations are further analyzed, and when the system deviation is greater than the set threshold, it is diagnosed that there may be problems with the photovoltaic system, and the abnormal results of the photovoltaic system are output.

9. The photovoltaic power generation diagnosis method based on multi-dimensional influence quantity according to claim 8, characterized in that: The reasons cited for the deviations included equipment failure and excessive temperature.

10. A photovoltaic power generation diagnosis system based on multi-dimensional influence quantities, used to execute a photovoltaic power generation diagnosis method based on multi-dimensional influence quantities according to any one of claims 1 to 9, characterized in that: The system includes a data acquisition module, a data processing module, a multi-dimensional influence quantity analysis module, a diagnosis and evaluation module and a user interaction interface; The data acquisition module includes a system operation parameter acquisition unit and an environmental parameter acquisition unit, which are used to collect relevant data of the operation of the photovoltaic power generation system in real time, including the power generation parameters and environmental parameter data of the photovoltaic power generation system, and transmit the data to the data processing module; the data processing module pre-processes the received data of different types, and transmits the processed data to the multi-dimensional influence quantity analysis module; the multi-dimensional influence quantity analysis module is used to establish a mapping relationship diagram based on real-time data, which is used to reflect the relationship between the performance of the photovoltaic system and various influence quantities; the diagnosis and evaluation module evaluates the performance of the photovoltaic power generation system according to the multi-dimensional influence quantity mapping relationship diagram, and outputs the diagnosis results; the user interaction interface is used to display the diagnosis results and provide a user operation interface, so that the user can perform corresponding system optimization and maintenance according to the diagnosis results.

Images