Distributed photovoltaic power generation amount prediction method and device, storage medium and electronic equipment

Abstract

This invention discloses a method, apparatus, storage medium, and electronic device for predicting distributed photovoltaic (PV) power generation. The method includes: acquiring a real-time operational dataset of the distributed PV system for the current period, wherein the real-time operational dataset includes real-time irradiance data, real-time ambient temperature data, and real-time module efficiency data; obtaining an initial power prediction result for the distributed PV system during the prediction period based on the real-time irradiance data; obtaining a quantified efficiency deviation value based on the real-time ambient temperature data, real-time module efficiency data, and a target baseline efficiency curve; and correcting the initial power prediction result according to the quantified efficiency deviation value to obtain the predicted power generation result for the distributed PV system during the prediction period. This invention solves the technical problem of inaccurate power generation prediction in distributed PV systems due to changes in environmental conditions in related technologies.

Description

Technical Field

[0001] This invention relates to the field of new energy power generation technology, and more specifically, to a method, apparatus, storage medium, and electronic equipment for predicting distributed photovoltaic power generation. Background Technology

[0002] With the increasing global demand for clean energy, distributed photovoltaic (PV) systems are playing an increasingly important role in the energy market due to their environmental friendliness and energy efficiency. Especially in power distribution networks, the penetration rate of distributed PV systems is continuously increasing due to their installation flexibility and localized energy consumption advantages. Distributed PV power generation prediction technologies primarily rely on meteorological data, particularly solar irradiance, to predict power generation. However, methods that rely solely on solar irradiance for power generation prediction, such as those commonly focusing on the direct correlation between irradiance and power output, do not consider all factors comprehensively, leading to systematic biases in the prediction results and decreased prediction accuracy.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, storage medium, and electronic device for predicting distributed photovoltaic power generation, in order to at least solve the technical problem of inaccurate power generation prediction in distributed photovoltaic systems due to changes in environmental conditions.

[0005] According to one aspect of the present invention, a method for predicting distributed photovoltaic (PV) power generation is provided, comprising: acquiring a real-time operating dataset of a distributed PV system in the current time period, wherein the real-time operating dataset includes real-time irradiance data, real-time ambient temperature data, and real-time module efficiency data, the real-time module efficiency data being used to indicate the actual conversion efficiency of PV modules in the distributed PV system under the current environmental conditions; obtaining an initial power prediction result for the distributed PV system in the prediction period based on the real-time irradiance data, wherein the prediction period is a predetermined duration after the current time period; obtaining an efficiency deviation quantification value based on the real-time ambient temperature data, the real-time module efficiency data, and a target baseline efficiency curve, wherein the target baseline efficiency curve represents a pre-set baseline relationship curve between PV module efficiency and ambient temperature that matches the weather pattern of the current time period; and correcting the initial power prediction result according to the efficiency deviation quantification value to obtain a power generation prediction result for the distributed PV system in the prediction period.

[0006] According to another aspect of the present invention, a distributed photovoltaic power generation prediction device is also provided, comprising: a data acquisition module, configured to acquire a real-time operating dataset of a distributed photovoltaic system in the current time period, wherein the real-time operating dataset includes real-time irradiance data, real-time ambient temperature data, and real-time module efficiency data, the real-time module efficiency data being used to indicate the actual conversion efficiency of the photovoltaic modules in the distributed photovoltaic system under the current environmental conditions; an initial power prediction module, configured to obtain an initial power prediction result of the distributed photovoltaic system in the prediction period based on the real-time irradiance data, wherein the prediction period is a period of predetermined duration after the current time period; a deviation quantification module, configured to obtain an efficiency deviation quantification value based on the real-time ambient temperature data, the real-time module efficiency data, and a target benchmark efficiency curve, wherein the target benchmark efficiency curve represents a pre-set benchmark relationship curve between the photovoltaic module efficiency and the ambient temperature that matches the weather pattern of the current time period; and a power generation prediction module, configured to correct the initial power prediction result according to the efficiency deviation quantification value, thereby obtaining a power generation prediction result of the distributed photovoltaic system in the prediction period.

[0007] According to another aspect of the present invention, a non-volatile storage medium is also provided, which stores a plurality of instructions adapted for a distributed photovoltaic power generation prediction method to be loaded by a processor and executed at any one of them.

[0008] According to another aspect of the present invention, an electronic device is also provided, including one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement any one of the distributed photovoltaic power generation prediction methods.

[0009] According to another aspect of the present invention, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of any one of the distributed photovoltaic power generation prediction methods.

[0010] In this embodiment of the invention, a real-time operational dataset of the distributed photovoltaic system for the current period is obtained. This dataset includes real-time irradiance data, real-time ambient temperature data, and real-time module efficiency data. The real-time module efficiency data indicates the actual conversion efficiency of the photovoltaic modules in the distributed photovoltaic system under the current environmental conditions. Based on the real-time irradiance data, an initial power prediction result for the distributed photovoltaic system during a predicted period is obtained, where the predicted period is a predetermined duration following the current period. Based on the real-time ambient temperature data, real-time module efficiency data, and a target baseline efficiency curve, a quantitative value for efficiency deviation is obtained. The target baseline efficiency curve represents a pre-set efficiency curve relative to the current weather conditions. The system obtains a baseline curve showing the relationship between photovoltaic module efficiency and ambient temperature. Based on the quantified efficiency deviation, the initial power prediction results are corrected to obtain the power generation prediction results of the distributed photovoltaic system during the prediction period. This achieves the goal of dynamically correcting the initial power prediction results by collecting the operating data of the photovoltaic system in real time and combining it with the baseline efficiency curve under historical weather patterns. This improves the accuracy of power generation prediction for distributed photovoltaic systems, ensures accurate estimation of photovoltaic system power generation under different environmental conditions, and provides a reliable basis for grid dispatch and power trading. Furthermore, it solves the technical problem of inaccurate power generation prediction in distributed photovoltaic systems due to changes in environmental conditions. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0012] Figure 1 This is a flowchart of a distributed photovoltaic power generation prediction method according to an embodiment of the present invention;

[0013] Figure 2 This is a flowchart of an optional distributed photovoltaic power generation prediction method according to an embodiment of the present invention;

[0014] Figure 3 This is a schematic diagram of a distributed photovoltaic power generation prediction device according to an embodiment of the present invention. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0016] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] According to an embodiment of the present invention, a method embodiment for predicting distributed photovoltaic power generation is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0018] Figure 1 This is a flowchart of a distributed photovoltaic power generation prediction method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0019] Step S102: Obtain the real-time operation dataset of the distributed photovoltaic system in the current time period. The real-time operation dataset includes real-time irradiance data, real-time ambient temperature data, and real-time component efficiency data. The real-time component efficiency data is used to indicate the actual conversion efficiency of the photovoltaic components in the distributed photovoltaic system under the current environmental conditions.

