A Big Data-Based System and Method for Insight and Analysis of Low-Performance Operation of Photovoltaic Strings

Abstract

This invention provides a big data-based system and method for analyzing and understanding the performance degradation and low-performance operation of photovoltaic (PV) strings. The system collects electrical serial numbers and production status data of string inverters or DC combiner boxes and PV strings; analyzes the electrical relationship between the PV strings and upstream equipment based on the serial numbers to determine power generation analysis units; aligns the data time, performs preprocessing, and verifies the valid model input data for preparing the performance analysis calculation model; uses the performance analysis calculation model to calculate the power generation performance evaluation indicators corresponding to each power generation analysis unit; optimizes the power generation string structure based on the calculated power generation performance evaluation indicators, and decides on operation and maintenance handling schemes matching different evaluation indicator levels. This solution overcomes the shortcomings of insufficient reliability in existing technology analysis results, employs specially designed data processing and analysis algorithms to analyze abnormal and inefficient states among numerous strings, and proposes improvement measures accordingly, thereby effectively improving the overall performance of the PV system.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic string performance analysis technology, and in particular to a system and method for insight analysis of low-performance operation of photovoltaic strings based on big data. Background Technology

[0002] Under sunlight, photovoltaic (PV) panels absorb light energy, resulting in the accumulation of opposite charges at both ends of the panel. This generates an electromotive force (EMF), converting light energy into electrical energy. As a crucial energy conversion device, PV modules are susceptible to damage from factors such as surface contamination and performance degradation. Theoretically, strings within the same power generation unit should maintain consistent power output (or current / voltage values) or exhibit only slight differences under identical operating conditions. However, environmental and human factors can prevent strings from reaching full capacity, significantly reducing the power generation efficiency of PV power plants. Therefore, effectively identifying inefficient strings using specialized string inefficiency algorithms is crucial for improving overall power generation efficiency.

[0003] Existing research includes several methods for identifying low-performance conditions in photovoltaic (PV) strings. The main approach involves starting with historical string data from specific time periods, using clustering algorithms to identify strings with defined characteristics, and then inferring the corresponding inefficient strings. This identification of abnormal PV strings allows for processing, thereby promoting economical and efficient operation and maintenance of power plants. For example, one existing technology provides a method for identifying inefficient PV strings. This method acquires current and voltage string data for each PV string over a preset time period within a historical period; processes the data to identify and remove abnormal data; then filters the remaining string data to find strings within the first preset time period, obtaining the number of filtered PV strings; finally, it selects clustering objects to identify inefficient PV strings in the photovoltaic system. While this method overcomes the impact of abnormal values ​​on the identification results to some extent, the limited current and voltage data for PV strings makes it difficult to provide a comprehensive data foundation for clustering algorithms, resulting in limitations in the accuracy of the identification results. Other research focuses on identifying and improving the power output of photovoltaic (PV) modules. For example, one existing technology provides a method for identifying and improving the power output of PV modules under multiple orientations and tilt angles. This method stores preprocessed data in the cloud and constructs an MLP-Mixer model using the TensorFlow deep learning framework. The output of the MLP-Mixer model is integrated with prior knowledge of the PV power plant to form a second dataset, which is then used to build an XGBoost model. During application, real-time operating data of the PV power plant is collected and fed into the MLP-Mixer model. The MLP-Mixer model identifies inefficient strings, and the XGBoost model generates maintenance suggestions. While this method combines the identification results with prior knowledge to build an abnormal maintenance suggestion analysis model, the analysis results depend on the identification results. The accuracy of inefficient string information identified solely based on historical string data is difficult to guarantee, leading to insufficient reliability of the maintenance suggestion analysis results. In addition, some photovoltaic equipment uses alarm information to identify inefficient photovoltaic modules or manually identifies inefficient photovoltaic modules based on the power generation curve of the equipment. However, the method of using alarm information lacks accuracy. General equipment remote signaling alarms usually do not report the status of string inefficient operation. Checking the power generation curve of the equipment will consume a lot of labor costs and on-site operation time costs. Moreover, these traditional methods will lead to missed judgments and misjudgments, and cannot well meet the operational needs of the photovoltaic field.

[0004] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a big data-based photovoltaic (PV) string low-performance operation insight and analysis system. This system overcomes the reliability limitations of existing technologies by employing specialized data processing and analysis algorithms to analyze abnormally inefficient states among numerous strings, effectively identifying poorly performing strings and providing early warnings and improvement measures, thereby facilitating an overall improvement in PV system performance. The system collects electrical serial numbers and production status data of string inverters or DC combiner boxes and PV strings; analyzes the electrical relationship between PV strings and upstream equipment based on serial numbers to determine power generation analysis units; aligns data time, preprocesses and verifies it to prepare valid model input data for the performance analysis calculation model; uses the performance analysis calculation model to calculate the power generation performance evaluation indicators corresponding to each power generation analysis unit; optimizes the power generation business string structure based on the calculated power generation performance evaluation indicators, and determines the operation and maintenance handling scheme matching different evaluation indicator levels. Preferably, in one embodiment, the system includes:

[0006] The data collection module is configured to acquire electrical serial numbers and production status data of all string inverters or DC combiner boxes and photovoltaic strings in the photovoltaic power station to be analyzed.

[0007] The unit structure determination module is configured to analyze the electrical relationship between the photovoltaic string and the upstream string inverter or DC combiner box based on the electrical number, and determine the power generation analysis unit based on the digital model of the power generation unit corresponding to the photovoltaic string and its upstream equipment.

[0008] The data processing module is configured to align the time of the acquired data with different electrical numbers, preprocess the data, and verify the data based on the preprocessed data to prepare effective model input data for the performance analysis calculation model.

