Distributed photovoltaic new energy data management method and system based on cloud computing

Abstract

The invention discloses a distributed photovoltaic new energy data management method and system based on cloud computing, and belongs to the field of new energy management, and the method comprises the steps: building a photovoltaic topology architecture; the method comprises the following steps: reading an operation record of a distributed photovoltaic system in a preset time zone, mining a multi-stage data topology node, and configuring a data acquisition array; controlling the data acquisition array, and determining photovoltaic data; the photovoltaic data are transmitted to a cloud computing platform, data modular storage and photovoltaic operation and maintenance evaluation are carried out, a photovoltaic analysis result is determined, and a photovoltaic evaluation model is built in the cloud computing platform; determining a photovoltaic system adjustment plan based on the photovoltaic analysis result; and carrying out terminal visualization display on the photovoltaic analysis result and the photovoltaic system adjustment plan, and releasing the photovoltaic system adjustment plan to carry out optimization regulation and control management on the distributed photovoltaic system. The technical problems of low system operation efficiency and complex maintenance caused by high data management and optimization control difficulty and poor real-time performance of a distributed photovoltaic system in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of new energy management, and in particular to a distributed photovoltaic new energy data management method and system based on cloud computing. Background Art

[0002] With the continuous growth of global energy demand and the improvement of environmental protection awareness, renewable energy, especially solar energy, has gradually become an important part of the global energy structure. Photovoltaic power generation, as a clean and renewable energy form, has been widely used in recent years. Distributed photovoltaic systems have been rapidly developed in residential areas, commercial areas, industrial areas and other fields due to their flexibility, easy deployment and direct access to user-side electricity.

[0003] However, at present, the operating data of distributed photovoltaic systems are usually scattered in different devices and locations, lacking a unified management platform. This leads to problems such as untimely data collection, inaccurate data analysis, and inflexible system control. In addition, traditional data management methods are difficult to cope with the large amount of data generated by photovoltaic systems in different geographical locations and different time periods. Especially in large-scale distributed photovoltaic systems, the real-time and accuracy of data are particularly important. Therefore, how to uniformly manage the data and intelligently control the distributed photovoltaic system has become the key to improving the efficiency and reliability of the photovoltaic system. Summary of the invention

[0004] This application provides a distributed photovoltaic new energy data management method and system based on cloud computing, aiming to solve the technical problems in the prior art that distributed photovoltaic system data management and optimization control are difficult and have poor real-time performance, resulting in low system operation efficiency and complex maintenance.

[0005] In view of the above problems, the present application provides a distributed photovoltaic new energy data management method and system based on cloud computing.

[0006] The first aspect disclosed in the present application provides a distributed photovoltaic new energy data management method based on cloud computing, the method comprising traversing a distributed photovoltaic system and building a photovoltaic topology architecture; reading the operation records of the preset time zone of the distributed photovoltaic system, mining multi-level data topology nodes, and configuring a data acquisition array, wherein the data acquisition array includes multiple types of sensor devices; controlling the data acquisition array to perform sensor monitoring of the distributed photovoltaic system and determine photovoltaic data, wherein the photovoltaic data carries a data topology node identifier and a time-space identifier; an auxiliary transfer module transmits the photovoltaic data to a cloud computing platform, performs data modular storage and photovoltaic operation and maintenance evaluation, and determines photovoltaic analysis results, wherein the cloud computing platform has a built-in photovoltaic evaluation model, and a communication connection is established between the cloud computing platform and the data acquisition array; based on the photovoltaic analysis results, a photovoltaic system adjustment plan is determined; the photovoltaic analysis results and the photovoltaic system adjustment plan are displayed visually on the terminal, and the photovoltaic system adjustment plan is decentralized to perform optimization, regulation and management of the distributed photovoltaic system.