[0020] In this step, real-time data is collected for the distributed photovoltaic (PV) system. This dataset includes real-time irradiance data, real-time ambient temperature data, and real-time module efficiency data. Real-time irradiance data reflects the current solar radiation intensity and is a direct factor affecting PV power generation. Real-time ambient temperature data records the temperature of the environment surrounding the PV modules, as the conversion efficiency of PV modules changes with ambient temperature. Real-time module efficiency data provides the actual conversion efficiency of the PV modules under current environmental conditions, serving as direct feedback on the operating status of the distributed PV system. Real-time acquisition of this data is crucial for accurately predicting future power generation.

[0021] As an optional embodiment, real-time irradiance data, real-time ambient temperature data, and real-time module efficiency data of a distributed photovoltaic system can be obtained in the following manner: At the distributed photovoltaic system site, select high-precision, stable irradiance sensors, ambient temperature sensors, and module efficiency monitoring equipment; install the irradiance sensors in an unobstructed location near the photovoltaic module array to ensure accurate measurement of the solar irradiance incident on the modules; install the ambient temperature sensors in a well-ventilated location, avoiding direct sunlight and heat sources, to obtain the actual ambient temperature; connect the module efficiency monitoring equipment to the photovoltaic modules to monitor their conversion efficiency in real time; configure data acquisition modules for each sensor and monitoring equipment, and set appropriate data acquisition frequencies. Simultaneously, a unified data storage format is established to facilitate subsequent data processing; wireless communication is used to transmit the collected data to the data processing center in real time; the data is encrypted before transmission to ensure data security; a data receiving server is set up in the data processing center to receive data transmitted from sensors and monitoring equipment in real time and perform preliminary verification to check whether the data is complete and reasonable; after receiving the data, the data processing center integrates the irradiance, ambient temperature, and component efficiency data according to the timestamp; the irradiance, ambient temperature, and component efficiency data at the same time point are combined into a single record to form a real-time operating dataset; the integrated data is then verified again to check whether the logical relationships between the data are reasonable.

[0022] In this embodiment, high-precision and stable irradiance sensors, ambient temperature sensors, and module efficiency monitoring equipment are selected and installed in reasonable locations to ensure accurate measurement of solar irradiance incident on the photovoltaic modules, acquisition of real-time ambient temperature, and real-time monitoring of module conversion efficiency. Data acquisition modules are configured for each device, with appropriate acquisition frequencies and unified storage formats set. This ensures data acquisition at a reasonable pace while facilitating subsequent processing and improving data processing efficiency. Real-time data transmission via wireless communication is employed and encrypted to ensure secure and rapid data delivery to the data processing center. Preliminary verification by the data receiving server promptly identifies problems, preventing erroneous data from entering subsequent processes. The data processing center integrates data by timestamp to form a real-time running dataset and performs further verification to ensure data integrity and logical coherence.

[0023] Step S104: Based on real-time irradiance data, obtain the initial power prediction result of the distributed photovoltaic system during the prediction period, where the prediction period is a predetermined duration after the current period.

[0024] This step, based on current real-time irradiance data and using a baseline power curve established from historical data, predicts the photovoltaic power generation over a predetermined future period. This prediction first identifies the baseline power curve that best matches the current real-time irradiance data. This curve reveals the expected power generation of the photovoltaic system under similar irradiance conditions. This step lays the foundation for subsequent accurate predictions.

[0025] In one optional embodiment, the initial power prediction result of the distributed photovoltaic system during the prediction period is obtained based on real-time irradiance data, including: determining a target weather pattern that matches the real-time irradiance data from multiple weather patterns; determining a target reference power curve that matches the target weather pattern from a reference mapping relationship library, wherein the reference mapping relationship library includes multiple reference power curves, and the multiple reference power curves correspond one-to-one with multiple weather patterns; and obtaining the initial power prediction result based on the target reference power curve.

[0026] In this embodiment, based on real-time irradiance data, a matching weather pattern can be determined, and then a corresponding benchmark power curve can be selected from the benchmark mapping relationship library to generate an initial power prediction result. This process uses pattern recognition to map real-time data to various weather patterns derived from historical data analysis, finding the benchmark power curve that best matches the current environmental conditions. The benchmark mapping relationship library contains multiple benchmark power curves, each corresponding to a weather pattern, ensuring the accuracy and relevance of the prediction results. By matching with real-time irradiance data, the prediction model can be dynamically adjusted to reflect the photovoltaic system's power generation performance under real environmental conditions, improving the real-time performance and accuracy of the prediction. In practice, this approach can reduce prediction bias, enhance adaptability to complex environmental changes, and provide reliable data support for the operation management and power dispatch of distributed photovoltaic power plants.

[0027] As an optional implementation, an initial power prediction sequence can be obtained by matching the corresponding benchmark power curve from a benchmark mapping database based on real-time irradiance data. Specifically, the real-time irradiance data can be preprocessed; the data integrity and rationality can be checked, and obvious outliers can be removed; for missing data, linear interpolation can be used to supplement it based on data from adjacent time points; the irradiance distribution range under different weather patterns in the benchmark mapping database can be analyzed, and the irradiance can be divided into several intervals, such as low irradiance interval, medium irradiance interval, and high irradiance interval. The interval division should reflect the effect of different irradiance levels on power generation. The impact of differences is assessed; the preprocessed real-time irradiance value is compared with the divided irradiance intervals to determine its interval; the benchmark power curve corresponding to the weather pattern of the irradiance interval is found in the benchmark mapping relationship library; if the real-time irradiance is near the interval boundary, such as within the predetermined neighborhood of the interval boundary, the benchmark power curves of adjacent intervals are comprehensively considered and fused according to a certain weight to obtain an accurate matching curve; based on the time axis, according to the matched benchmark power curve and combined with real-time time information, the corresponding power value is extracted from the benchmark power curve at the same time interval to form an initial power prediction sequence.