[0009] The string performance analysis module is configured to start the constructed performance analysis calculation model according to the set analysis calculation start time, and input the valid model input data of the power generation analysis unit to determine the power generation performance evaluation index corresponding to each power generation analysis unit.

[0010] The business optimization application module is configured to select the optimal power generation business string structure based on calculation-based power generation performance evaluation indicators, and to decide on the operation and maintenance handling scheme matching different evaluation indicator levels.

[0011] In an optional embodiment, the data processing module includes a data time alignment and partitioning unit, which is configured to perform time alignment processing on the collected equipment time-series production status data using a data information model of the same adjacent time, and to select power generation status data objects for storage according to preset effective data time periods and analysis nodes.

[0012] Furthermore, in one embodiment, the data processing module includes a data preprocessing unit for performing data cleaning and data completion processing on the data object; the data completion processing includes using interpolation to supplement missing data values.

[0013] In a preferred embodiment, the data processing module includes a validity check unit, which performs checks on the preprocessed data according to the following logic to prepare valid model input data for the performance analysis computation model:

[0014] Determine whether the data volume of each photovoltaic string corresponding to different analysis nodes meets the set data volume conditions, select the data of photovoltaic strings that do not meet the data volume conditions, and use the data of the remaining photovoltaic strings as standby power generation status data.

[0015] Determine whether the number of photovoltaic strings under the corresponding power generation analysis unit of the standby power generation status data meets the set string quantity condition, select the power generation analysis unit that meets the string quantity condition, and use its standby power generation status data as valid model input data.

[0016] In one embodiment, during the process of the string performance analysis module calculating the power generation performance evaluation index of different power generation analysis units, the start time of the analysis model is first set. The time when the irradiance of the photovoltaic power station location is greater than the set condition for the last time of the day is set as the calculation start time point of the daily performance analysis calculation model.

[0017] Furthermore, in one embodiment, the string performance analysis module calculates the average capacity utilization rate and standard deviation of different power generation analysis units above the multi-photovoltaic string as power generation performance evaluation indicators based on the performance analysis calculation model.

[0018] The performance analysis calculation model adopts the following logic:

[0019] ,

[0020] ,

[0021] ,

[0022] in, It is the standard deviation of the daily average capacity utilization rate of the photovoltaic strings corresponding to the current power generation analysis unit. m It is the number of strings under the power generation analysis unit. Indicates the first generation in the power generation analysis unit i The average daily capacity utilization of each string group. This is the result of averaging the daily capacity utilization rate of all strings under the power generation analysis unit again. This represents the average daily capacity utilization rate for each string.n This represents the number of string capacity utilization rates calculated by a single analysis node each day. Indicates the number of days j Capacity utilization data for each analysis node. This represents the capacity utilization rate of a single analysis node in a photovoltaic string. VI The DC-side power of the photovoltaic string represents the power of a single analysis node. Wp This represents the installed capacity of the photovoltaic string.

[0023] Preferably, in one embodiment, the business optimization application module includes a string structure optimization unit, which is configured to combine the power generation performance evaluation index with the set optimization standard to compare and analyze the results and eliminate low-performance photovoltaic strings, thereby realizing the optimization and marking of the power generation operation string structure.

[0024] In an optional embodiment, the business optimization application module includes an operation and maintenance solution decision unit, which is configured to determine different levels of performance evaluation levels based on set performance index division rules, corresponding to different degrees of performance degradation, and different forms of early warning and display for different performance evaluation levels, matching operation and maintenance handling solutions with different degrees of urgency.

[0025] Preferably, in one embodiment, the system further includes an operation and maintenance effect analysis module, which is configured to mark and record operation and maintenance information of photovoltaic strings that have adopted operation and maintenance schemes, update the data tags of the operation and maintenance strings, compare the power generation performance evaluation indicators of the power generation analysis unit before and after operation and maintenance, analyze the optimization effect of the operation and maintenance scheme, and provide data support for the operation and maintenance scheme decision-making unit.

[0026] Based on the application aspects of the system described in any one or more of the above embodiments, the present invention also provides a method for insight analysis of photovoltaic string performance degradation and low-performance operation based on big data. This method is applied to the system described in any one or more of the above embodiments. In a preferred embodiment, the method includes:

[0027] The data collection module is used to obtain the electrical serial numbers of all string inverters or DC combiner boxes in the photovoltaic power station to be analyzed, as well as the photovoltaic string production status data corresponding to all measurement points in the field.

[0028] The electrical relationship between the photovoltaic string and the upstream string inverter or DC combiner box is analyzed based on the electrical numbering, and the power generation analysis unit is determined based on the digital model of the power generation unit corresponding to the photovoltaic string and its upstream equipment.

[0029] After aligning the time of the acquired electrical number data, the data is preprocessed and verified based on the preprocessed data to prepare effective model input data for the performance analysis calculation model;

[0030] The performance analysis and calculation model is started according to the set analysis and calculation start time. The effective model input data of the power generation analysis unit is input to determine the power generation performance evaluation index corresponding to each power generation analysis unit.

[0031] The calculation-based power generation performance evaluation index is used to select the optimal power generation service string structure and determine the operation and maintenance handling scheme matching different evaluation index levels.

[0032] Based on other aspects of the methods described in any one or more of the above embodiments, the present invention also provides a storage medium storing program code capable of implementing the methods described in the above embodiments.