[0007] Another aspect disclosed in the present application provides a distributed photovoltaic new energy data management system based on cloud computing, the system comprising a photovoltaic topology architecture module: used to traverse the distributed photovoltaic system and build a photovoltaic topology architecture; a data acquisition array configuration module: used to read the operation records of the preset time zone of the distributed photovoltaic system, mine multi-level data topology nodes, and configure the data acquisition array, wherein the data acquisition array includes multiple types of sensor devices; a photovoltaic data determination module: used to control the data acquisition array, perform sensor monitoring of the distributed photovoltaic system, and determine photovoltaic data, wherein the photovoltaic data carries a data topology node identifier and a time-space identifier; a photovoltaic analysis result determination module: used to assist the transfer module, transmit the photovoltaic data to the cloud computing platform, perform data modular storage and photovoltaic operation and maintenance evaluation, and determine the photovoltaic analysis results, wherein the cloud computing platform has a built-in photovoltaic evaluation model, and the cloud computing platform establishes a communication connection with the data acquisition array; a photovoltaic system adjustment plan determination module: used to determine the photovoltaic system adjustment plan based on the photovoltaic analysis results; a terminal visualization display module: used to perform terminal visualization display of the photovoltaic analysis results and the photovoltaic system adjustment plan, and delegate the photovoltaic system adjustment plan to perform optimization and regulation management of the distributed photovoltaic system.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] Due to the adoption of a distributed photovoltaic new energy data management method based on cloud computing, the technical solution of traversing the photovoltaic system, building a topological architecture, reading operation records, mining multi-level data topological nodes, configuring a data acquisition array, and transmitting photovoltaic data to a cloud computing platform for modular storage and operation and maintenance evaluation through an auxiliary transfer module solves the technical problems in the existing technology of difficult data management and optimization control of distributed photovoltaic systems, poor real-time performance, resulting in low system operation efficiency and complex maintenance, and achieves the technical effect of efficient collection, accurate analysis and intelligent regulation of photovoltaic system data, thereby improving the overall operation efficiency of the system, reducing maintenance costs and enhancing system reliability.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A flow chart of a distributed photovoltaic new energy data management method based on cloud computing is provided for an embodiment of the present application.

[0012] Figure 2 A schematic diagram of the structure of a distributed photovoltaic new energy data management system based on cloud computing is provided for the embodiment of the present application.

[0013] Explanation of the reference numerals: photovoltaic topology architecture module 11 , data acquisition array configuration module 12 , photovoltaic data determination module 13 , photovoltaic analysis result determination module 14 , photovoltaic system adjustment plan determination module 15 , terminal visualization display module 16 . DETAILED DESCRIPTION

[0014] The overall idea of ​​the technical solution provided by this application is as follows:

[0015] The embodiment of the present application provides a distributed photovoltaic new energy data management method and system based on cloud computing. By traversing the photovoltaic system and building a topological architecture, reading the operation records of the preset time zone, mining multi-level data topological nodes, and configuring the data acquisition array of multiple types of sensor equipment. The data is transmitted to the cloud computing platform through the auxiliary transfer module for modular storage and operation and maintenance evaluation. Based on the evaluation results, a photovoltaic system adjustment plan is formulated and displayed visually through the terminal, ultimately achieving the optimization and control management of the distributed photovoltaic system.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0017] Embodiment 1, as Figure 1As shown, the embodiment of the present application provides a distributed photovoltaic new energy data management method based on cloud computing, the method comprising:

[0018] Step S100: traverse the distributed photovoltaic system and build a photovoltaic topology architecture.

[0019] Specifically, first, all power generation units in the entire distributed photovoltaic system are traversed. This is usually done through automated network scanning tools (such as Nmap) or dedicated software (such as SCADA systems). The purpose is to collect information such as the location, device type, connection method, and operating status of each power generation unit in the system.

[0020] During the traversal, the system collects detailed data about each PV unit, including its geographic location, electrical connections, current operating status, and connection relationships with other devices. This data can be collected by sensors or controllers installed on each unit.

[0021] After completing the data collection, the system will establish the topology of the photovoltaic system based on this data. This process is usually implemented through professional software tools, such as network topology design tools, GIS (geographic information system) software or power system simulation software. The topology will show the connection relationship between each photovoltaic unit, such as how the cables connect each solar panel to the inverter, how the inverter is connected to the local power grid, and even how to communicate with the remote monitoring center.

