Distributed photovoltaic power generation centralized operation management and control method and system

Abstract

The application discloses a centralized operation management and control method and system for distributed photovoltaic power generation, relates to the technical field of operation management and control, and comprises the following steps: performing photovoltaic unit division and computing node deployment on a target photovoltaic region to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes; collecting N multi-source operation data streams of the N distributed photovoltaic power generation units, performing operation state evaluation, and obtaining N photovoltaic unit operation state parameters; loading a photovoltaic power generation operation strategy library, performing mode matching and integrated analysis on the N photovoltaic unit operation state parameters, determining target power generation operation strategy parameters, and performing photovoltaic power generation operation management and control through the target power generation operation strategy parameters. The application solves the technical problem that the prior art cannot uniformly and efficiently centrally manage and optimally dispatch the distributed photovoltaic power generation units, and achieves the technical effect of realizing centralized and intelligent management and control of distributed photovoltaic power generation.

Description

Technical Field

[0001] This invention relates to the field of operation and management technology, specifically to a centralized operation and management method and system for distributed photovoltaic power generation. Background Technology

[0002] Distributed photovoltaic (PV) power generation units are typically numerous, widely distributed, and vary significantly in equipment type and operating environment. Each unit largely relies on independent operation and management, with operational data scattered across different nodes. A lack of a unified aggregation and analysis mechanism hinders timely information sharing between units. The absence of a centralized operation and management system makes it difficult to perform global optimization scheduling based on the overall operational status, resulting in untimely scheduling responses, low operational efficiency, and difficulty in combining multi-source data for real-time optimization of power generation strategies. Consequently, achieving global optimization scheduling for distributed PV power generation units remains challenging. Summary of the Invention

[0003] This application provides a centralized operation and management method and system for distributed photovoltaic power generation, which is used to address the technical problem that existing technologies are unable to achieve unified and efficient centralized management and optimized scheduling of distributed photovoltaic power generation units.

[0004] In view of the above problems, this application provides a centralized operation and management method and system for distributed photovoltaic power generation.

[0005] The first aspect of this application provides a centralized operation and management method for distributed photovoltaic power generation, the method comprising:

[0006] The target photovoltaic area is divided into photovoltaic units and computing nodes are deployed to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes. N multi-source operation data streams are collected from the N distributed photovoltaic power generation units, including equipment layer data and grid layer data. Based on the N photovoltaic unit edge nodes, the operation status of the N multi-source operation data streams is evaluated to obtain N photovoltaic unit operation status parameters. These N photovoltaic unit operation status parameters are centrally uploaded to the photovoltaic operation control center, which loads the photovoltaic power generation operation strategy library. Based on the photovoltaic power generation operation strategy library, pattern matching and integration analysis are performed on the N photovoltaic unit operation status parameters to determine the target power generation operation strategy parameters, and photovoltaic power generation operation management is carried out using these target power generation operation strategy parameters.

[0007] A second aspect of this application provides a centralized operation and management system for distributed photovoltaic power generation, the system comprising:

[0008] The system comprises the following modules: a node deployment module for dividing the target photovoltaic area into photovoltaic units and deploying computational nodes to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes; a data stream acquisition module for acquiring N multi-source operational data streams from the N distributed photovoltaic power generation units, including equipment-level data and grid-level data; an operational status assessment module for assessing the operational status of the N multi-source operational data streams based on the N photovoltaic unit edge nodes to obtain N photovoltaic unit operational status parameters; a data transmission module for centrally uploading the N photovoltaic unit operational status parameters to the photovoltaic operation control center, which then loads the photovoltaic power generation operation strategy library; and an operation management module for performing pattern matching and integrated analysis on the N photovoltaic unit operational status parameters based on the photovoltaic power generation operation strategy library, determining target power generation operation strategy parameters, and performing photovoltaic power generation operation management based on these target power generation operation strategy parameters.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application divides a target photovoltaic area into photovoltaic units and deploys computing nodes to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes. It collects N multi-source operational data streams from the N distributed photovoltaic power generation units, including equipment-level data and grid-level data. Based on the N photovoltaic unit edge nodes, it evaluates the operational status of the N multi-source operational data streams to obtain N photovoltaic unit operational status parameters. These N photovoltaic unit operational status parameters are then centrally uploaded to a photovoltaic operation control center, which loads a photovoltaic power generation operation strategy library. Based on the photovoltaic power generation operation strategy library, it performs pattern matching and integrated analysis on the N photovoltaic unit operational status parameters to determine target power generation operation strategy parameters, and then uses these parameters for photovoltaic power generation operation management and control. This invention solves the technical problem of the difficulty in achieving unified and efficient centralized management and optimized scheduling of distributed photovoltaic power generation units in existing technologies. By centrally collecting and analyzing the operational status of each photovoltaic unit and matching the optimal operation strategy, it achieves the technical effect of centralized and intelligent management and control of distributed photovoltaic power generation. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1A schematic diagram of the centralized operation and management method for distributed photovoltaic power generation provided in the embodiments of this application;

[0013] Figure 2 This is a schematic diagram of the centralized operation and management system structure for distributed photovoltaic power generation provided in an embodiment of this application.

[0014] Figure labeling: Node deployment module 11, data stream acquisition module 12, operation status assessment module 13, data transmission module 14, operation management module 15. Detailed Implementation

[0015] This application provides a centralized operation and management method and system for distributed photovoltaic power generation, addressing the technical problem that existing technologies struggle to achieve unified and efficient centralized management and optimized scheduling of distributed photovoltaic power generation units. By centrally collecting and analyzing the operating status of each photovoltaic unit and matching the optimal operating strategy, it achieves the technical effect of centralized and intelligent management and control of distributed photovoltaic power generation.