[0033] Compared with the closest prior art, the present invention also has the following beneficial effects:

[0034] This invention provides a big data-based photovoltaic string low-performance operation insight and analysis system, including a data collection module, a unit structure determination module, a data processing module, a string performance analysis module, and a business optimization application module;

[0035] This system collects electrical serial numbers and production status data of string inverters or DC combiner boxes and photovoltaic strings. Based on these serial numbers, it analyzes the electrical relationship between the photovoltaic strings and upstream equipment to determine power generation analysis units. After aligning the data time, it performs preprocessing and verifies the valid model input data for preparing the performance analysis calculation model. The performance analysis calculation model is used to calculate the power generation performance evaluation indicators corresponding to each power generation analysis unit. Based on the calculated power generation performance evaluation indicators, it optimizes the power generation business string structure and decides on operation and maintenance handling schemes matching different evaluation indicator levels. This invention utilizes the established performance analysis calculation module to calculate performance evaluation indicators for insight analysis of low-performance operating strings in the plant. It deeply analyzes the performance characteristics of low photovoltaic strings in the standard deviation, and provides appropriate early warning levels and corresponding handling methods for the identified low-performance strings based on business experience and on-site measurements. This improves the accuracy of low-performance operation insight identification and saves computational and handling resources caused by misjudgments.

[0036] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0037] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0038] Figure 1This is a schematic diagram illustrating the structure of a big data-based photovoltaic string low-performance operation insight and analysis system provided in an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the operation process of a method for insight analysis of low-performance operation of photovoltaic strings based on big data, provided in another embodiment of the present invention. Detailed Implementation

[0040] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples. Those skilled in the art will then fully understand how the present invention uses technical means to solve technical problems and achieve technical effects, and will be able to implement the present invention specifically based on the above-described implementation process. It should be noted that, as long as there is no conflict, the various embodiments and features of the present invention can be combined with each other, and the resulting technical solutions are all within the protection scope of the present invention.

[0041] Although the flowchart describes the operations as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. The order of the operations can be rearranged. A process can terminate when its operation is complete, but it may also have additional steps not included in the diagram. A process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0042] Computer equipment includes user equipment and network equipment. User equipment or clients include, but are not limited to, computers, smartphones, and PDAs (Personal Digital Assistants); network equipment includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Computer equipment can operate independently to implement this invention, or it can connect to a network and implement this invention through interaction with other computer equipment within the network. The network in which the computer equipment resides includes, but is not limited to, the Internet, wide area networks (WANs), metropolitan area networks (MANs), local area networks (LANs), and VPN networks.

[0043] The terms “first,” “second,” etc., may be used herein to describe various units, but these units should not be limited by these terms; they are used merely to distinguish one unit from another. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. When a unit is referred to as “connected” or “coupled” to another unit, it may be directly connected or coupled to said other unit, or there may be intermediate units present.

[0044] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms “a” and “an” as used herein are also intended to include the plural. It should also be understood that the terms “comprising” and / or “including” as used herein specify the presence of the stated features, integers, steps, operations, units, and / or components, without excluding the presence or addition of one or more other features, integers, steps, operations, units, components, and / or combinations thereof.

[0045] In operation, multiple photovoltaic (PV) modules are typically connected in series to form a PV string. Multiple PV strings are further connected in parallel to form a whole power generation unit. Multiple power generation units support the operation of PV projects. Theoretically, under the same operating conditions, the power generation (or current / voltage value) of strings within the same power generation unit should always be consistent or have only slight differences. However, certain environmental or human factors can cause strings to not generate full power, which will significantly reduce the power generation efficiency of PV power plants. Effectively identifying inefficient strings through professional string inefficiency algorithms plays an important role in improving power generation efficiency.

[0046] Existing research includes several methods for identifying inefficient photovoltaic (PV) strings. The main approach involves starting with historical string data from specific time periods. This historical data is then filtered to remove outliers and used as a baseline for clustering algorithms to identify strings with defined characteristics. This process helps identify inefficient strings and allows for their processing, thus promoting more economical and efficient power plant operation and maintenance. For example, one existing technology provides a method for identifying inefficient PV strings. This method acquires current and voltage string data for a preset duration within a historical time period; processes the data to identify and remove outliers; then filters the remaining string data to find strings within the first preset time period, determining the number of filtered strings; and finally, selects a clustering object to identify inefficient PV strings. While this method overcomes the impact of outliers to some extent, the limited current and voltage data for PV strings makes it difficult to provide a comprehensive data foundation for clustering algorithms, resulting in limitations in the accuracy of the identification results.

[0047] Another area of ​​research concerns the identification and power generation improvement of photovoltaic (PV) equipment components. For example, one existing technology provides a method for identifying inefficiencies and improving power generation of PV modules under multiple orientations and tilt angles. This method collects historical operating data from PV power plants, stores the pre-processed data in cloud space for further processing, and constructs an MLP-Mixer model using the TensorFlow deep learning framework. The output of the MLP-Mixer model is integrated with prior knowledge of the PV power plant to form a second dataset, which is then used to build an XGBoost model. During application, real-time operating data from the PV power plant is collected and fed into the MLP-Mixer model. The MLP-Mixer model identifies inefficient strings, and the XGBoost model generates maintenance suggestions. While this method combines the identification results with prior knowledge to build an abnormal maintenance suggestion analysis model, the analysis results depend on the identification results. The accuracy of inefficient string information identified solely based on historical string data is difficult to guarantee, leading to insufficient reliability of the maintenance suggestion analysis results.

[0048] In addition, some photovoltaic equipment uses alarm information to identify inefficient photovoltaic modules or manually identifies inefficient photovoltaic modules based on the equipment's power generation curve. However, the method of using alarm information lacks accuracy. General equipment remote signaling alarms usually do not report the status of string inefficient operation. Checking the equipment's power generation curve will consume a lot of labor costs and on-site operation time costs, and these traditional methods will lead to missed judgments and misjudgments.