[0028] In this embodiment, a preliminary power prediction sequence can be generated by accurately matching the irradiance with the baseline power curve. Preprocessing the real-time irradiance data ensures its integrity and accuracy. Outliers are removed, and missing data is supplemented using linear interpolation, avoiding prediction errors caused by data issues and ensuring data quality. Dividing the irradiance value into multiple intervals effectively reflects the differences in the impact of different irradiance levels on power generation, accurately reflecting the power generation characteristics of the photovoltaic system under different weather conditions. By comparing the preprocessed real-time irradiance with each irradiance interval, the interval to which it belongs can be accurately determined, and the corresponding baseline power curve can be matched from the baseline mapping database. If the real-time irradiance value is at the boundary of an interval, weighted fusion is performed based on the power curves of adjacent intervals, further improving the matching accuracy. The corresponding power value is extracted from the baseline power curve according to the time axis to generate the initial power prediction sequence. This not only ensures the accuracy of power prediction but also improves the model's adaptability, effectively responding to changes under different weather patterns and reducing errors caused by environmental changes.

[0029] In an optional embodiment, before determining the target reference power curve matching the target weather pattern from the reference mapping relationship library, the method further includes: performing cluster analysis based on the historical irradiance data and historical power generation data corresponding to the distributed photovoltaic system in multiple historical periods to obtain clustering results, wherein the clustering results are used to indicate multiple weather patterns and the historical power generation curves corresponding to each of the multiple weather patterns, and the multiple weather patterns correspond to different irradiance ranges; and obtaining multiple reference power curves corresponding to the multiple weather patterns based on the historical power generation curves corresponding to each of the multiple weather patterns.

[0030] In this embodiment, cluster analysis is performed based on historical irradiance data and historical power generation data corresponding to the distributed photovoltaic system in multiple historical periods to obtain clustering results. These results indicate various weather patterns and their corresponding historical power generation curves, with each weather pattern corresponding to a different irradiance range. This step forms the basis for constructing a benchmark mapping database. Through in-depth analysis of historical data, the power generation characteristics of the photovoltaic system under different weather patterns can be identified and classified, providing a basis for subsequent benchmark power curve matching. Based on these classified historical power generation curves, further data processing and analysis generate multiple benchmark power curves corresponding to various weather patterns. These benchmark power curves represent the average power generation performance of the photovoltaic system under specific weather patterns, providing an important reference standard for real-time power generation prediction. Through this series of data processing steps, the weather pattern of the current photovoltaic system operation can be identified more accurately, and the most matching benchmark power curve can be selected, thereby improving the accuracy of power generation prediction and effectively ensuring the stability of the power grid operation and the economic benefits of distributed photovoltaic power stations. In other embodiments, the clustering analysis methods and parameters can be adjusted and optimized according to actual needs to better adapt to different regions and different types of photovoltaic systems, ensuring the universality and accuracy of the prediction model.

[0031] In one optional embodiment, multiple reference power curves corresponding to various weather patterns are obtained based on their respective historical power generation curves. This includes: when there are multiple historical power generation curves for any given weather pattern, obtaining the reference power curve for any given weather pattern is achieved by: constructing a distance matrix and determining the optimal path search range based on preset constraints. The distance matrix indicates the degree of difference between the multiple historical power generation curves at various time points, the preset constraints indicate the maximum permissible slope on the normalized path, and the optimal path search range indicates the relative slope between two historical power generation curves when calculating the normalized distance. The allowable offset range of the connecting line at the points on the time axis and power axis; within the optimal path search range, calculate the cumulative normalized distance between multiple historical curves of any weather pattern, where the cumulative normalized distance is used to indicate the cumulative value of the difference between the curves calculated along the optimal normalized path after normalizing multiple historical power generation curves of any weather pattern to the same length through a dynamic time normalization algorithm; based on the cumulative normalized distance between multiple historical curves, generate a reference power curve corresponding to any weather pattern; based on the historical power generation curves corresponding to each of the multiple weather patterns, obtain multiple reference power curves by adopting the method of obtaining the reference power curve corresponding to any weather pattern.

[0032] In this embodiment, the normalized path indicates the optimal matching sequence formed by nonlinearly aligning historical power generation curves of different lengths in the time domain. The normalized distance indicates the cumulative sum of squares of the power differences between corresponding points under the optimal matching sequence, characterizing the overall morphological similarity between curves. For historical power generation curves under different weather patterns, a distance matrix is ​​constructed using a dynamic time normalization algorithm. The optimal path search range is determined based on preset constraints, which indicate the maximum allowable slope on the normalized path. The optimal path search range indicates the allowable offset range of the line connecting corresponding points between two historical power generation curves on the time and power axes when calculating the normalized distance. Within the determined search range, the cumulative normalized distance between multiple historical curves is calculated. The cumulative normalized distance indicates the cumulative value of the differences between curves calculated along the optimal normalized path after normalization to the same length. Based on the calculated cumulative normalized distance, a reference power curve is generated for each weather pattern. In this way, multiple reference power curves are obtained for the historical power generation curves corresponding to various weather patterns. This process effectively quantifies the similarity between different historical power generation curves, enabling the extracted benchmark power curves to represent typical power generation performance under specific weather patterns. This improves the accuracy and reliability of distributed photovoltaic power generation prediction based on big data analysis. By constructing a benchmark mapping relationship library, the relationship between complex environmental factors and power generation is systematized, providing an accurate reference benchmark for real-time prediction. This ensures that the prediction results can adapt to various actual weather conditions, enhancing the practicality and stability of the prediction method.

[0033] As an optional implementation, cluster analysis is performed on historical irradiance data and historical power generation data. Using the K-means clustering algorithm, the data is divided into different weather patterns such as sunny, cloudy, rainy, and foggy days based on the characteristics of irradiance and power generation. Multiple meteorological indicators are considered to comprehensively determine the weather pattern category, ensuring that the classification accurately reflects different environmental conditions. For each weather pattern, a dynamic time warping algorithm is used to extract feature curves from all historical power generation curves belonging to that weather pattern. A distance matrix is ​​constructed, and the search range of the optimal path is determined based on preset constraints. Within the search range, the cumulative warped distance between multiple historical curves is calculated. The historical curve with the smallest cumulative warped distance is selected, or multiple historical curves are fitted to generate feature curves. For each weather pattern, the relationship between historical ambient temperature data and historical component efficiency data is analyzed. A regression analysis method is used to establish a mathematical model of ambient temperature and component efficiency, determining the benchmark correspondence between ambient temperature and component efficiency under different weather patterns, forming a benchmark efficiency curve. The benchmark power curve and benchmark efficiency curve corresponding to each weather pattern are stored in a database to construct a benchmark mapping relationship library.