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0018] Example 1, as Figure 1 As shown, this application provides a centralized operation and management method for distributed photovoltaic power generation, the method comprising:

[0019] Step S100: Divide the target photovoltaic area into photovoltaic units and deploy computing nodes to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes.

[0020] In this embodiment, when dividing the target photovoltaic area into photovoltaic units and deploying computing nodes, the operational attribute information of the target photovoltaic area is first obtained, including geographical distribution data, equipment characteristic data, historical meteorological data, and grid connection conditions. Then, based on this operational attribute information, the target photovoltaic area is clustered to complete the photovoltaic unit division, resulting in N distributed photovoltaic power generation units. Next, the computing requirements of each distributed photovoltaic power generation unit are analyzed sequentially to determine the corresponding analysis requirement parameters. Finally, computing nodes are deployed according to the analysis requirement parameters, and corresponding photovoltaic unit edge nodes are configured for each distributed photovoltaic power generation unit, thereby forming N photovoltaic unit edge nodes.

[0021] Furthermore, in the method provided in the application embodiments, obtaining N distributed photovoltaic power generation units and N photovoltaic unit edge nodes further includes:

[0022] The operation attribute information of the target photovoltaic area is obtained, including geographical distribution data, equipment characteristic data, historical meteorological data, and grid connection conditions. Based on the operation attribute information, the target photovoltaic area is clustered to obtain N distributed photovoltaic power generation units. Demand analysis is performed on the N distributed photovoltaic power generation units in sequence to determine the analysis demand parameters of the N photovoltaic power generation units. Edge nodes are deployed on the N distributed photovoltaic power generation units according to the analysis demand parameters of the N photovoltaic power generation units to obtain the N photovoltaic unit edge nodes.

[0023] In this embodiment, the operational attribute information of the target photovoltaic area is first obtained. This operational attribute information includes geographical distribution data, equipment characteristic data, historical meteorological data, and grid connection conditions. The geographical distribution data is obtained through a GIS system, accurate to the latitude, longitude, elevation, and relative location of each photovoltaic facility. The equipment characteristic data is provided by a pre-set operation and maintenance database, including the model, rated power, and conversion efficiency of photovoltaic modules, as well as the rated capacity and output characteristics of inverters. The historical meteorological data is obtained from a meteorological monitoring system or a third-party meteorological service, including indicators such as multi-year solar irradiance, sunshine duration, temperature, and humidity. The grid connection conditions are provided by a power dispatching system or grid connection agreement documents, which describe the geographical location, capacity limitations, and power quality requirements of the grid connection point.

[0024] Subsequently, the target photovoltaic (PV) region is clustered based on operational attribute information. In this process, key features are extracted from the operational attribute information of the target PV region to form a key feature set, which is then used to classify and label the region, resulting in a set of PV region operational feature parameters. Then, K-means clustering analysis is used based on this set of PV region operational feature parameters to obtain a set of PV region feature clusters. Finally, the region is divided into N distributed PV power generation units according to this feature cluster set.

[0025] After obtaining N distributed photovoltaic (PV) power generation units, computational demand analysis is performed on each of the N units to determine the required parameters for each unit. Specifically, historical power generation data and real-time operating data of each distributed PV power generation unit are retrieved, and the data acquisition frequency, daily data volume, and data type are statistically analyzed. Combined with information such as the model, rated power, and output characteristics of the equipment within the unit, the required computing power, storage capacity, and network bandwidth are directly determined according to preset rules. The preset rules are set as follows: when the data acquisition frequency is higher than 1 time / second and the daily data volume is greater than 5GB, a 4-core processor, 8GB of RAM, and at least 256GB of local storage are configured; when the acquisition frequency is between 1 time / minute and 1 time / second and the daily data volume is between 500MB and 5GB, a 2-core processor, 4GB of RAM, and 128GB of local storage are configured; when the acquisition frequency is lower than 1 time / minute and the daily data volume is less than 500MB, a 1-core processor, 2GB of RAM, and 64GB of local storage are configured. By using this matching method based on fixed conditions, we can obtain the N photovoltaic power generation unit analysis requirements parameters that correspond one-to-one with the N distributed photovoltaic power generation units.

[0026] Subsequently, based on the demand parameters of N photovoltaic power generation units, edge nodes were deployed for these N distributed photovoltaic power generation units. During deployment, edge nodes were installed in locations close to the main data acquisition equipment, for example, no more than 50 meters from the main combiner box or inverter, to reduce data transmission latency and signal loss. During deployment, hardware configurations were directly selected according to the aforementioned preset rules: high-performance units were equipped with industrial-grade edge nodes featuring multi-core processors, large-capacity memory, and high-speed solid-state storage; medium-performance units were equipped with medium-configuration edge nodes; and low-performance units were equipped with low-power, miniaturized edge devices. Before installation, stable power supply and communication conditions were ensured at the node locations, and data acquisition programs, data preprocessing programs, and operational status assessment functions were pre-installed to ensure local data processing and status assessment even in the event of communication interruptions or significant delays. Through direct configuration based on preset rules, each distributed photovoltaic power generation unit was equipped with a photovoltaic unit edge node matching its operational data scale and frequency, ultimately resulting in N photovoltaic unit edge nodes.