[0049] To address the aforementioned issues, this invention utilizes big data processing technology for power generation performance combined with a matching algorithm model to analyze string inefficiency, thereby improving the comprehensiveness and reliability of the analysis results. By combining the big data processing logic for power generation performance with the algorithm model to assess the power stability of the strings, it effectively filters out strings with poor power generation and issues early warnings accordingly. This plays a crucial role in enabling users to identify problematic strings as soon as possible and arrange maintenance in a timely manner.

[0050] The proposed big data-based photovoltaic string low-performance operation insight analysis system and method first acquires the current, voltage, and capacity data of photovoltaic strings under the DC combiner box or string inverter of the entire field, as well as the irradiance data of the field. Based on the electrical numbering information of each string, the electrical relationships between the strings and their upstream equipment are organized to form a digital model of the power generation unit, which serves as the power generation analysis unit. Data within the effective power generation time period is selected for each unit, and all data is preprocessed for data quality judgment and optimization control. Then, the established performance analysis and calculation module is used to calculate performance evaluation indicators to conduct insight analysis on the low-performance operating strings in the field. For the identified low-performance strings, appropriate warning levels and corresponding handling methods are given based on business experience and on-site measurements. This invention improves the effective insight into low-performance operation by deeply understanding the performance characteristics of photovoltaic strings in the standard deviation, combined with the construction of string electrical digital models, reducing misjudgments and the loss of significant computing resources.

[0051] The structural components, connection methods, and functional principles of the system according to embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Although the logical order of each operation is shown in the description of the system's structural operation, in some cases, the operations shown or described may be performed in a different order than that shown here.

[0052] Example 1

[0053] Figure 1 This diagram illustrates the structure of the photovoltaic string low-performance operation insight and analysis system based on big data provided in Embodiment 1 of the present invention. (Refer to...) Figure 1 It can be seen that the system includes:

[0054] The data collection module is configured to acquire electrical serial numbers and production status data of all string inverters or DC combiner boxes and photovoltaic strings in the photovoltaic power station to be analyzed.

[0055] The unit structure determination module is configured to analyze the electrical relationship between the photovoltaic string and the upstream string inverter or DC combiner box based on the electrical number, and determine the power generation analysis unit based on the digital model of the power generation unit corresponding to the photovoltaic string and its upstream equipment.

[0056] The data processing module is configured to align the time of the acquired data with different electrical numbers, preprocess the data, and verify the data based on the preprocessed data to prepare effective model input data for the performance analysis calculation model.

[0057] The string performance analysis module is configured to start the constructed performance analysis calculation model according to the set analysis calculation start time, and input the valid model input data of the power generation analysis unit to determine the power generation performance evaluation index corresponding to each power generation analysis unit.

[0058] The business optimization application module is configured to select the optimal power generation business string structure based on calculation-based power generation performance evaluation indicators, and to decide on the operation and maintenance handling scheme matching different evaluation indicator levels.

[0059] The photovoltaic string low-performance operation insight and analysis system based on big data, as described in this invention, monitors and collects the working status information of photovoltaic strings in real time, uses specific data processing and analysis algorithms to identify abnormal and inefficient states hidden among millions of strings, and proposes improvement measures accordingly, thereby effectively improving the overall performance of the photovoltaic system.

[0060] In an optional embodiment, the photovoltaic string production status data includes photovoltaic string number data, string current and voltage data, meteorological irradiance data, collector line power data, and unit string installed capacity data;

[0061] In practical applications, the following operations are performed through the data collection module:

[0062] Obtain the electrical serial numbers of all DC combiner boxes or string inverters (depending on the actual electrical structure of the photovoltaic power station) and string electrical serial numbers.

[0063] Based on the corresponding measurement points, obtain the current and voltage data of different photovoltaic strings under the DC combiner box or string inverter;

[0064] Real-time irradiance data from the meteorological station of the photovoltaic power station to be analyzed are obtained based on the corresponding measuring points.

[0065] Active power data of different collector lines of the photovoltaic power station to be analyzed are collected based on the corresponding measuring points;

[0066] Obtain the basic parameters of the installed capacity of different strings.

[0067] The unit structure determination module analyzes the electrical numbering of the photovoltaic string and the string inverter or DC combiner box, determines the upstream equipment matched to each photovoltaic string, and constructs a digital model of the power generation unit as a power generation analysis unit based on the docking relationship between the photovoltaic string and its upstream equipment.

[0068] In practical applications, the electrical numbers of combiner boxes or string inverters (depending on the actual electrical structure of the photovoltaic power station) and string electrical numbers are organized; a digital model of the power generation unit relationship between the string and its upstream equipment is constructed, and each digital model of the power generation unit is used as a power generation analysis unit.

[0069] For example, when this embodiment of the invention is applied to a 4.2MWp photovoltaic project for railway construction, the parent device (string inverter or DC combiner box) of all photovoltaic strings is used as a power generation analysis unit to calculate the power performance of the photovoltaic strings.

[0070] The data processing module is configured to align the time of the acquired different electrical number data, preprocess the data, and verify the preprocessed data to prepare effective model input data for the performance analysis calculation model.

[0071] The data processing module includes a data time alignment and partitioning unit, which is configured to perform time alignment processing on the collected equipment time-series production status data using a data information model with the same adjacent time. The unit also selects power generation status data objects for storage based on preset effective data time periods and analysis nodes.