[0022] This step traverses the distributed photovoltaic system and establishes the photovoltaic topology architecture, so that the system can fully understand and grasp the operating status of the photovoltaic system and achieve efficient management and optimization. It ensures that all parts of the system can work together seamlessly, thereby maximizing the power generation efficiency and operational reliability of the photovoltaic system.

[0023] Step S200: reading the operation records of the preset time zone of the distributed photovoltaic system, mining multi-level data topology nodes, and configuring a data acquisition array, wherein the data acquisition array includes multiple types of sensor devices.

[0024] Specifically, the preset time zone refers to a specific time zone set in a distributed photovoltaic system, which is used to record and analyze the operating data of the system in different time periods. Different time zones can reflect the performance changes of the system at different times of the day, such as the impact of changes in sunlight intensity on power generation. Operation records refer to various operating data of the photovoltaic system in the preset time zone, including parameters such as power generation, voltage, current, and temperature. These data are used to evaluate the performance and health status of the system and identify potential problems. Multi-level data topology nodes refer to multiple levels of nodes divided in the system according to the importance and relevance of data. Each node represents an important source or transit point of data, including solar panels, inverters, sensors and other devices. The data acquisition array is a system composed of multiple sensor devices, which is used to monitor the operating status of the photovoltaic system in real time. Sensor devices include temperature sensors, light intensity sensors, voltage sensors, etc., which are used to collect various key operating data.

[0025] First, the system reads the operation records of the distributed photovoltaic system according to the preset time zone. These records contain detailed data of the photovoltaic system in different time periods, such as power generation, voltage and temperature at different time points from sunrise to sunset. This step is usually completed by energy management systems (EMS) or IoT platforms, which can regularly collect and store the operation data of photovoltaic systems.

[0026] Next, the system will analyze these operation records to mine multi-level data topology nodes. Identify which nodes are critical, such as an inverter that plays a key role in the entire system because it directly affects the power generation efficiency of multiple photovoltaic panels. Based on these analysis results, the data nodes are divided into different levels to determine which nodes need to be monitored.

[0027] After determining the multi-level data topology nodes, the system will configure the data acquisition array. This step involves allocating appropriate sensor devices to each important node. For example, at an important inverter location, the system installs multiple sensors, such as temperature sensors, voltage sensors, and current sensors, to fully monitor the operating status of the node.

[0028] This step can effectively monitor and manage the distributed photovoltaic system, ensure its operating efficiency in different time periods, and promptly detect and deal with potential problems.

[0029] Step S300: controlling the data acquisition array to perform sensor monitoring of the distributed photovoltaic system and determine photovoltaic data, wherein the photovoltaic data carries a data topology node identifier and a time-space identifier.

[0030] Specifically, the data topology node identifier refers to the node identifier corresponding to the data in the photovoltaic system, which is used to determine the specific location or device of the data source. For example, the output voltage data of an inverter will carry the node identifier of the inverter to identify which device or location the data comes from. The spatiotemporal identifier refers to the time and geographic location identifier of data collection, which is used to record when and where the data was collected. This helps with subsequent data analysis, especially when analyzing data changes in different time periods or locations.

[0031] The configured data acquisition array is controlled to perform sensor monitoring on the distributed photovoltaic system. During the sensor monitoring process, various sensors will collect the operating data of the photovoltaic system in real time. These data are called photovoltaic data. The collected data usually includes electrical parameters such as voltage, current, power, and environmental parameters such as temperature and light intensity. In order to ensure the traceability and accuracy of the data, data topology node identifiers and time-space identifiers are added to the photovoltaic data. The data topology node identifier is used to indicate which specific device or node the data comes from.

[0032] By controlling the data acquisition array for sensor monitoring and collecting photovoltaic data with data topology node identification and time-space identification, the system can achieve comprehensive, accurate and real-time monitoring of the photovoltaic system, ensuring the integrity and traceability of photovoltaic data, and providing a reliable basis for subsequent analysis and optimization.

[0033] Step S400: The auxiliary transfer module transmits the photovoltaic data to the cloud computing platform, performs modular data storage and photovoltaic operation and maintenance evaluation, and determines the photovoltaic analysis results, wherein the cloud computing platform has a built-in photovoltaic evaluation model, and the cloud computing platform establishes a communication connection with the data acquisition array.