[0027] Furthermore, in the method provided in the application embodiments, obtaining N distributed photovoltaic power generation units further includes:

[0028] Key features are extracted from each operational attribute in the operational attribute information to obtain a key feature set of operational attributes; the target photovoltaic area is classified and identified according to the key feature set of operational attributes to obtain a set of operational feature parameters of the photovoltaic area; K-means clustering analysis is performed based on the set of operational feature parameters of the photovoltaic area to obtain a set of feature clusters of the photovoltaic area; the target photovoltaic area is divided into units according to the set of feature clusters of the photovoltaic area to obtain the N distributed photovoltaic power generation units.

[0029] In this embodiment, key features are first extracted from each operational attribute in the operational attribute information. This involves filtering and formatting geographical distribution data, equipment characteristic data, historical meteorological data, and grid connection conditions to transform the raw information into numerical or standardized parameter values ​​suitable for cluster analysis. For example, geographical distribution data is parsed using a GIS system to obtain the latitude and longitude coordinates and altitude of each photovoltaic facility; equipment characteristic data, such as rated power, conversion efficiency, and inverter capacity, is derived from the operation and maintenance database; historical meteorological data, including annual average sunshine duration, annual average irradiance, and average temperature, is obtained through a meteorological monitoring system interface; and grid connection conditions are provided by the power dispatching system and transformed into quantitative indicators such as grid connection point capacity, transmission distance, and allowable voltage fluctuation range. Through this method, the raw data of each operational attribute are transformed into characteristic parameters representing operational differences, forming a key feature set of operational attributes.

[0030] Subsequently, the target photovoltaic areas are classified and identified according to the key feature set of operational attributes, that is, each photovoltaic facility or group of facilities is assigned a unique feature label. This process uses feature coding methods to convert the parameters in the key feature set of operational attributes into corresponding classification codes. For example, geographical coordinates are converted into two-dimensional numerical vectors, equipment power and efficiency are expressed as percentage values, and meteorological and grid parameters are standardized to comparable units. Through this classification and identification method, a set of operational feature parameters for the photovoltaic area is generated.

[0031] Next, K-means clustering analysis is performed based on the photovoltaic area operation characteristic parameter set. In this process, weights are assigned to each key feature in the key feature set of operation attributes to form operation attribute feature weight factor information; then, the number of cluster centers K is initialized, and K-means clustering weighted calculation is performed on the photovoltaic area operation characteristic parameter set based on this number and weight factor information to obtain an initial set of regional feature clusters; finally, iterative clustering optimization is performed on the initial set until a preset termination condition is met, generating a photovoltaic area feature cluster set.

[0032] Finally, the target photovoltaic area is divided into units according to the photovoltaic area feature cluster set. Photovoltaic facilities belonging to the same feature cluster are grouped into an independent operating unit and assigned a unique unit identifier. The geographical scope of each unit is marked in the GIS system, and the equipment parameters, meteorological conditions, and grid connection information of all photovoltaic facilities within the unit are integrated to form a standardized management file. Through the above steps, N distributed photovoltaic power generation units are finally obtained.

[0033] Furthermore, in the method provided in the application embodiments, obtaining the photovoltaic region feature cluster set further includes:

[0034] Weights are assigned to each key feature in the set of key operational attributes to determine the operational attribute feature weight factor information; the number of cluster centers K is initialized, and K-means clustering weighted calculation is performed on the photovoltaic area operational feature parameter set using the number of cluster centers K according to the operational attribute feature weight factor information to obtain an initial set of regional feature clusters; the initial set of regional feature clusters is iteratively clustered and optimized until a preset termination condition is met to obtain the photovoltaic area feature cluster set.

[0035] In this embodiment, the key features in the operational attribute key feature set are first weighted. A preset weighting method is used to set fixed numerical weights for different features to reflect their importance in clustering calculations. For example, the weight of geographical distribution data is preset to 0.4, the weight of equipment characteristic data is preset to 0.25, the weight of historical meteorological data is preset to 0.2, and the weight of power grid access conditions is preset to 0.15. The above weights are then normalized so that their sum is 1.0, thereby forming the operational attribute feature weight factor information.

[0036] The number of cluster centers K is then initialized using a preset value, for example, setting K to 10 to divide the target photovoltaic area into 10 initial clusters in subsequent cluster analysis. After determining the value of K, K-means clustering weighted calculation is performed on the photovoltaic area's operational characteristic parameter set using the number of cluster centers K combined with operational attribute feature weighting factors. In this calculation process, K photovoltaic facilities are first randomly selected as initial cluster centers. Then, the weighted Euclidean distance between each photovoltaic facility and each cluster center is calculated (i.e., each feature value is multiplied by its corresponding weighting factor before being included in the distance calculation). The photovoltaic facilities are assigned to the cluster with the smallest weighted distance, and the cluster center positions are updated based on the average value of the samples within each cluster, completing one full clustering iteration.

[0037] Next, iterative clustering optimization is performed on the initial set of regional feature clusters. This involves repeatedly executing the photovoltaic facility allocation and cluster center update steps until a preset termination condition is met, such as stopping the calculation when the change in the position of the old and new cluster centers is less than 0.001 or the number of iterations reaches 300. Through the above steps, the final set of photovoltaic regional feature clusters is obtained, which includes groups of photovoltaic facilities that are highly similar in terms of geographical location, equipment performance, meteorological conditions, and grid connection characteristics.

[0038] Step S200: Collect N multi-source operation data streams from the N distributed photovoltaic power generation units, wherein the N multi-source operation data streams include equipment layer data and grid layer data.