[0072] The effective data period determines the length of the window for using production status data during calculation, i.e., the data range used by the daily performance analysis calculation model. In a preferred embodiment, the power generation status data within the effective data period of the daily power station is selected as the effective power generation status data. The effective data (collection) period of the daily power station is set from the time when the local irradiance is greater than the set value for the first time to the time when the local irradiance is greater than the set value for the last time of the day. In a preferred embodiment, the set value can be set to 120W / ㎡. In actual application, the set value is flexibly set according to seasonal factors, meteorological factors, and geographical factors. It should be noted that, based on the technical logic of this invention, technicians can set the set value to other values ​​or data ranges according to their needs.

[0073] In an optional embodiment, after aligning the time of all data, each interval is set as an analysis node to determine the storage time of the data object; the set interval can be set to 5 minutes. In actual application, the set interval is flexibly set according to the performance analysis status and expected analysis needs of each power generation analysis unit. It should be noted that, based on the technical logic of the present invention, technicians can set the set interval to other values ​​or data ranges according to their needs.

[0074] Based on this, when applying the data, the data closest to the exact moment can be selected every 5 minutes. Usually, the first data point every 5 minutes is selected for storage as the collected power generation status data object.

[0075] On the other hand, for photovoltaic power plants with multiple power generation analysis units, technicians can set different time intervals for each unit based on its dynamic performance analysis status (e.g., the previous day's performance analysis results) and expected analysis needs. For example, for a power generation analysis unit exhibiting a low performance trend, its time interval can be set shorter than that of a unit with normal performance, to generate sufficiently dense and effective power generation status data. Conversely, for a power generation analysis unit exhibiting high or normal performance throughout the set time period, its time interval can be set longer, dynamically saving data acquisition and computation resources without affecting the reliability of the analysis results.

[0076] The data processing module includes a data preprocessing unit, which is used to perform data cleaning and data completion processing on the data objects.

[0077] In an optional embodiment, during the data preprocessing process, the data preprocessing unit performs data quality judgment on the collected data, and performs data cleaning and data completion processing based on the judgment results. The data cleaning process includes removing constant values ​​and out-of-limit values; the data completion process includes using interpolation to supplement missing data values.

[0078] The data processing module includes a validity check unit, which performs checks on the preprocessed data according to the following logic to prepare valid model input data for the performance analysis calculation model:

[0079] Determine whether the data volume of each photovoltaic string corresponding to different analysis nodes meets the set data volume conditions, select the data of photovoltaic strings that do not meet the data volume conditions, and use the data of the remaining photovoltaic strings as standby power generation status data.

[0080] Determine whether the number of photovoltaic strings under the corresponding power generation analysis unit meets the set string quantity condition. Select the power generation analysis unit that meets the string quantity condition and use its standby power generation status data as valid model input data. If it does not meet the condition, it indicates that the current standby power generation status of the power generation analysis unit does not meet the input data quality requirements of the performance analysis calculation model.

[0081] The technical means described above can effectively avoid interference with the accuracy and authenticity of the performance model calculation results caused by data quality issues and insufficient data volume.

[0082] The string performance analysis module is configured to start the constructed performance analysis calculation model according to the set analysis calculation start time, and use the effective model input data of the selected power generation analysis station as the calculation input to determine the power generation performance evaluation index corresponding to each power generation analysis unit.

[0083] In the process of calculating the power generation performance evaluation index of different power generation analysis units, the string performance analysis module first sets the start time of the analysis model. In an optional embodiment, it sets the last irradiance at the location of the photovoltaic power station each day to be greater than a set condition (e.g., 120W / m). 2 The time is the calculation start time of the daily performance analysis calculation model.

[0084] The string performance analysis module calculates the average capacity utilization rate and standard deviation of different power generation analysis units above the multi-PV string based on the performance analysis calculation model, which are used as power generation performance evaluation indicators.

[0085] The performance analysis calculation model adopts the following logic:

[0086] ,

[0087] ,

[0088] ,

[0089] in, It is the standard deviation of the daily average capacity utilization rate of the photovoltaic strings corresponding to the current power generation analysis unit. m It is the number of strings under the power generation analysis unit. Indicates the first generation in the power generation analysis unit i The average daily capacity utilization of each string group. This is the result of averaging the daily capacity utilization rate of all strings under the power generation analysis unit again. This represents the average daily capacity utilization rate for each string. n This represents the number of string capacity utilization rates calculated by a single analysis node each day. Indicates the number of days j Capacity utilization data for each analysis node. This represents the capacity utilization rate of a single analysis node in the photovoltaic string (every 5 minutes). VI The DC-side power of the photovoltaic string represents the power of a single analysis node. Wp This represents the installed capacity of the photovoltaic string.

[0090] The performance analysis calculation model based on this invention first calculates the capacity utilization rate of each string within the daily effective power generation segment of each analysis node, then calculates the average daily capacity utilization rate of each string, and then, based on the average daily capacity utilization rate, introduces the average daily capacity utilization rate of all strings under the power generation analysis unit, calculates the standard deviation index of the capacity utilization rate mean data distribution, and uses it as the power generation performance evaluation index corresponding to the power generation analysis unit.

[0091] The business optimization application module is configured to select the optimal power generation business string structure based on calculation-based power generation performance evaluation indicators, and to decide on the operation and maintenance handling scheme matching different evaluation indicator levels.

[0092] The business optimization application module includes a string structure optimization unit, which is configured to combine the power generation performance evaluation index with the set optimization standard to compare and analyze the results and eliminate low-performance photovoltaic strings, thereby realizing the optimal marking of the power generation operation string structure.