[0034] Specifically, the auxiliary relay module is an intermediate device or system module for data transmission, which is responsible for transmitting the photovoltaic data obtained from the data acquisition array to the cloud computing platform. During the data transmission process, the auxiliary relay module also performs data preprocessing, such as data compression, filtering or adding timestamps. The photovoltaic evaluation model is an algorithm or calculation model built into the cloud computing platform, which is specifically used to evaluate the operating status of the photovoltaic system. The model generates evaluation results on the system performance and health status by analyzing the input data.

[0035] First, the auxiliary transfer module receives the photovoltaic data transmitted from the data acquisition array. The auxiliary transfer module not only transmits data, but also pre-processes the data, such as compressing the data to reduce the transmission bandwidth, filtering useless data to reduce the storage burden, or adding a timestamp to the data to ensure its timeliness. Then, the photovoltaic data processed by the auxiliary transfer module is transmitted to the cloud computing platform. After receiving the data, the cloud computing platform will store it modularly. The purpose of modular storage is to classify and store data of different types and sources according to a predetermined structure for subsequent retrieval and analysis.

[0036] After the data is successfully stored, the cloud computing platform will call the built-in photovoltaic evaluation model to analyze and evaluate the data. The photovoltaic evaluation model is based on machine learning algorithms, statistical analysis methods or physical models, and can extract valuable information from the data. Among them, the data transmission channel established between the data acquisition array and the cloud computing platform ensures that the photovoltaic data can be seamlessly transmitted from the front-end sensor equipment to the cloud for processing. The communication connection can be achieved through wired or wireless networks.

[0037] By transmitting photovoltaic data to the cloud computing platform through the auxiliary transfer module and using the cloud platform for modular data storage and photovoltaic operation and maintenance evaluation, efficient management and optimization of distributed photovoltaic systems can be achieved. This method ensures the integrity of the data and the accuracy of the analysis, providing strong support for the continuous optimization and maintenance of the photovoltaic system.

[0038] Step S500: Determine a photovoltaic system adjustment plan based on the photovoltaic analysis result.

[0039] Specifically, first, the current operating status of the PV system is evaluated based on the PV analysis results. The PV analysis results are generated by the PV evaluation model on the cloud computing platform, and a specific adjustment plan is formulated based on these analysis results. The adjustment plan is a detailed operation plan that explains how to adjust the PV system to optimize its performance. After the adjustment plan is determined, the system will convert it into specific operation instructions and directly apply it to the PV system through the automated control system.

[0040] The process of determining the PV system adjustment plan based on the PV analysis results includes analyzing the current system status, formulating a detailed adjustment plan, and converting the plan into specific operating instructions to optimize the performance of the PV system. This approach helps improve the system's power generation efficiency and operational reliability.

[0041] Step S600: Visually display the photovoltaic analysis results and the photovoltaic system adjustment plan on the terminal, and delegate the photovoltaic system adjustment plan to perform optimized regulation and management of the distributed photovoltaic system.

[0042] Specifically, first, the generated photovoltaic analysis results and photovoltaic system adjustment plans are visualized in the terminal. The purpose of the visualization is to allow system administrators or operators to intuitively view the analysis results and adjustment plans, understand the current status of the system and the operations that need to be performed. Commonly used visualization tools include dashboard tools (such as Grafana, Tableau) and data visualization libraries (such as D3.js, Plotly), which can convert data into graphics, charts, or other forms of visualization output.

[0043] After the visualization is completed, the system will delegate the PV system adjustment plan to the control unit of the PV system. Delegating the adjustment plan means transmitting the predetermined operation plan to the various components of the PV system, such as the inverter controller, regulator, etc., so that they can perform corresponding adjustments according to the plan.

[0044] Finally, the system monitors and adjusts the photovoltaic system in real time through optimized regulation management. The goal of optimized regulation management is to ensure the smooth implementation of the regulation plan and make necessary adjustments based on actual conditions.

[0045] Through the visual display and decentralized execution of photovoltaic analysis results and adjustment plans, the system can effectively manage the operating status of the photovoltaic system, ensure timely and accurate adjustment and optimization, and improve overall operating efficiency and system stability.