[0039] In this embodiment, for the devices within the N distributed photovoltaic power generation units, such as photovoltaic modules, inverters, and batteries, their operating parameters are monitored and recorded in real time using built-in sensors. These sensors periodically or in real time collect device-level data including output power, voltage, current, temperature, and conversion efficiency. Simultaneously, grid-level data, including grid voltage, current, power factor, and grid frequency, is collected through the interface connecting the photovoltaic unit to the grid. Through this process, N multi-source operating data streams are obtained.

[0040] Step S300: Based on the edge nodes of the N photovoltaic units, evaluate the operating status of the N multi-source operating data streams to obtain the operating status parameters of the N photovoltaic units.

[0041] In this embodiment, when evaluating the operational status of N multi-source operational data streams based on N photovoltaic unit edge nodes, the N multi-source operational data streams are processed by calling N data preprocessing programs and N photovoltaic operational status evaluators. Specifically, firstly, the collected multi-source operational data streams are preprocessed according to the data preprocessing programs to remove noise, correct outliers, and format them, resulting in N usable multi-source operational data streams. Subsequently, the operational status of each photovoltaic unit is evaluated based on the processed data streams using the photovoltaic operational status evaluators, ultimately yielding N photovoltaic unit operational status parameters.

[0042] Furthermore, in the method provided in the application embodiments, obtaining the operating status parameters of N photovoltaic units further includes:

[0043] Based on the N photovoltaic unit edge nodes, N data preprocessing programs and N photovoltaic operation status evaluators are invoked; the N multi-source operation data streams are preprocessed according to the N data preprocessing programs to obtain N usable multi-source operation data streams; the operation status of the N usable multi-source operation data streams is evaluated based on the N photovoltaic operation status evaluators to obtain the N photovoltaic unit operation status parameters.

[0044] In this embodiment of the application, firstly, based on the N photovoltaic unit edge nodes, the corresponding N pre-prepared data preprocessing programs and N photovoltaic operation status evaluators are invoked.

[0045] Next, N multi-source running data streams are preprocessed according to N data preprocessing procedures. Each data preprocessing procedure is responsible for cleaning, denoising, outlier correction, time alignment, and standardization of these multi-source data to ensure data quality and consistency. After these steps, N usable multi-source running data streams are obtained.

[0046] Subsequently, based on N pre-trained photovoltaic (PV) operation status evaluators, the operation status of N available multi-source operation data streams is evaluated. Each PV operation status evaluator uses its pre-learned model or rules to determine the operation status of each PV unit by analyzing the processed multi-source data streams. For example, the PV operation status evaluator compares the output power of the PV unit with a preset normal range, or identifies whether there are any abnormalities in the equipment based on grid parameters (such as voltage and current fluctuations). The status of each PV unit is categorized as normal, faulty, or inefficient. Through this evaluation process, the N available multi-source operation data streams and the corresponding status labels are integrated to obtain N PV unit operation status parameters.

[0047] Furthermore, in the method provided in the application embodiments, the step of calling N data preprocessing programs and N photovoltaic operating status evaluators further includes:

[0048] Historical power generation data is mined based on the N photovoltaic unit edge nodes to collect N historical power generation datasets; preprocessing steps are parsed on the N historical power generation datasets to determine N data preprocessing programs; operation status identification and evaluation training optimization are performed on the N historical power generation datasets to obtain N photovoltaic operation status evaluators; the N data preprocessing programs and the N photovoltaic operation status evaluators are stored in the N photovoltaic unit edge nodes.

[0049] In this embodiment, historical power generation datasets are first collected from each photovoltaic unit based on N photovoltaic unit edge nodes through historical power generation data mining. Specifically, the operating data of devices such as photovoltaic modules, inverters, and batteries, including output power, voltage, current, temperature, and irradiance, are recorded via a SCADA system (Supervisory and Data Acquisition System). During the historical data mining process, data query and extraction techniques are used to extract the required historical data from the database, thereby obtaining historical power generation datasets for N photovoltaic units.

[0050] Next, based on the collected historical power generation datasets of N photovoltaic units, the preprocessing steps are analyzed. Specifically, firstly, the data integrity of the N photovoltaic unit historical power generation datasets is checked, and missing values ​​are identified and processed. If data is missing for certain time periods, linear interpolation or nearest-neighbor interpolation is used to fill in the missing parts. Then, outliers in the data are processed. Outliers may be caused by equipment failure or sensor errors. Algorithms such as the Z-score method are used to detect and remove these irregular outliers to ensure the accuracy of the dataset. After that, time alignment is performed because the data collection frequency of different photovoltaic units may be different, so time synchronization of the data is required to ensure that the timestamps of all data are consistent. Finally, all data are standardized to unify the values ​​of different dimensions (such as voltage, current, and power) to the same scale for easy comparison and analysis. After these steps, N data preprocessing programs are obtained, each corresponding to the processing flow of a photovoltaic unit's historical power generation dataset, ensuring the quality, integrity, and consistency of the data.

[0051] Subsequently, based on the historical power generation datasets of N photovoltaic (PV) units, operational status identification and evaluation training optimization are performed. In this process, technical experts first identify the operational status of each PV unit's historical power generation dataset, that is, associate the input data in the historical dataset with the corresponding operational status. Specifically, technical experts determine the operational status (e.g., normal, fault, or inefficient operation) of the PV unit based on historical data such as output power, current, and voltage. Technical experts determine the status identifier for each PV unit by comparing feature patterns in historical data and analyzing grid-level data. The identified operational status serves as output data, forming a complete dataset with the historical power generation dataset of the PV units, for subsequent training of machine learning models. After identification, neural network models or other machine learning algorithms are used to evaluate, train, and optimize the identified data. Specifically, the historical power generation datasets of N PV units (as input data) and the expert-identified operational status (as output data) are used to train N PV operational status evaluators. Each evaluator, based on its corresponding historical dataset, learns the operational characteristics of the photovoltaic unit (such as the relationship between output power, voltage, current, etc., and fault or inefficient operation) to determine whether real-time data indicates that the photovoltaic unit is in a fault, inefficient, or normal operating state. Through the aforementioned process, N photovoltaic operating status evaluators are finally obtained.