[0093] The string structure optimization unit eliminates low-performance photovoltaic strings based on the following logic, combining the power generation performance evaluation indicators with the set optimization standards and comparative analysis results:

[0094] The dynamically calculated power generation performance evaluation index is compared with the set optimization standard. If the power generation performance evaluation index does not meet the set optimization standard, the photovoltaic string with the smallest average daily capacity utilization rate under the power generation analysis unit is selected as the current low-performance photovoltaic string. After being removed, the power generation performance evaluation index of the power generation analysis unit is recalculated based on the effective model input data of the remaining photovoltaic strings. The process of judging and selecting low-performance photovoltaic strings for removal is repeated until the calculated power generation performance evaluation index of the power generation analysis unit meets the set optimization standard. The remaining photovoltaic strings are marked as high-performance power generation operation string structures, which facilitates the station to carry out targeted scheduling operations according to needs.

[0095] In practical applications, if the calculated standard deviation does not meet the optimization criteria (e.g., the standard deviation is less than 5%), the calculated standard deviation can be marked on the string with the lowest daily average capacity utilization, and then that string can be removed. The calculation is then performed again based on the remaining strings in the power generation analysis unit. The results are compared with the set optimization criteria and the calculated standard deviation is marked on the string with the lowest daily average capacity utilization in this calculation. The loop continues until the last standard deviation meets the optimization criteria, at which point the loop stops.

[0096] The business optimization application module includes an operation and maintenance solution decision unit, which is configured to determine different levels of performance evaluation grades based on set performance index division rules, corresponding to different degrees of performance degradation, and different forms of early warning and display for different performance evaluation grades, matching operation and maintenance handling solutions with different urgency levels.

[0097] In an optional embodiment, based on the analysis of the historical model calculation results and corresponding on-site measurement experience, the standard deviation value distribution range corresponding to different performance evaluation levels can be set as the performance index division rule. For example, the performance index division rule can be set as follows: when the standard deviation results are ≤5%, >5% and ≤10%, >10% and ≤20%, and >20%, respectively, they correspond to four performance evaluation levels from low to high: first, second, third, and fourth. Correspondingly, they correspond to four warning levels of the low performance operation impact of the string: normal, slightly inefficient, moderately inefficient, and severely inefficient. Different warning levels correspond to different operation and maintenance methods.

[0098] When the model detects an anomaly in a string, the system automatically triggers an alarm, which is displayed on the interface as an indicator light. A report containing fault analysis and suggested solutions is also generated. The report is detailed and easy to understand, facilitating rapid action by maintenance personnel.

[0099] The system also includes an operation and maintenance effect analysis module, which is configured to mark and record operation and maintenance information of the photovoltaic strings that have undergone operation and maintenance, compare the power generation performance evaluation indicators of the power generation analysis unit before and after operation and maintenance, record the effect of each measure taken, analyze the optimization effect of the operation and maintenance plan, and provide data support for the operation and maintenance plan decision-making unit. Accordingly, after the photovoltaic strings have undergone operation and maintenance, the markings should be updated in a timely manner, and low-performance markings should be removed to participate in the evaluation process of the power generation analysis unit and the optimization process of the power generation string structure.

[0100] The photovoltaic string low-performance operation insight and analysis system based on big data provided in this invention displays the results information in response to user operation commands through a visual interactive interface. Users can query the performance analysis and calculation results of the photovoltaic string of the required equipment through the interactive interface.

[0101] By employing a data filtering mechanism, normal photovoltaic (PV) string objects and information are removed, and the output results can intuitively display problematic strings and their types or levels of problems. In an optional embodiment, strings with poor power generation can be filtered out based on a visual interface and displayed with different levels of warnings. Simultaneously, the visual interface uses diverse color rendering for different types and levels of degradation warnings. Based on this intuitive output information, users can ensure that problematic strings are addressed and resolved promptly, improving power generation efficiency and ensuring stable operation.

[0102] This invention utilizes big data combined with algorithmic models to analyze string inefficiencies, effectively improving the comprehensiveness and reliability of the analysis results. Furthermore, it refines the classification according to the degree of string inefficiency, allowing for more accurate scheduling of operations and maintenance, thus guiding string maintenance work more precisely. This further reduces maintenance costs and improves the efficiency of photovoltaic string maintenance.

[0103] In the photovoltaic string low-performance operation insight analysis system based on big data provided in this embodiment of the invention, each module or unit structure can operate independently or in combination according to actual data processing needs and model calculation needs to achieve corresponding technical effects.

[0104] Example 2

[0105] The above-described embodiments of the present invention have provided a detailed description of the system. Based on other aspects of the system described in any one or more of the above embodiments, the present invention also provides a method for insight analysis of low-performance operation of photovoltaic strings based on big data. This method is applied to the system for insight analysis of low-performance operation of photovoltaic strings based on big data described in any one or more of the above embodiments. Specific embodiments are given below for detailed description.

[0106] Specifically, Figure 2 The diagram illustrates a flowchart of the method for analyzing low-performance operation of photovoltaic strings based on big data, as provided in an embodiment of the present invention. Figure 2 As shown, the method includes the following steps:

[0107] Data collection steps: Obtain the electrical serial numbers and production status data of all string inverters or DC combiner boxes and photovoltaic strings in the photovoltaic power station to be analyzed;

[0108] Unit structure determination steps: Analyze the electrical relationship between the photovoltaic string and the upstream string inverter or DC combiner box based on the electrical number, and determine the power generation analysis unit based on the digital model of the power generation unit corresponding to the photovoltaic string and its upstream equipment;

[0109] Data processing steps: After aligning the time of the acquired data, preprocess the data and verify the preprocessed data to prepare effective model input data for the performance analysis calculation model;

[0110] String performance analysis steps: Start the constructed performance analysis calculation model according to the set analysis calculation start time, input the valid model input data of the power generation analysis unit to determine the power generation performance evaluation index corresponding to each power generation analysis unit;

[0111] Business optimization application steps: Based on the calculation of power generation performance evaluation indicators, the optimal marking of power generation business string structure is selected, and the operation and maintenance handling scheme matching different evaluation indicator levels is determined.