[0034] In this embodiment, from the perspective of prediction accuracy, the precise clustering of historical irradiance and power generation data into weather patterns can meticulously capture the unique characteristics of photovoltaic power generation under different environmental conditions. Based on this, the dynamic time warping algorithm is used to extract the baseline power curve for each weather pattern, and regression analysis is used to determine the baseline efficiency curve. This accurately depicts the intrinsic relationship between power generation and ambient temperature and component efficiency under different weather patterns, reducing prediction bias caused by uncoupled ambient temperature effects and improving the accuracy of power generation prediction. The comprehensive consideration of multiple meteorological indicators in weather pattern classification ensures that the classification results accurately reflect complex and changing environmental conditions. This allows the generated baseline mapping relationship library to adapt to various practical scenarios, enhancing the predictive method's adaptability to different environments. Regarding data processing and utilization efficiency, after systematic preprocessing of historical data, scientific clustering, warping, and regression analysis methods are used to fully explore the data value. The processed baseline power curve and baseline efficiency curve are stored in the database for easy subsequent rapid querying and matching, improving data utilization efficiency and providing efficient data support for real-time prediction. The accurate and reliable baseline mapping relationship library provides a solid foundation for power generation prediction, helps maintain the stability of the power grid operation, and ensures the economic benefits of distributed photovoltaic power stations.

[0035] As an optional implementation, historical datasets of distributed photovoltaic (PV) systems can be obtained, but are not limited to, through the following methods: utilizing PV system hardware monitoring equipment to directly acquire historical data from the PV system; the PV system hardware monitoring equipment records indicators such as irradiance, ambient temperature, system efficiency, and power generation of the PV modules; collaborating with meteorological data providers to acquire historical meteorological data of the surrounding area, including irradiance, temperature, humidity, and wind speed, to assist in assessing the external environmental impact of PV system power generation; developing detailed data collection plans based on the characteristics of different weather patterns; for example, setting different collection frequencies for different weather conditions such as sunny, cloudy, rainy, and foggy days; appropriately increasing the collection frequency for periods of rapid weather change to capture rapid changes in irradiance and ambient temperature; and ensuring the accuracy of the timestamps on the collected data for subsequent time-series analysis. The process involves: establishing a data storage system using distributed databases or cloud storage technology to store large amounts of historical data; classifying and storing the collected data, indexing it according to weather patterns, time, etc., to facilitate subsequent queries and analysis; establishing a data quality monitoring mechanism to verify the collected data in real time, removing abnormal data to ensure data reliability; preprocessing the historical data after collection, including data cleaning to remove missing values, outliers, and noisy data; data normalization to unify data of different dimensions into the same range to facilitate subsequent pattern recognition and time-series processing; and supplementing missing data points using appropriate interpolation methods to ensure data integrity. The historical dataset includes historical irradiance data, historical power generation data, historical ambient temperature data, and historical component efficiency data collected under various weather patterns.

[0036] In this embodiment, data is directly acquired from the photovoltaic system using hardware monitoring equipment, accurately recording key indicators such as irradiance, ambient temperature, system efficiency, and power generation of the photovoltaic module, providing a real and direct information foundation for subsequent analysis. Collaboration with a meteorological data provider platform acquires historical meteorological data from the surrounding area, covering irradiance, temperature, humidity, and wind speed, which helps assess the impact of the external environment on the photovoltaic system's power generation, making the consideration of factors affecting power generation more comprehensive. Furthermore, a detailed data acquisition plan is developed based on the characteristics of different weather patterns, setting different acquisition frequencies for different weather conditions, increasing the frequency when the weather changes rapidly, capturing rapid changes in irradiance and ambient temperature, and ensuring accurate data timestamps, providing accurate and orderly data for subsequent time-series processing. A data storage system is established, employing distributed database or cloud storage technology, capable of storing large amounts of historical data. The data is categorized and indexed by weather pattern, time, etc., facilitating subsequent querying and analysis, and improving data utilization efficiency. Simultaneously, a data quality monitoring mechanism is established to verify the collected data in real time and remove abnormal data, ensuring data reliability.

[0037] Step S106: Based on real-time ambient temperature data, real-time module efficiency data, and target baseline efficiency curve, obtain the efficiency deviation quantification value. The target baseline efficiency curve represents a pre-set baseline relationship curve between photovoltaic module efficiency and ambient temperature that matches the weather pattern of the current period.

[0038] This step involves quantifying the efficiency deviation, aiming to assess the difference between the photovoltaic module efficiency under real-time operating conditions and the historical baseline efficiency curve. By comparing real-time ambient temperature data and real-time module efficiency data with the theoretical values ​​from the target baseline efficiency curve under the same environmental conditions, a quantified value of the efficiency deviation can be calculated. This quantified value reflects the degree of influence of environmental factors on the actual conversion efficiency of the photovoltaic module and is a key parameter for subsequent correction of prediction results.

[0039] In one optional embodiment, based on real-time ambient temperature data, real-time module efficiency data, and a target baseline efficiency curve, an efficiency deviation quantification value is obtained, including: determining a baseline module efficiency value that matches the initial power prediction result in time from the target baseline efficiency curve; performing a difference calculation on the real-time module efficiency data and the baseline module efficiency value to obtain an initial efficiency deviation; converting the real-time ambient temperature data into a corresponding theoretical efficiency impact value based on a correlation model between ambient temperature and module efficiency, wherein the correlation model between ambient temperature and module efficiency is used to indicate a specific proportional relationship in which the photovoltaic module efficiency decreases as the ambient temperature increases, and is used to reflect the change law of the conversion efficiency of the photovoltaic module under different ambient temperatures; and weightedly fusing the initial efficiency deviation and the theoretical efficiency impact value to obtain the efficiency deviation quantification value.