[0052] Finally, the N data preprocessing programs and N photovoltaic operation status evaluators are stored in the corresponding N photovoltaic unit edge nodes.

[0053] Step S400: Upload the operating status parameters of the N photovoltaic units to the photovoltaic operation control center, and load the photovoltaic power generation operation strategy library through the photovoltaic operation control center.

[0054] In this embodiment, the edge nodes corresponding to the photovoltaic units upload the operating status parameters of N photovoltaic units to the photovoltaic operation control center via a secure communication protocol (such as MQTT or HTTPS). The photovoltaic operation control center is a system platform that centrally manages and monitors the operating status of the photovoltaic power generation system, responsible for real-time data reception, status assessment, strategy execution, and optimized scheduling.

[0055] Once the photovoltaic operation control center receives all uploaded photovoltaic unit operating status parameters, it loads the pre-prepared photovoltaic power generation operation strategy library. This library contains various operation optimization schemes and scheduling strategies, such as fault handling, performance optimization, and load balancing.

[0056] Step S500: Based on the photovoltaic power generation operation strategy library, perform pattern matching and integrated analysis on the operating status parameters of the N photovoltaic units to determine the target power generation operation strategy parameters, and perform photovoltaic power generation operation control through the target power generation operation strategy parameters.

[0057] In this embodiment, based on a photovoltaic power generation operation strategy library, the operating status parameters of N photovoltaic units are first matched with operating modes. By comparing the current operating status with preset modes in the strategy library, the corresponding photovoltaic unit power generation operation strategy is triggered. Subsequently, the triggered N photovoltaic unit power generation operation strategies are used to perform integrated analysis on the operating status parameters of the N photovoltaic units, and the most suitable target power generation operation strategy parameters are comprehensively evaluated.

[0058] Finally, based on the determined target power generation operation strategy parameters, the photovoltaic operation control center manages and controls photovoltaic power generation operations, implements corresponding scheduling and optimization measures, ensures the stable and efficient operation of the photovoltaic power generation system, and responds promptly to any potential faults or efficiency degradation issues.

[0059] Furthermore, in the method provided in the application embodiments, determining the target power generation operation strategy parameters further includes:

[0060] Based on the photovoltaic power generation operation strategy library, the operation mode is matched for the operating status parameters of the N photovoltaic units, triggering the power generation operation strategies of the N photovoltaic units; the operating status parameters of the N photovoltaic units are integrated and analyzed using the power generation operation strategies of the N photovoltaic units to determine the target power generation operation strategy parameters.

[0061] In this embodiment, the photovoltaic (PV) operation control center first performs operation mode matching on the operating status parameters of N PV units based on a PV power generation operation strategy library. The operating status parameters of each PV unit include equipment-level data (such as power, voltage, current, etc.), grid-level data (such as grid voltage, current, power factor, etc.), and corresponding status identifiers (such as normal operation, fault, inefficiency, etc.). Based on the status identifier of each PV unit, the PV operation control center selects multiple PV power generation operation strategies corresponding to that status identifier from the PV power generation operation strategy library. These strategies reflect the coping solutions under different states. Subsequently, the PV operation control center uses cosine similarity calculation or other similarity measurement methods to calculate the similarity between the operating status data of each PV unit and multiple related strategies in the PV power generation operation strategy library. Through this calculation, the similarity between the current PV unit's state and multiple strategies in the strategy library is identified, thereby selecting the strategy that best matches the current PV unit's operating state. Through this process, N PV unit power generation operation strategies are obtained.

[0062] Next, an integrated analysis of the operating state parameters of N photovoltaic (PV) units is performed using N PV unit power generation operation strategies. This process begins by analyzing the operating state parameters of the N PV units using these strategies, and determining the power generation operation strategy threshold for each PV unit. Then, preset PV system power generation operation objectives are established, such as improving overall system power generation efficiency, optimizing load distribution, and balancing grid demand. Based on these objectives, a global optimization of the N PV unit power generation operation strategy thresholds is performed, and the optimal strategy combination is calculated using an optimization algorithm to determine the target power generation operation strategy parameters.

[0063] Furthermore, in the method provided in the application embodiments, the step of integrating and analyzing the operating status parameters of the N photovoltaic units using the power generation operation strategies of the N photovoltaic units to determine the target power generation operation strategy parameters further includes:

[0064] The operating status parameters of the N photovoltaic units are analyzed using the power generation operation strategies of the N photovoltaic units to obtain N photovoltaic power generation operation strategy thresholds; a photovoltaic system power generation operation target is preset, and the N photovoltaic power generation operation strategy thresholds are globally optimized according to the photovoltaic system power generation operation target to determine the target power generation operation strategy parameters.