[0112] In an optional embodiment, the data processing step includes:

[0113] The collected equipment time-series production status data is time-aligned using a data information model based on the same proximity time. Power generation status data objects are selected and stored according to the preset effective data period and analysis node.

[0114] Furthermore, in one embodiment, the data processing step includes:

[0115] Data preprocessing steps: used to perform data cleaning and data completion processing on data objects; the data completion processing includes using interpolation to fill in missing data values.

[0116] In a preferred embodiment, the data preprocessing step further includes:

[0117] Validity verification steps: Based on the preprocessed data, verification is performed to prepare valid model input data for the performance analysis computation model, including the following operations:

[0118] Determine whether the data volume of each photovoltaic string corresponding to different analysis nodes meets the set data volume conditions, select the data of photovoltaic strings that do not meet the data volume conditions, and use the data of the remaining photovoltaic strings as standby power generation status data.

[0119] Determine whether the number of photovoltaic strings under the corresponding power generation analysis unit of the standby power generation status data meets the set string quantity condition, select the power generation analysis unit that meets the string quantity condition, and use its standby power generation status data as valid model input data.

[0120] In one embodiment, the string performance analysis step includes:

[0121] Set the start time for the analysis model. Set the time when the irradiance at the location of the photovoltaic power station is greater than the set condition for the last time each day as the calculation start time point for the daily performance analysis calculation model.

[0122] Furthermore, in one embodiment, the string performance analysis step includes:

[0123] The average capacity utilization rate and standard deviation of different power generation analysis units above the multi-PV string are calculated based on the performance analysis calculation model as power generation performance evaluation indicators.

[0124] The string performance analysis step uses the following performance analysis calculation model:

[0125] ,

[0126] ,

[0127] ,

[0128] in, It is the standard deviation of the daily average capacity utilization rate of the photovoltaic strings corresponding to the current power generation analysis unit. m It is the number of strings under the power generation analysis unit. Indicates the first generation in the power generation analysis uniti The average daily capacity utilization of each string group. This is the result of averaging the daily capacity utilization rate of all strings under the power generation analysis unit again. This represents the average daily capacity utilization rate for each string. n This represents the number of string capacity utilization rates calculated by a single analysis node each day. Indicates the number of days j Capacity utilization data for each analysis node. This represents the capacity utilization rate of a single analysis node in a photovoltaic string. VI The DC-side power of the photovoltaic string represents the power of a single analysis node. Wp This represents the installed capacity of the photovoltaic string.

[0129] Preferably, in one embodiment, the service optimization application step includes:

[0130] String structure optimization steps: By combining the power generation performance evaluation indicators with the set optimization standards, low-performance photovoltaic strings are eliminated, thereby achieving the optimization and marking of the power generation operation string structure.

[0131] In an optional embodiment, the service optimization application step further includes:

[0132] Operation and maintenance solution decision steps: Based on the established performance index classification rules, determine the performance evaluation level of different levels, corresponding to different degrees of performance degradation. Different performance evaluation levels adopt different forms of early warning and display, and match operation and maintenance handling solutions with different urgency levels.

[0133] Preferably, in one embodiment, the method further includes an operation and maintenance effect analysis step: marking and recording the operation and maintenance information of the photovoltaic strings that have adopted the operation and maintenance plan, updating the data tags of the operation and maintenance strings, comparing the power generation performance evaluation indicators of the power generation analysis unit before and after operation and maintenance, analyzing the optimization effect of the operation and maintenance plan, and providing data support for the operation and maintenance plan decision-making unit.

[0134] For the foregoing method embodiments, in order to simplify the description, they are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0135] It should be noted that, in other embodiments of the present invention, the method can also combine one or more of the above embodiments to obtain a new big data-based method for insight analysis of low-performance operation of photovoltaic strings, so as to achieve comprehensive and reliable monitoring of the performance of photovoltaic equipment components.

[0136] Example 3

[0137] It should be noted that, based on the methods in any one or more embodiments of the present invention described above, the present invention also provides a storage medium storing program code that can implement the methods described in any one or more embodiments. When the program code is executed by the operating system, it can implement the above-described method for insight analysis of low-performance operation of photovoltaic strings based on big data.

[0138] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should be extended to equivalent substitutions of these features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0139] The phrase "an embodiment" in the specification means that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0140] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A big data-based system for analyzing and understanding the performance degradation and low-performance operation of photovoltaic strings, characterized in that, The system includes: The data collection module is configured to acquire electrical serial numbers and production status data of all string inverters or DC combiner boxes and photovoltaic strings in the photovoltaic power station to be analyzed. The unit structure determination module is configured to analyze the electrical relationship between the photovoltaic string and the upstream string inverter or DC combiner box based on the electrical number, and determine the power generation analysis unit based on the digital model of the power generation unit corresponding to the photovoltaic string and its upstream equipment. The data processing module is configured to align the acquired data with the time, preprocess the data, and verify the preprocessed data to prepare effective model input data for the performance analysis calculation model. The string performance analysis module is configured to start the constructed performance analysis calculation model according to the set analysis calculation start time, and input the valid model input data of the power generation analysis unit to determine the power generation performance evaluation index corresponding to each power generation analysis unit. The business optimization application module is configured to select the optimal power generation business string structure based on the calculation of power generation performance evaluation indicators, and to decide on the operation and maintenance processing scheme matching different evaluation indicator levels. The string performance analysis module calculates the average capacity utilization rate and standard deviation of different power generation analysis units above the multi-photovoltaic string based on the performance analysis calculation model as power generation performance evaluation indicators. The business optimization application module includes a string structure optimization unit, which is configured to combine the power generation performance evaluation index with the set optimization standard to compare and analyze the results and eliminate low-performance photovoltaic strings, thereby realizing the optimization and marking of the string structure for power generation operations. If the calculated standard deviation does not meet the optimization criteria, the standard deviation is marked on the string with the lowest daily average capacity utilization, and then that string is removed. The calculation is then performed again based on the remaining strings in the power generation analysis unit. The process is repeated, comparing the results with the set optimization criteria and marking the standard deviation on the string with the lowest daily average capacity utilization in this calculation. The loop continues until the final standard deviation meets the optimization criteria, at which point the loop stops.