[0046] Furthermore, the mining of multi-level data topology nodes includes: traversing the operation records to determine the effective monitoring nodes; based on the data criticality, traversing the effective monitoring nodes to divide the multi-level data nodes; based on the data correlation, traversing the effective monitoring nodes to divide the data batches, wherein the data correlation includes synchronization correlation and time correlation degree; fitting the multi-level data nodes and the data batches to determine the multi-level data topology nodes.

[0047] Specifically, effective monitoring nodes refer to nodes in the photovoltaic system that are screened and considered to have a significant impact on system performance and stability. These nodes include key inverters, solar panels or sensor locations, which are usually used for key monitoring and data collection. Data criticality is a classification of the importance of different data nodes, which determines the weight of each node in the overall system. Data batches refer to groups or clusters into which data is divided according to the relevance of the data. Each batch of data comes from different time periods or different devices, but has some relevance, such as synchronization or temporal correlation. Synchronous correlation refers to whether the data is generated synchronously or almost synchronously in time.

[0048] First, the system will traverse the operation records of the distributed photovoltaic system to determine which nodes are effective monitoring nodes. This process is usually completed through big data analysis tools or IoT platforms, which can filter out nodes that have a significant impact on system performance from a large amount of historical operation data.

[0049] Next, the system will classify and grade these valid monitoring nodes based on the data criticality. Data classification algorithms (such as cluster analysis or decision trees) can be used to divide these nodes into different levels.

[0050] Then, the system further divides the data of these effective monitoring nodes into batches based on the correlation of the data. The synchronous correlation of the data can be identified by time series analysis tools, and the time correlation degree can be detected by correlation analysis or time lag analysis tools.

[0051] Finally, the multi-level data nodes and data batches are fitted. Data fitting tools (such as linear regression or machine learning models) are used to integrate these nodes and data batches to form a multi-level data topology network. This network shows the complex relationships and data flow paths between the nodes in the system.

[0052] This step enables in-depth analysis of data relationships in the PV system, identification of nodes with the greatest impact on system performance, and improvement of the efficiency and stability of the entire PV system through optimized management of these key nodes.

[0053] Furthermore, the configuration of the data acquisition array includes: traversing the valid monitoring nodes, and configuring node monitoring devices based on the node monitoring type, wherein each valid monitoring node corresponds to a monitoring device of at least one monitoring type; based on the node monitoring device, assembling it in the distributed photovoltaic system, and performing upper-level management integration to determine the data acquisition array.

[0054] Specifically, node monitoring type refers to the specific monitoring parameters or categories required for different effective monitoring nodes. For example, for photovoltaic modules, monitoring types include voltage, current, and temperature; for inverters, monitoring types include input and output power, efficiency, etc.

[0055] First, we traverse the valid monitoring nodes that have been determined and configure the corresponding monitoring devices based on the monitoring requirements of each node. This is done through field survey tools to ensure that the monitoring devices equipped at each node can accurately collect the required data.

[0056] Next, the system will select the appropriate monitoring device according to the monitoring type of different nodes. Each valid monitoring node is equipped with at least one type of monitoring device to ensure the comprehensiveness and accuracy of the data. The sensor configuration tool can help select the appropriate sensor device and determine its installation location and connection method.

[0057] After configuration is complete, the system will install these monitoring devices on the corresponding nodes and perform upper management integration, which means integrating the monitoring devices in the data acquisition array with the central control system so that all collected data can be centrally managed and analyzed. This process usually involves an IoT platform or SCADA system to achieve communication and data transmission between the data acquisition device and the central control system. Through upper management integration, the data collected by all monitoring devices can be transmitted to the cloud platform in real time for centralized analysis and management. Finally, the system will debug and optimize the entire data acquisition array to ensure that all monitoring devices can operate normally and are seamlessly integrated with the central system.

[0058] The process of configuring the data acquisition array includes determining monitoring requirements, selecting appropriate sensor equipment, performing on-site installation, and upper management integration, ensuring that key nodes in the photovoltaic system can be effectively monitored, thereby improving the overall operating efficiency and reliability of the system.