[0065] In this embodiment, the operating status parameters of N photovoltaic (PV) units are first analyzed using existing PV unit power generation operation strategies. These PV unit operating status parameters include equipment-level data, grid-level data, and the status identifier of each PV unit. During the analysis, these parameters are evaluated based on the current operating status of each PV unit and the PV power generation operation strategy. Based on preset thresholds in the strategy, the operating status of each PV unit is adjusted accordingly. For example, if the output power of a PV unit is lower than a set threshold, a corresponding adjustment strategy is triggered, such as adjusting the inverter's output power or limiting the maximum output power of the PV unit, to avoid equipment damage and improve system stability. That is, by comparing the actual operating data of these PV units with the preset strategy, the operating limits of each PV unit under the current conditions are extracted; these operating limits are the PV power generation operation strategy thresholds. Through this process, N PV power generation operation strategy thresholds are obtained.

[0066] Next, preset the photovoltaic system power generation operation goals. These goals include multiple aspects, such as maximizing photovoltaic power generation efficiency and optimizing load distribution.

[0067] Subsequently, a global optimization is performed on N photovoltaic power generation operation strategy thresholds according to the photovoltaic system's power generation operation objectives. In this process, firstly, a photovoltaic power generation fitness function is constructed. Then, based on the obtained N photovoltaic power generation operation strategy thresholds, a photovoltaic power generation operation strategy particle space is initialized, where each particle represents a possible strategy configuration. Using the photovoltaic power generation fitness function, the control center performs a global optimization within the particle space, finding the optimal strategy parameters through iterative calculation, and finally determining the target power generation operation strategy parameters.

[0068] Furthermore, in the method provided in the application embodiments, the step of globally optimizing the N photovoltaic power generation operation strategy thresholds according to the photovoltaic system power generation operation target to determine the target power generation operation strategy parameters further includes:

[0069] Based on the photovoltaic system power generation operation target, a photovoltaic power generation fitness function is constructed; based on the N photovoltaic power generation operation strategy thresholds, a photovoltaic power generation operation strategy particle space is initialized; the photovoltaic power generation fitness function is used to perform global optimization within the photovoltaic power generation operation strategy particle space to determine the target power generation operation strategy parameters.

[0070] In this embodiment, a photovoltaic (PV) power generation fitness function is first constructed based on the PV system's power generation operation objectives. The PV power generation fitness function is a mathematical function used to evaluate the performance of different strategies in achieving predetermined objectives (such as maximizing power generation efficiency, balancing load, and reducing energy consumption). The fitness function weights multiple objectives, and the value of each objective is normalized (usually to between 0 and 1) to ensure that all objectives are compared on the same scale. For example, after normalizing the power generation efficiency, load balancing, and energy loss values, the fitness function can be expressed as: fitness value = Among them, the values ​​of power generation efficiency, load balance, and energy loss are normalized to ensure that their values ​​are within the same range. , and These are the weights of each objective.

[0071] Next, based on N photovoltaic power generation operation strategy thresholds, a photovoltaic power generation operation strategy particle space is generated, where each particle represents a possible strategy combination. Within this space, each particle contains multiple parameters, such as the output power of the photovoltaic unit, inverter regulation parameters, and load allocation. Randomly generating initial particles provides a broad search range for the optimization algorithm.

[0072] Then, using the photovoltaic power generation fitness function, the control center performs global optimization within the particle space of the photovoltaic power generation operation strategy. Particle Swarm Optimization (PSO) is used here. Each particle represents a strategy combination and is optimized based on its fitness value. The particle swarm continuously updates the parameters of each particle, ultimately finding the strategy combination that maximizes the fitness value. The fitness value of each particle is calculated by the fitness function, indicating the merits of the current strategy combination in satisfying the system objective.

[0073] After multiple iterations and optimizations, the target power generation operation strategy parameters were finally obtained. These parameters include the output power of each photovoltaic unit, inverter regulation settings, and load distribution.

[0074] In summary, the embodiments of this application have at least the following technical effects:

[0075] This application divides a target photovoltaic area into photovoltaic units and deploys computing nodes to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes. It collects N multi-source operational data streams from the N distributed photovoltaic power generation units, including equipment-level data and grid-level data. Based on the N photovoltaic unit edge nodes, it evaluates the operational status of the N multi-source operational data streams to obtain N photovoltaic unit operational status parameters. These N photovoltaic unit operational status parameters are then centrally uploaded to a photovoltaic operation control center, which loads a photovoltaic power generation operation strategy library. Based on the photovoltaic power generation operation strategy library, it performs pattern matching and integrated analysis on the N photovoltaic unit operational status parameters to determine target power generation operation strategy parameters, and then uses these parameters for photovoltaic power generation operation management and control. This invention solves the technical problem of the difficulty in achieving unified and efficient centralized management and optimized scheduling of distributed photovoltaic power generation units in existing technologies. By centrally collecting and analyzing the operational status of each photovoltaic unit and matching the optimal operation strategy, it achieves the technical effect of centralized and intelligent management and control of distributed photovoltaic power generation.

[0076] Example 2, based on the same inventive concept as the centralized operation and management method for distributed photovoltaic power generation in the foregoing examples, such as... Figure 2 As shown, this application provides a centralized operation and management system for distributed photovoltaic power generation. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0077] The module 11 is used to divide the target photovoltaic area into photovoltaic units and deploy computing nodes to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes; the data stream acquisition module 12 is used to acquire N multi-source operation data streams of the N distributed photovoltaic power generation units, the N multi-source operation data streams including equipment layer data and grid layer data; the operation status assessment module 13 is used to assess the operation status of the N multi-source operation data streams based on the N photovoltaic unit edge nodes to obtain N photovoltaic unit operation status parameters; the data transmission module 14 is used to centrally upload the N photovoltaic unit operation status parameters to the photovoltaic operation control center, and load the photovoltaic power generation operation strategy library through the photovoltaic operation control center; the operation management module 15 is used to perform pattern matching and integrated analysis on the N photovoltaic unit operation status parameters based on the photovoltaic power generation operation strategy library, determine the target power generation operation strategy parameters, and perform photovoltaic power generation operation management through the target power generation operation strategy parameters.