2. The system according to claim 1, characterized in that, The data processing module includes a data time alignment and partitioning unit, which is configured to perform time alignment processing on the collected equipment time-series production status data using a data information model with the same adjacent time. The unit also selects power generation status data objects for storage based on preset effective data time periods and analysis nodes.

3. The system according to claim 2, characterized in that, The data processing module includes a data preprocessing unit, used to perform data cleaning and data completion processing on the data object; the data completion processing includes using interpolation to supplement missing data values.

4. The system according to claim 3, characterized in that, The data processing module includes a validity check unit, which checks the preprocessed data according to the following logic to prepare valid model input data for the performance analysis calculation model: Determine whether the data volume of each photovoltaic string corresponding to different analysis nodes meets the set data volume conditions, select the data of photovoltaic strings that do not meet the data volume conditions, and use the data of the remaining photovoltaic strings as standby power generation status data. Determine whether the number of photovoltaic strings under the corresponding power generation analysis unit of the standby power generation status data meets the set string quantity condition, select the power generation analysis unit that meets the string quantity condition, and use its standby power generation status data as valid model input data.

5. The system according to claim 1 or 4, characterized in that, In the process of calculating the power generation performance evaluation index of different power generation analysis units, the string performance analysis module first sets the start time of the analysis model. The time when the irradiance at the location of the photovoltaic power station is greater than the set condition for the last time of the day is set as the calculation start time point of the daily performance analysis calculation model.

6. The system according to claim 1, characterized in that, The performance analysis calculation model adopts the following logic: ， ， ， in, It is the standard deviation of the daily average capacity utilization rate of the photovoltaic strings corresponding to the current power generation analysis unit. m It is the number of strings under the power generation analysis unit. Indicates the first generation in the power generation analysis unit i The average daily capacity utilization of each string group. This is the result of averaging the daily capacity utilization rate of all strings under the power generation analysis unit again. This represents the average daily capacity utilization rate for each string. n This represents the number of string capacity utilization rates calculated by a single analysis node each day. Indicates the number of days j Capacity utilization data for each analysis node. This represents the capacity utilization rate of a single analysis node in a photovoltaic string. VI The DC-side power of the photovoltaic string represents the power of a single analysis node. Wp This represents the installed capacity of the photovoltaic string.

7. The system according to claim 1, characterized in that, The business optimization application module includes an operation and maintenance solution decision unit, which is configured to determine different levels of performance evaluation grades based on set performance index division rules, corresponding to different degrees of performance degradation, and different forms of early warning and display for different performance evaluation grades, matching operation and maintenance handling solutions with different urgency levels.

8. The system according to claim 7, characterized in that, The system also includes an operation and maintenance effect analysis module, which is configured to mark and record operation and maintenance information of photovoltaic strings that have adopted operation and maintenance schemes, update the data tags of the operation and maintenance strings, compare the power generation performance evaluation indicators of the power generation analysis unit before and after operation and maintenance, analyze the optimization effect of the operation and maintenance scheme, and provide data support for the operation and maintenance scheme decision-making unit.

9. A method for insightful analysis of photovoltaic string performance degradation and low-performance operation based on big data, characterized in that, The method is applied to the system according to any one of claims 1 to 8, and the method includes: The data collection module is used to obtain the electrical serial numbers of all string inverters or DC combiner boxes in the photovoltaic power station to be analyzed, as well as the photovoltaic string production status data corresponding to all measurement points in the field. The electrical relationship between the photovoltaic string and the upstream string inverter or DC combiner box is analyzed based on the electrical numbering, and the power generation analysis unit is determined based on the digital model of the power generation unit corresponding to the photovoltaic string and its upstream equipment. After aligning the time of the acquired electrical number data, the data is preprocessed and verified based on the preprocessed data to prepare effective model input data for the performance analysis calculation model; The performance analysis and calculation model is started according to the set analysis and calculation start time. The effective model input data of the site to be analyzed is input to determine the power generation performance evaluation index corresponding to each power generation analysis unit. Based on the performance analysis and calculation model, the average capacity utilization rate and standard deviation of different power generation analysis units above the multi-photovoltaic string are calculated as power generation performance evaluation index. The calculation-based power generation performance evaluation index is used to optimize and label the power generation business string structure, and to determine the operation and maintenance handling scheme matching different evaluation index levels. By combining the power generation performance evaluation index with the set optimization standards, low-performance photovoltaic strings are eliminated, thus realizing the optimization and labeling of the power generation operation string structure. If the calculated standard deviation does not meet the optimization criteria, the standard deviation is marked on the string with the lowest daily average capacity utilization, and then that string is removed. The calculation is then performed again based on the remaining strings in the power generation analysis unit. The process is repeated, comparing the results with the set optimization criteria and marking the standard deviation on the string with the lowest daily average capacity utilization in this calculation. The loop continues until the final standard deviation meets the optimization criteria, at which point the loop stops.

10. A storage medium, characterized in that, The storage medium stores program code that can implement the method as described in claim 9.

Images