[0059] Furthermore, the auxiliary transfer module includes: transmitting the photovoltaic data to the auxiliary transfer module, performing data preprocessing, and determining valid photovoltaic data; traversing the valid photovoltaic data, and performing timestamp constraints on data transmission.

[0060] Specifically, timestamp constraints refer to attaching or checking timestamps for each set of data during data transmission to ensure the timeliness and order of the data. Timestamp constraints can help the system maintain the temporal consistency of data during analysis, which is especially important when processing real-time data.

[0061] First, the raw data generated by the photovoltaic system is transmitted to the auxiliary relay module through the data acquisition array. After receiving the data, the auxiliary relay module will perform data preprocessing on it. The purpose of preprocessing is to clean and optimize the data so that it is ready for subsequent transmission and analysis. These operations include filtering out invalid or erroneous data, such as abnormal data generated by sensor failure. It also includes compressing the data to reduce the bandwidth required for transmission, or format conversion to ensure data compatibility.

[0062] After preprocessing, the module determines which data is valid PV data. This means that the cleaned and filtered data is marked as "valid" and ready for further processing. For example, if some points in the voltage data recorded by a sensor within an hour show abnormally high or low values, these abnormal values ​​will be filtered out to ensure that the data transmitted to the cloud platform is accurate and reliable.

[0063] Next, the system will traverse all valid PV data and attach or check timestamps to these data. Timestamp constraints ensure the timeliness and order of data, that is, each piece of data has an accurate collection time and these times are correctly sorted.

[0064] After the timestamp constraints are completed, the data will be prepared and transmitted to the cloud computing platform. This process ensures that the transmitted data is cleaned, verified and time-corrected, and can be directly used for further analysis and processing on the cloud platform.

[0065] The auxiliary transfer module ensures the quality and timeliness of data transmission by preprocessing and timestamping photovoltaic data, laying a solid foundation for subsequent cloud data analysis and ensuring the efficiency and reliability of photovoltaic system management.

[0066] Furthermore, the cloud computing platform has a built-in photovoltaic evaluation model, including: determining data management requirements, and mining sample evaluation records based on the operation records; traversing the data management requirements, clustering the sample evaluation records based on the requirement similarity, and determining multiple groups of samples; based on the multiple groups of samples and the data management requirements, performing data-driven modeling to generate the photovoltaic evaluation model.

[0067] Specifically, demand similarity refers to the degree to which the system measures the similarity between data and current demand based on certain criteria (such as time, environmental conditions, equipment status, etc.) when processing data. Data with high demand similarity is considered to be more in line with the needs of the current analysis.

[0068] First, the cloud computing platform determines the data management requirements. This step involves defining the specific data types and conditions that need to be analyzed and evaluated, such as the time period that needs to be evaluated, the type of PV equipment that needs to be monitored, or the performance indicators. System administrators can set these requirements through the data management platform or cloud management interface.

[0069] Then, the system will mine relevant sample evaluation records based on these requirements and the operation records of the photovoltaic system. Data mining tools or database query tools (such as SQL query, data mining algorithm) can be used to extract qualified sample records from historical data.

[0070] The system then goes through all data management requirements and performs clustering on the extracted sample evaluation records. The purpose of clustering is to group similar sample data into groups in order to better understand the patterns and trends of the data. Clustering algorithms such as K-means clustering or hierarchical clustering can help divide the data into multiple groups, each containing similar samples. For example, sample data can be divided into multiple groups based on different sunlight conditions, ambient temperature, and equipment type for more accurate evaluation.

[0071] After the clustering process is completed, the system will perform data-driven modeling based on these multiple sets of sample data and data management requirements. This step usually uses machine learning algorithms or statistical modeling tools (such as linear regression, neural networks, support vector machines, etc.) to build a photovoltaic evaluation model. The training and generation process of the model depends on the input data. By analyzing this data, the model can learn the operating rules of the system.

[0072] The cloud computing platform uses the built-in photovoltaic evaluation model to extract samples, cluster processing and data-driven modeling based on data management needs and historical data, and finally generates a model that can evaluate the performance of the photovoltaic system. This method provides a scientific basis and decision-making support for the operation optimization of the photovoltaic system.