[0078] Furthermore, the system is also used to implement the following functions:

[0079] The operation attribute information of the target photovoltaic area is obtained, including geographical distribution data, equipment characteristic data, historical meteorological data, and grid connection conditions. Based on the operation attribute information, the target photovoltaic area is clustered to obtain N distributed photovoltaic power generation units. Demand analysis is performed on the N distributed photovoltaic power generation units in sequence to determine the analysis demand parameters of the N photovoltaic power generation units. Edge nodes are deployed on the N distributed photovoltaic power generation units according to the analysis demand parameters of the N photovoltaic power generation units to obtain the N photovoltaic unit edge nodes.

[0080] Furthermore, the system is also used to implement the following functions:

[0081] Key features are extracted from each operational attribute in the operational attribute information to obtain a key feature set of operational attributes; the target photovoltaic area is classified and identified according to the key feature set of operational attributes to obtain a set of operational feature parameters of the photovoltaic area; K-means clustering analysis is performed based on the set of operational feature parameters of the photovoltaic area to obtain a set of feature clusters of the photovoltaic area; the target photovoltaic area is divided into units according to the set of feature clusters of the photovoltaic area to obtain the N distributed photovoltaic power generation units.

[0082] Furthermore, the system is also used to implement the following functions:

[0083] Weights are assigned to each key feature in the set of key operational attributes to determine the operational attribute feature weight factor information; the number of cluster centers K is initialized, and K-means clustering weighted calculation is performed on the photovoltaic area operational feature parameter set using the number of cluster centers K according to the operational attribute feature weight factor information to obtain an initial set of regional feature clusters; the initial set of regional feature clusters is iteratively clustered and optimized until a preset termination condition is met to obtain the photovoltaic area feature cluster set.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] Based on the N photovoltaic unit edge nodes, N data preprocessing programs and N photovoltaic operation status evaluators are invoked; the N multi-source operation data streams are preprocessed according to the N data preprocessing programs to obtain N usable multi-source operation data streams; the operation status of the N usable multi-source operation data streams is evaluated based on the N photovoltaic operation status evaluators to obtain the N photovoltaic unit operation status parameters.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] Historical power generation data is mined based on the N photovoltaic unit edge nodes to collect N historical power generation datasets; preprocessing steps are parsed on the N historical power generation datasets to determine N data preprocessing programs; operation status identification and evaluation training optimization are performed on the N historical power generation datasets to obtain N photovoltaic operation status evaluators; the N data preprocessing programs and the N photovoltaic operation status evaluators are stored in the N photovoltaic unit edge nodes.

[0088] Furthermore, the system is also used to implement the following functions:

[0089] Based on the photovoltaic power generation operation strategy library, the operation mode is matched for the operating status parameters of the N photovoltaic units, triggering the power generation operation strategies of the N photovoltaic units; the operating status parameters of the N photovoltaic units are integrated and analyzed using the power generation operation strategies of the N photovoltaic units to determine the target power generation operation strategy parameters.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] The operating status parameters of the N photovoltaic units are analyzed using the power generation operation strategies of the N photovoltaic units to obtain N photovoltaic power generation operation strategy thresholds; a photovoltaic system power generation operation target is preset, and the N photovoltaic power generation operation strategy thresholds are globally optimized according to the photovoltaic system power generation operation target to determine the target power generation operation strategy parameters.

[0092] Furthermore, the system is also used to implement the following functions:

[0093] Based on the photovoltaic system power generation operation target, a photovoltaic power generation fitness function is constructed; based on the N photovoltaic power generation operation strategy thresholds, a photovoltaic power generation operation strategy particle space is initialized; the photovoltaic power generation fitness function is used to perform global optimization within the photovoltaic power generation operation strategy particle space to determine the target power generation operation strategy parameters.

[0094] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0095] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0096] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A centralized operation and management method for distributed photovoltaic power generation, characterized in that, The method includes: The target photovoltaic area is divided into photovoltaic units and computing nodes are deployed to obtain N distributed photovoltaic power generation units and N photovoltaic unit edge nodes, including: Obtain the operational attribute information of the target photovoltaic area, including geographical distribution data, equipment characteristic data, historical meteorological data, and grid connection conditions; Based on the operational attribute information, the target photovoltaic area is clustered and divided to obtain N distributed photovoltaic power generation units; The demand analysis is performed sequentially on the N distributed photovoltaic power generation units to determine the analysis demand parameters for the N photovoltaic power generation units; Based on the analysis of the demand parameters of the N photovoltaic power generation units, edge nodes are deployed for the N distributed photovoltaic power generation units to obtain the edge nodes of the N photovoltaic units; Collect N multi-source operation data streams from the N distributed photovoltaic power generation units, wherein the N multi-source operation data streams include equipment layer data and grid layer data; Based on the edge nodes of the N photovoltaic units, the operating status of the N multi-source operating data streams is evaluated to obtain the operating status parameters of the N photovoltaic units; The operating status parameters of the N photovoltaic units are uploaded to the photovoltaic operation control center, and the photovoltaic power generation operation strategy library is loaded through the photovoltaic operation control center. Based on the photovoltaic power generation operation strategy library, pattern matching and integrated analysis are performed on the operating status parameters of the N photovoltaic units to determine the target power generation operation strategy parameters, and photovoltaic power generation operation management and control are carried out through the target power generation operation strategy parameters.