[0040] In this embodiment, the process of obtaining the quantified value of efficiency deviation based on real-time ambient temperature data, real-time component efficiency data, and the target baseline efficiency curve includes determining a baseline component efficiency value that matches the initial power prediction result in time from the target baseline efficiency curve. The initial efficiency deviation is obtained by calculating the difference between the real-time component efficiency data and the baseline component efficiency value. This step directly reflects the difference between the current component efficiency and the historical baseline efficiency. Next, based on the correlation model between ambient temperature and component efficiency, the real-time ambient temperature data is converted into a corresponding theoretical efficiency impact value. This model indicates a specific proportional relationship between the photovoltaic component efficiency and the decrease in efficiency as the ambient temperature increases, reflecting the change in conversion efficiency of the photovoltaic component under different ambient temperatures and quantifying the impact of ambient temperature on component efficiency. Finally, the initial efficiency deviation and the theoretical efficiency impact value are weighted and fused to obtain the quantified value of efficiency deviation. This fusion process comprehensively considers both the actual component efficiency and the theoretical efficiency impact, making the quantified value closer to the efficiency deviation under real-time operating conditions, thereby improving the accuracy of power generation prediction. Through the above steps, the system in this embodiment can dynamically adapt to changes in ambient temperature, effectively correct component efficiency prediction, and ensure the accuracy and real-time performance of distributed photovoltaic power generation prediction.

[0041] In an optional embodiment, before obtaining the efficiency deviation quantification value based on real-time ambient temperature data, real-time component efficiency data, and target baseline efficiency curve, the method further includes: determining a target weather pattern that matches the real-time irradiance data from multiple weather patterns; and determining a target baseline efficiency curve that matches the target weather pattern from a baseline mapping relationship library, wherein the baseline mapping relationship library includes multiple baseline efficiency curves, and the multiple baseline efficiency curves correspond one-to-one with multiple weather patterns.

[0042] In this embodiment, a target weather pattern matching the real-time irradiance data is determined from multiple weather patterns, and then a target baseline efficiency curve corresponding to the target weather pattern is selected from the baseline mapping relationship library. This process ensures accurate matching between real-time environmental parameters and historical data benchmarks, which is a key prerequisite for generating efficiency deviation quantification values. By carefully comparing the characteristics of real-time irradiance data with the boundaries of multiple weather patterns defined in the historical dataset, the weather pattern that best matches the current environmental conditions can be automatically identified. Subsequently, the baseline efficiency curve associated with the identified weather pattern is retrieved from the baseline mapping relationship library. This step provides an important reference baseline for the calculation of efficiency deviation quantification. In this series of operations, selecting the most appropriate historical data pattern based on the characteristics of real-time data effectively reduces the negative impact of weather pattern misjudgment on the prediction results, improves the calculation accuracy of efficiency deviation quantification values, and thus enhances the accuracy and reliability of the entire distributed photovoltaic power generation prediction method. Of course, the weather pattern identification strategy can also be dynamically adjusted according to the changing trends of real-time data to further enhance the flexibility and response speed of the prediction method. Through intelligent identification and dynamic adjustment, it is possible to better adapt to the complex and ever-changing actual environment, ensuring that the prediction results are always close to reality and improving the scientific nature of distributed photovoltaic power station operation and maintenance decisions.

[0043] Optionally, statistical analysis can be performed on the historical ambient temperature data and historical component efficiency data of the distributed photovoltaic system for each of the multiple historical periods to obtain the correlation between photovoltaic component efficiency and ambient temperature for each of the various weather patterns. Using machine learning or statistical modeling methods, based on the historical ambient temperature data and historical component efficiency data, a curve of photovoltaic component efficiency changing with ambient temperature under each weather pattern can be fitted, i.e., multiple benchmark efficiency curves.

[0044] As an optional embodiment, real-time ambient temperature data and real-time component efficiency data can be compared with a matched benchmark efficiency curve to generate a quantitative value of efficiency deviation. Specifically, the real-time ambient temperature data and real-time component efficiency data can be preprocessed; data integrity can be checked and obvious outliers can be removed; for missing data, linear interpolation can be used to supplement it based on data from adjacent time points; a time benchmark for comparison can be determined; the preprocessed real-time ambient temperature data and real-time component efficiency data can be mapped to time points at the same time interval as the generated benchmark efficiency curve; benchmark component efficiency values ​​at the same time points can be extracted from the matched benchmark efficiency curve; the real-time component efficiency value can be subtracted from the benchmark component efficiency value at the corresponding time point to obtain a preliminary efficiency deviation value; considering the influence of ambient temperature on component efficiency, a correlation model between ambient temperature and component efficiency can be established based on the characteristics of photovoltaic modules; using this model, the real-time ambient temperature data can be converted into the corresponding theoretical efficiency influence value; the preliminary efficiency deviation value and the theoretical efficiency influence value can be weighted and fused to finally obtain a comprehensive efficiency deviation quantitative value.

[0045] In this embodiment, preprocessing of real-time ambient temperature data and real-time component efficiency data ensures data integrity and accuracy, removes obvious outliers, and supplements missing data, improving the reliability of subsequent comparison processes. Time alignment using the same time interval as the benchmark efficiency curve ensures the correspondence between real-time and benchmark data at the same point in time, thereby improving comparison accuracy. When generating preliminary efficiency deviation values, comparing real-time component efficiency with benchmark component efficiency clearly identifies efficiency fluctuations caused by environmental changes in the system. Establishing a correlation model between temperature and component efficiency, and converting real-time ambient temperature data into theoretical efficiency impact values, accurately describes the actual impact of ambient temperature on component efficiency. By weighted fusion of the preliminary efficiency deviation value and the theoretical efficiency impact value, a comprehensive efficiency deviation quantification value is generated, comprehensively reflecting the dynamic changes in ambient temperature and component efficiency, effectively eliminating efficiency prediction errors caused by temperature fluctuations, and ensuring the accuracy of power prediction. This dynamic correction improves the prediction accuracy and operational stability of the distributed photovoltaic system.

[0046] Step S108: Based on the efficiency deviation quantification value, the initial power prediction result is corrected to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period.

[0047] This step dynamically corrects the initial power forecast based on the quantified efficiency deviation value to reflect the actual impact of environmental changes on power generation. This means that if a rise in real-time ambient temperature leads to a decrease in module efficiency, or if module efficiency deviates from the baseline efficiency curve due to other factors, the forecast will be adjusted accordingly to more closely approximate the actual situation. The corrected result is the final power generation forecast, which combines real-time data and historical benchmarks, improving the accuracy and reliability of the forecast.

[0048] In one optional embodiment, the initial power prediction result is corrected based on the efficiency deviation quantization value to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period. This includes: correcting the initial power prediction result based on the efficiency deviation quantization value in the following manner to obtain the power generation prediction result:

[0049] ;in, This indicates the power generation forecast result. This indicates the initial power prediction result. This represents the quantified value of efficiency deviation. This represents the system characteristic coefficients obtained by calibration based on historical datasets of distributed photovoltaic systems.