[0073] Furthermore, the generation of the photovoltaic evaluation model includes: based on the multiple groups of samples, extracting a group of sample data and dividing it into a training set and a verification set; training a first evaluation branch based on the training set, and verifying the convergence of the first evaluation branch based on the verification set; if the convergence does not meet the preset convergence, re-dividing and training the group of sample data; traversing the multiple groups of sample data to complete the training of the Nth evaluation branch; and integrating the first evaluation branch in parallel until the Nth evaluation branch to obtain the photovoltaic evaluation model.

[0074] Specifically, convergence is an indicator to measure whether the model has reached a stable state during the training process. If the model error continues to decrease and stabilize during the training process, it means that the model has reached a converged state. If the model error fluctuates greatly, it means that it has not yet converged and further training is needed. Parallel integration refers to combining multiple trained evaluation branches into a comprehensive model, using the advantages of each branch to improve the performance of the overall model.

[0075] First, a set of sample data is extracted from multiple sets of sample data and divided into a training set and a validation set. The training set is used to learn the model, while the validation set is used to test the performance of the model. Commonly used tools such as data segmentation tools (such as the train_test_split function in scikit-learn) can help divide the sample data in proportion.

[0076] Next, the system trains the first evaluation branch based on the training set. This process is usually implemented through machine learning algorithms (such as linear regression, neural networks, etc.). The algorithm learns the data patterns from the training set and gradually adjusts the model parameters to optimize its performance on the training set. After training is completed, the system uses the validation set to verify the convergence of the first evaluation branch. Convergence testing is usually performed through the validation loss function. If the validation loss of the model tends to be stable during the training process, it means that the model has converged.

[0077] If the verification result shows that the convergence of the model does not meet the preset requirements, the system will re-partition the sample data and train the model again. The purpose of re-partitioning is to adjust the distribution of the training set and the verification set so that the model can better learn and converge under the new data combination. This process needs to be repeated many times until the model reaches the ideal convergence state.

[0078] The system then traverses multiple sets of sample data and completes the training of each evaluation branch one by one. Each evaluation branch focuses on a different subtask or different data characteristics in order to make full use of all sample data.

[0079] When all evaluation branches are trained, the system will integrate them in parallel. The purpose of parallel integration is to combine the prediction results of multiple sub-models to form a more powerful comprehensive model. This process can be achieved through ensemble learning techniques (such as random forests, gradient boosting trees, etc.) to ensure that the advantages of each branch model are fully utilized.

[0080] Ultimately, the system obtained a comprehensive photovoltaic evaluation model, which can accurately evaluate the performance of the photovoltaic system under different conditions and provide a scientific basis for the optimization and maintenance of the system.

[0081] Furthermore, based on the photovoltaic analysis results, a photovoltaic system adjustment plan is determined, including: traversing the photovoltaic analysis results to determine pre-adjustment points; clustering the pre-adjustment points based on the adjustment type, and configuring adjustment point plans; performing adjustment collision analysis and priority sorting on the pre-adjustment points to determine the adjustment order; and generating the photovoltaic system adjustment plan based on the adjustment order and the adjustment point plan.

[0082] Specifically, pre-regulation points are key nodes or devices identified in the PV analysis results that need to be adjusted or controlled. For example, a certain inverter is identified as a pre-regulation point that needs to have its operating parameters adjusted. Regulation type refers to the category of specific regulation measures taken during the regulation process. Regulation types include equipment parameter adjustment, fault repair, energy distribution optimization, etc. Regulation collision analysis refers to the analysis of potential conflicts or mutual influences between different regulation operations. Collision analysis ensures that multiple regulation operations do not interfere with each other or lead to undesirable results.

[0083] First, the photovoltaic analysis results are traversed to determine the pre-regulation points that need to be regulated. Pre-regulation points are usually key devices or nodes in the photovoltaic system, and the analysis results show that the performance of these nodes needs to be adjusted or optimized.

[0084] Next, the system will cluster these pre-control points based on the control type. The purpose of clustering is to group pre-control points with similar control requirements for unified control. For example, all pre-control points that need to adjust the inverter output power can be divided into one group, while points that need to replace cables can be divided into another group. Clustering algorithms (such as K-means clustering or hierarchical clustering) can help effectively group pre-control points according to their control type.