2. The centralized operation and management method for distributed photovoltaic power generation as described in claim 1, characterized in that, The acquisition of N distributed photovoltaic power generation units includes: Key features are extracted from each operational attribute in the operational attribute information to obtain a set of key features for operational attributes. The target photovoltaic area is classified and identified according to the key feature set of operational attributes, and the set of operational feature parameters of the photovoltaic area is obtained. K-means clustering analysis was performed based on the photovoltaic area operation characteristic parameter set to obtain a photovoltaic area characteristic cluster set; The target photovoltaic region is divided into units according to the photovoltaic region feature cluster set to obtain the N distributed photovoltaic power generation units.

3. The centralized operation and management method for distributed photovoltaic power generation as described in claim 2, characterized in that, The obtained photovoltaic region feature cluster set includes: Weights are assigned to each key feature in the set of key features of operational attributes to determine the weight factor information of operational attribute features; Initialize the number of cluster centers K, and use the number of cluster centers K to perform K-means clustering weighted calculation on the photovoltaic area operation feature parameter set according to the operation attribute feature weight factor information to obtain an initial set of regional feature clusters; The initial set of regional feature clusters is iteratively clustered and optimized until a preset termination condition is met, thus obtaining the set of photovoltaic regional feature clusters.

4. The centralized operation and management method for distributed photovoltaic power generation as described in claim 1, characterized in that, The obtained operating status parameters of N photovoltaic units include: Based on the N photovoltaic unit edge nodes, call N data preprocessing programs and N photovoltaic operation status evaluators; The N multi-source running data streams are preprocessed according to the N data preprocessing procedures to obtain N usable multi-source running data streams; The operating status of the N photovoltaic units is evaluated based on the N available multi-source operating data streams by the N photovoltaic operating status evaluators, and the operating status parameters of the N photovoltaic units are obtained.

5. The centralized operation and management method for distributed photovoltaic power generation as described in claim 4, characterized in that, The invocation of N data preprocessing programs and N photovoltaic operation status evaluators includes: Historical power generation data is mined based on the edge nodes of the N photovoltaic units to collect historical power generation datasets for the N photovoltaic units; The N photovoltaic unit historical power generation datasets are analyzed through preprocessing steps to determine N data preprocessing procedures; The historical power generation datasets of the N photovoltaic units are used for operation status identification and evaluation training and optimization to obtain N photovoltaic operation status evaluators; The N data preprocessing programs and the N photovoltaic operating status evaluators are stored in the N photovoltaic unit edge nodes.

6. The centralized operation and management method for distributed photovoltaic power generation as described in claim 1, characterized in that, The determination of the target power generation operation strategy parameters includes: Based on the photovoltaic power generation operation strategy library, the operation mode is matched with the operating status parameters of the N photovoltaic units, and the power generation operation strategy of the N photovoltaic units is triggered; The operating status parameters of the N photovoltaic units are integrated and analyzed using the power generation operation strategies of the N photovoltaic units to determine the target power generation operation strategy parameters.

7. The centralized operation and management method for distributed photovoltaic power generation as described in claim 6, characterized in that, The process of integrating and analyzing the operating status parameters of the N photovoltaic units using the power generation operation strategies of the N photovoltaic units to determine the target power generation operation strategy parameters includes: The operating status parameters of the N photovoltaic units are analyzed using the power generation operation strategies of the N photovoltaic units to obtain the threshold values ​​of the N photovoltaic power generation operation strategies; A preset photovoltaic system power generation operation target is established, and the N photovoltaic power generation operation strategy thresholds are globally optimized according to the photovoltaic system power generation operation target to determine the target power generation operation strategy parameters.

8. The centralized operation and management method for distributed photovoltaic power generation as described in claim 7, characterized in that, The step of globally optimizing the N photovoltaic power generation operation strategy thresholds according to the photovoltaic system power generation operation target to determine the target power generation operation strategy parameters includes: Based on the photovoltaic system's power generation operation objectives, a photovoltaic power generation fitness function is constructed. Based on the N photovoltaic power generation operation strategy thresholds, the photovoltaic power generation operation strategy particle space is initialized and generated. The photovoltaic power generation fitness function is used to perform global optimization within the particle space of the photovoltaic power generation operation strategy to determine the parameters of the target power generation operation strategy.

9. A centralized operation and management system for distributed photovoltaic power generation, characterized in that, The system is used to execute the centralized operation and management method for distributed photovoltaic power generation as described in any one of claims 1-8, and the system includes: The node deployment module is used to divide the target photovoltaic area into photovoltaic units and calculate the node deployment, resulting in N distributed photovoltaic power generation units and N photovoltaic unit edge nodes; The data stream acquisition module is used to acquire N multi-source operation data streams from the N distributed photovoltaic power generation units, wherein the N multi-source operation data streams include equipment layer data and grid layer data; The operation status assessment module is used to assess the operation status of the N multi-source operation data streams based on the N photovoltaic unit edge nodes, and obtain the N photovoltaic unit operation status parameters. The data transmission module is used to centrally upload the operating status parameters of the N photovoltaic units to the photovoltaic operation control center, and load the photovoltaic power generation operation strategy library through the photovoltaic operation control center; The operation and management module is used to perform pattern matching and integrated analysis on the operating status parameters of the N photovoltaic units based on the photovoltaic power generation operation strategy library, determine the target power generation operation strategy parameters, and perform photovoltaic power generation operation and management through the target power generation operation strategy parameters.

Images