[0050] In this embodiment, the initial power prediction result is corrected based on the efficiency deviation quantification value to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period. This correction process is implemented through the formula in this embodiment, wherein... This indicates the final power generation forecast. This represents the initial power prediction result obtained based on real-time irradiance data. The efficiency deviation is quantified by comparing real-time ambient temperature data and real-time component efficiency data with a baseline efficiency curve. 'k' represents the system characteristic coefficient, calibrated based on historical datasets of distributed photovoltaic systems, used to quantify the impact of ambient temperature on component efficiency. This correction method dynamically adjusts the initial power prediction results, fully considering the influence of actual ambient temperature and component efficiency on power generation, thus improving prediction accuracy. In implementation, this technical solution effectively reduces prediction deviations caused by temperature changes or extreme weather conditions, ensuring that prediction results are closer to actual power generation, thereby providing more reliable data support for power dispatch and operation and maintenance management. In other embodiments, the calibration of the system characteristic coefficient 'k' can also be combined with other influencing factors, such as wind speed and humidity, to further improve the complexity and accuracy of the prediction model.

[0051] It should be noted that, based on the quantified value of efficiency deviation, the initial power prediction sequence is dynamically corrected to output the final power generation prediction result; the rationality of the numerical range of the quantified value of efficiency deviation is checked, and if a value exceeds the normal physical range, it is corrected according to the distribution pattern of historical data to adjust it to a reasonable range; the time granularity of the correction is determined to be consistent with the time interval of the initial power prediction sequence; then, a correction model is established, considering the degree of influence of efficiency deviation on power, and a system characteristic coefficient of efficiency deviation and power correction is set according to the characteristics of photovoltaic system, and the power generation prediction result is obtained through the method and formula of the above embodiment. In this embodiment, by checking and correcting the quantified value of efficiency deviation, it is ensured that the value remains within a reasonable physical range during the correction process, thereby avoiding inaccurate corrections due to data anomalies. By comparing with historical data distribution patterns, outliers can be effectively handled, ensuring the stability and reliability of the correction process. A correction granularity consistent with the time interval of the initial power prediction sequence is determined to ensure that the corrected power value is consistent with the original predicted value in time, ensuring the temporal continuity and accuracy of the prediction results. When establishing the correction model, a conversion coefficient between efficiency deviation and power correction is set according to the actual characteristics of the photovoltaic system, enabling the model to reasonably quantify the impact of efficiency deviation on power. After combining the quantified value of efficiency deviation with the conversion coefficient, the power correction amount corresponding to each time point is obtained. The corrected power value is added to the initial predicted value and arranged in chronological order. The final generated power generation prediction result sequence can more accurately reflect the power generation capacity of the photovoltaic system under different environmental conditions. This improves the accuracy of power generation prediction, and the corrected result can effectively eliminate errors caused by environmental fluctuations.

[0052] Optionally, but not limited to, the system characteristic coefficient k can be obtained as follows: First, select data from the historical dataset of distributed photovoltaic systems operating under the same or similar weather patterns to ensure that the relationship between efficiency deviation and power variation is established under similar conditions. Record the difference between the actual power generation corresponding to each efficiency deviation value and the predicted value based on the baseline power prediction. Based on the results of historical data analysis, establish a model relating efficiency deviation and power correction. This model describes the relationship between efficiency deviation and power correction. This model can be a linear model, in the form of power correction = k Efficiency bias: By adjusting the value of k, the model can best fit the efficiency bias and corresponding power changes in the historical dataset. A calibration method is used to determine the optimal value of k; for example, the least squares method can be used to minimize the sum of squared errors between the model's predicted values ​​and the actual historical power values. During calibration, the influence of various variables, such as photovoltaic module type, system size, and geographical location, can be further considered, as these factors may affect the value of k. After initial calibration of k, the model's predictive ability is tested using a retained validation dataset to ensure the applicability and accuracy of k under different conditions. Based on the validation results, the value of k is fine-tuned until the model's predicted results are closest to the actual power generation in the validation dataset. In practical applications, the determined k value is used to correct power predictions based on the specific parameters of the currently operating distributed photovoltaic system. Over time, the system will collect more real-time operating data, and the value of k can be updated periodically to adapt to the potential impact of system aging, environmental changes, and other factors on the relationship between efficiency bias and power correction, ensuring the continued accuracy of the prediction results. For example, through historical data analysis, it is found that when the efficiency deviation quantification value is 0.05, the actual power generation is 2% lower than the predicted value. Therefore, the value of k can be adjusted to make k... 0.05 = 0.02, thus determining that the updated value of k is 0.4.

[0053] Through the above steps S102 to S108, the real-time acquisition of photovoltaic system operation data and the combination of baseline efficiency curves under historical weather patterns can achieve the purpose of dynamic correction of the initial power prediction results. This improves the accuracy of large-scale power prediction for distributed photovoltaic systems, ensures accurate estimation of photovoltaic system power generation under different environmental conditions, and provides a reliable basis for grid dispatch and power trading. In turn, it solves the technical problem of inaccurate power generation prediction in distributed photovoltaic systems due to changes in environmental conditions.

[0054] With the energy structure shifting towards cleaner and distributed energy, distributed photovoltaic (PV) systems are seeing a continuous increase in penetration rate in power distribution networks due to their advantages such as flexible installation and local consumption. Power generation forecasting, as a supporting technology for the operation and maintenance management of distributed PV systems, grid consumption scheduling, and power trading, directly impacts the stability of grid operation and the economic benefits of power plants. Current distributed PV power generation forecasting methods generally focus on the direct correlation between irradiance and power generation, failing to effectively couple the dynamic impact mechanism of ambient temperature on PV module conversion efficiency. Since the power generation efficiency of PV modules is negatively correlated with ambient temperature, under the same irradiance conditions, differences in ambient temperature can cause significant fluctuations in the actual output power of the modules. Current methods cannot accurately correct for these physical processes, leading to systematic biases in forecast results under conditions of drastic temperature changes or extreme high temperatures, thus affecting forecast accuracy.

[0055] To address the aforementioned problems, and in conjunction with the above embodiments and optional embodiments, the present invention proposes an optional implementation method. Figure 2 This is a flowchart of an optional distributed photovoltaic power generation prediction method according to an embodiment of the present invention, such as... Figure 2 As shown, the method includes:

[0056] S1. Obtain the historical dataset of the distributed photovoltaic system. The historical dataset includes historical irradiance data, historical power generation data, historical ambient temperature data, and historical component efficiency data collected under various weather conditions.

[0057] S2. Based on historical datasets, a benchmark mapping relationship library is generated through pattern recognition and time series processing. The benchmark mapping relationship library contains benchmark power curves and associated benchmark efficiency curves for different weather patterns.

[0058] S3. Obtain the real-time operation dataset of the photovoltaic system. The real-time operation dataset includes real-time irradiance data, real-time ambient temperature data, and real-time component efficiency data.

[0059] S4. Based on real-time irradiance data, match the corresponding reference power curve from the reference mapping relationship library to obtain the initial power prediction sequence;

[0060] S5. Compare the real-time ambient temperature data and real-time component efficiency data with the matched baseline efficiency curve to generate a quantitative value of efficiency deviation.

[0061] S6. Based on the efficiency deviation quantification value, dynamically correct the initial power prediction sequence and output the final power generation prediction result.

[0062] It should be noted that the specific implementation process of steps S1 to S6 is the same as that of the aforementioned embodiments, and will not be repeated here.

[0063] In this embodiment, the method acquires historical datasets of distributed photovoltaic systems under various weather conditions, covering historical irradiance, power generation, ambient temperature, and component efficiency data, thus fully considering the impact of different weather conditions on power generation. Based on the historical datasets, a benchmark mapping relationship library is generated through pattern recognition and time-series processing, systematically sorting out and quantifying the complex relationships between environmental factors and power generation and component efficiency. Real-time operation datasets of the photovoltaic system are acquired to reflect the current state of the system, enabling predictions to closely align with actual conditions. Based on real-time irradiance data, the corresponding benchmark power curve is matched from the benchmark mapping relationship library to obtain an initial power prediction sequence, preliminarily predicting the power output under the current irradiance conditions. The method calculates the power generation status; compares real-time ambient temperature data and real-time component efficiency data with the matched benchmark efficiency curve to generate a quantitative value of efficiency deviation; since it considers the dynamic influence of ambient temperature on component efficiency, it can accurately quantify the degree of deviation between the current environment and the benchmark state; it dynamically corrects the initial power prediction sequence based on the quantitative value of efficiency deviation and outputs the final power generation prediction result; this method fully considers the dynamic influence of ambient temperature on photovoltaic module conversion efficiency, overcomes the prediction deviation problem of related technologies under drastic temperature changes or extreme high temperature conditions, and can more accurately predict distributed photovoltaic power generation in complex and ever-changing real environments, effectively improving prediction accuracy and ensuring the stability of grid operation.

[0064] This embodiment also provides a distributed photovoltaic power generation prediction device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described previously. As used below, the terms "module" and "device" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0065] According to an embodiment of the present invention, an apparatus embodiment for implementing the above-described distributed photovoltaic power generation prediction method is also provided. Figure 3 This is a schematic diagram of the structure of a distributed photovoltaic power generation prediction device according to an embodiment of the present invention, as shown below. Figure 3 As shown, the above-mentioned distributed photovoltaic power generation prediction device includes: a data acquisition module 300, an initial power prediction module 302, a deviation quantification module 304, and a power generation prediction module 306, wherein:

[0066] The data acquisition module 300 is used to acquire the real-time operating dataset of the distributed photovoltaic system in the current period. The real-time operating dataset includes real-time irradiance data, real-time ambient temperature data, and real-time component efficiency data. The real-time component efficiency data is used to indicate the actual conversion efficiency of the photovoltaic components in the distributed photovoltaic system under the current environmental conditions.

[0067] The initial power prediction module 302 is connected to the data acquisition module 300 and is used to obtain the initial power prediction result of the distributed photovoltaic system during the prediction period based on real-time irradiance data, wherein the prediction period is a period of predetermined duration after the current period.

[0068] The deviation quantification module 304 is connected to the initial power prediction module 302 and is used to obtain the efficiency deviation quantification value based on real-time ambient temperature data, real-time component efficiency data and target reference efficiency curve. The target reference efficiency curve represents a pre-set reference relationship curve between photovoltaic component efficiency and ambient temperature that matches the weather pattern of the current period.

[0069] The power generation prediction module 306 is connected to the deviation quantization module 304 and is used to correct the initial power prediction result based on the efficiency deviation quantization value to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period.

[0070] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0071] It should be noted that the data acquisition module 300, initial power prediction module 302, deviation quantification module 304, and power generation prediction module 306 mentioned above correspond to steps S102 to S108 in the embodiments. The instances and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run on a computer terminal.

[0072] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0073] The aforementioned distributed photovoltaic power generation prediction device may also include a processor and a memory. The aforementioned data acquisition module 300, initial power prediction module 302, deviation quantification module 304, power generation prediction module 306, etc., are all stored in the memory as program modules, and the processor executes the aforementioned program modules stored in the memory to realize the corresponding functions.

[0074] The processor contains a core that retrieves the corresponding program modules from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0075] According to an embodiment of this application, an embodiment of a non-volatile storage medium is also provided. Optionally, in this embodiment, the non-volatile storage medium includes a stored program, wherein, when the program runs, it controls the device where the non-volatile storage medium is located to execute any of the distributed photovoltaic power generation prediction methods described above.

[0076] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals, and the non-volatile storage medium includes stored programs.

[0077] Optionally, a program that controls the device containing the non-volatile storage medium to execute any of the above-mentioned distributed photovoltaic power generation prediction method steps during program execution.

[0078] According to an embodiment of this application, an embodiment of a processor is also provided. Optionally, in this embodiment, the processor is used to run a program, wherein the program executes any of the above-described distributed photovoltaic power generation prediction methods.

[0079] According to an embodiment of this application, an embodiment of a computer program product is also provided, which, when executed on a data processing device, is adapted to execute a program that initializes the distributed photovoltaic power generation prediction method steps described above.

[0080] This invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of any of the above-described distributed photovoltaic power generation prediction methods.

[0081] The order of the above embodiments of the present invention is merely for description and does not represent the superiority or inferiority of the embodiments.

[0082] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0083] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of modules described above can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between modules, and may be electrical or other forms.

[0084] The modules described above as separate components may or may not be physically separate. Similarly, the components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple modules. Some or all of the modules can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0085] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0086] If the aforementioned integrated modules are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable non-volatile storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a non-volatile storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned non-volatile storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0087] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for predicting distributed photovoltaic power generation, characterized in that, include: Obtain the real-time operation dataset of the distributed photovoltaic system in the current time period, wherein the real-time operation dataset includes real-time irradiance data, real-time ambient temperature data, and real-time component efficiency data, and the real-time component efficiency data is used to indicate the actual conversion efficiency of the photovoltaic components in the distributed photovoltaic system under the current environmental conditions; Based on the real-time irradiance data, the initial power prediction result of the distributed photovoltaic system in the prediction period is obtained, wherein the prediction period is a period of predetermined duration after the current period; Based on the real-time ambient temperature data, the real-time component efficiency data, and the target baseline efficiency curve, an efficiency deviation quantification value is obtained, wherein the target baseline efficiency curve represents a pre-set baseline relationship curve between photovoltaic component efficiency and ambient temperature that matches the weather pattern of the current time period. Based on the efficiency deviation quantification value, the initial power prediction result is corrected to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period.

2. The method according to claim 1, characterized in that, The process of obtaining the initial power prediction result of the distributed photovoltaic system during the prediction period based on the real-time irradiance data includes: Determine the target weather pattern that matches the real-time irradiance data from multiple weather patterns; The target reference power curve matching the target weather model is determined from the reference mapping relationship library, wherein the reference mapping relationship library includes multiple reference power curves, and the multiple reference power curves correspond one-to-one with the multiple weather models; The initial power prediction result is obtained based on the target reference power curve.

3. The method according to claim 2, characterized in that, Before determining the target reference power curve matching the target weather model from the reference mapping relation library, the method further includes: Cluster analysis is performed on the historical irradiance data and historical power generation data of the distributed photovoltaic system in multiple historical periods to obtain cluster results. The cluster results are used to indicate multiple weather patterns and the historical power generation curves corresponding to each of the multiple weather patterns. The multiple weather patterns correspond to different irradiance ranges. Based on the historical power generation curves corresponding to each of the various weather patterns, the multiple reference power curves corresponding to the various weather patterns are obtained.

4. The method according to claim 3, characterized in that, The process of obtaining multiple reference power curves corresponding to the various weather patterns based on their respective historical power generation curves includes: When there are multiple historical power generation curves for any given weather pattern, the baseline power curve for that weather pattern is obtained as follows: A distance matrix is ​​constructed, and the optimal path search range is determined based on preset constraints. The distance matrix is ​​used to indicate the degree of difference between multiple historical power generation curves at each time point, the preset constraints are used to indicate the maximum allowable slope on the normalized path, and the optimal path search range is used to indicate the allowable offset range of the line connecting corresponding points between two historical power generation curves on the time axis and the power axis when calculating the normalized distance. Within the optimal path search range, the cumulative normalized distance between multiple historical curves of any weather pattern is calculated, wherein the cumulative normalized distance is used to indicate the cumulative value of the difference between curves calculated along the optimal normalized path after normalizing multiple historical power generation curves of any weather pattern to the same length through a dynamic time normalization algorithm. Based on the cumulative normalized distance between the multiple historical curves, a baseline power curve corresponding to any weather pattern is generated. Based on the historical power generation curves corresponding to each of the various weather patterns, the multiple reference power curves are obtained by using the method of obtaining the reference power curve corresponding to any one of the weather patterns.

5. The method according to claim 1, characterized in that, The efficiency deviation quantification value is obtained based on the real-time ambient temperature data, the real-time component efficiency data, and the target baseline efficiency curve, including: Determine a reference component efficiency value that matches the time of the initial power prediction result from the target reference efficiency curve; The initial efficiency deviation is obtained by performing a difference calculation between the real-time component efficiency data and the benchmark component efficiency value. Based on the correlation model between ambient temperature and module efficiency, the real-time ambient temperature data is converted into the corresponding theoretical efficiency impact value. The correlation model between ambient temperature and module efficiency is used to indicate a specific proportional relationship in which the efficiency of photovoltaic modules decreases as the ambient temperature increases, and is used to reflect the change law of the conversion efficiency of photovoltaic modules under different ambient temperatures. The initial efficiency deviation and the theoretical efficiency impact value are weighted and fused to obtain the quantified value of the efficiency deviation.

6. The method according to claim 1, characterized in that, Before obtaining the quantified value of efficiency deviation based on the real-time ambient temperature data, the real-time component efficiency data, and the target baseline efficiency curve, the method further includes: Determine the target weather pattern that matches the real-time irradiance data from multiple weather patterns; The target baseline efficiency curve matching the target weather model is determined from the baseline mapping relationship library, wherein the baseline mapping relationship library includes multiple baseline efficiency curves, and the multiple baseline efficiency curves correspond one-to-one with the multiple weather models.

7. The method according to any one of claims 1 to 6, characterized in that, The step of correcting the initial power prediction result based on the efficiency deviation quantification value to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period includes: Based on the quantified efficiency deviation value, the initial power prediction result is corrected in the following manner to obtain the power generation prediction result: ； in, This indicates the predicted power generation result. This indicates the initial power prediction result. This represents the quantified value of the efficiency deviation. This represents the system characteristic coefficients obtained by calibration based on the historical dataset of the distributed photovoltaic system.

8. A distributed photovoltaic power generation prediction device, characterized in that, include: The data acquisition module is used to acquire the real-time operating dataset of the distributed photovoltaic system in the current time period. The real-time operating dataset includes real-time irradiance data, real-time ambient temperature data, and real-time component efficiency data. The real-time component efficiency data is used to indicate the actual conversion efficiency of the photovoltaic components in the distributed photovoltaic system under the current environmental conditions. An initial power prediction module is used to obtain the initial power prediction result of the distributed photovoltaic system during the prediction period based on the real-time irradiance data, wherein the prediction period is a period of predetermined duration after the current period; The deviation quantification module is used to obtain an efficiency deviation quantification value based on the real-time ambient temperature data, the real-time component efficiency data, and the target benchmark efficiency curve, wherein the target benchmark efficiency curve represents a pre-set benchmark relationship curve between photovoltaic component efficiency and ambient temperature that matches the weather pattern of the current time period. The power generation prediction module is used to correct the initial power prediction result based on the efficiency deviation quantification value, so as to obtain the power generation prediction result of the distributed photovoltaic system during the prediction period.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores multiple instructions, which are adapted to be loaded by a processor and executed by the distributed photovoltaic power generation prediction method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, It includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the distributed photovoltaic power generation prediction method according to any one of claims 1 to 7.

Images