[0085] After clustering is completed, the system generates a control point plan for each cluster. The control point plan is a specific action plan for the control points in each group, which describes how to adjust, repair or optimize them.

[0086] Next, the system will perform control collision analysis and priority sorting. The purpose of control collision analysis is to identify conflicting control operations and ensure that different operations do not interfere with each other. For example, if two control operations involve the same photovoltaic module at the same time, the system needs to analyze the relationship between them and adjust the execution order. Conflict detection tools and priority sorting algorithms can help complete this step. The basis for sorting includes the urgency of the control and the degree of impact on the system.

[0087] Finally, based on the control order and control point plan, the system generates the final PV system control plan. The control plan is a comprehensive operation plan that details the adjustment method, execution time and execution order of each control point. The system will use scheduling algorithms or optimization tools to ensure that the execution of the control plan can achieve the best results.

[0088] In summary, the distributed photovoltaic new energy data management method based on cloud computing provided in the embodiments of the present application has the following technical effects:

[0089] 1. By traversing the photovoltaic system, building a topological architecture and configuring a data acquisition array, claim 1 achieves the technical effect of comprehensively monitoring the operating status of the photovoltaic system. The data is transmitted to the cloud computing platform for modular storage and evaluation, and ultimately the operating efficiency and reliability of the system are improved by formulating adjustment plans and optimizing control management.

[0090] 2. By building a photovoltaic evaluation model into the cloud computing platform, clustering processing and data-driven modeling are performed based on data management needs and sample evaluation records, which enables accurate evaluation and optimization of photovoltaic system performance, and enhances the system's intelligent analysis capabilities and decision-making support.

[0091] 3. By traversing the photovoltaic analysis results, clustering pre-control points, performing collision analysis and priority sorting, a photovoltaic system adjustment plan is finally generated, which realizes the refined management and optimized control of the photovoltaic system, ensures the rationality and efficiency of system adjustment, and improves the overall operation performance.

[0092] Embodiment 2 is based on the same inventive concept as the distributed photovoltaic new energy data management method based on cloud computing in the above embodiment. Figure 2 As shown, the embodiment of the present application provides a distributed photovoltaic new energy data management system based on cloud computing, the system comprising:

[0093] Building a photovoltaic topology architecture module 11: used to traverse the distributed photovoltaic system and build a photovoltaic topology architecture;

[0094] Configuration data acquisition array module 12: used to read the operation records of the preset time zone of the distributed photovoltaic system, mine multi-level data topology nodes, and configure the data acquisition array, wherein the data acquisition array includes multiple types of sensor devices;

[0095] The photovoltaic data determination module 13 is used to control the data acquisition array, perform sensor monitoring of the distributed photovoltaic system, and determine photovoltaic data, wherein the photovoltaic data carries a data topology node identifier and a time-space identifier;

[0096] A photovoltaic analysis result determination module 14 is used to assist the transfer module, transmit the photovoltaic data to the cloud computing platform, perform data modular storage and photovoltaic operation and maintenance evaluation, and determine the photovoltaic analysis results, wherein the cloud computing platform has a built-in photovoltaic evaluation model, and the cloud computing platform establishes a communication connection with the data acquisition array;

[0097] A photovoltaic system adjustment plan determination module 15 is used to determine a photovoltaic system adjustment plan based on the photovoltaic analysis result;

[0098] Terminal visualization display module 16: used for performing terminal visualization display of the photovoltaic analysis results and the photovoltaic system adjustment plan, and delegating the photovoltaic system adjustment plan to perform optimization, regulation and management of the distributed photovoltaic system.

[0099] Furthermore, the system is also used to perform the following steps:

[0100] Traversing the operation records to determine valid monitoring nodes;

[0101] Based on the data criticality level, traverse the valid monitoring nodes and divide the data nodes into multiple levels;

[0102] Based on data relevance, traverse the valid monitoring nodes and divide the data batches, wherein the data relevance includes synchronization relevance and time relevance;

[0103] The multi-level data nodes are fitted with the data batches to determine the multi-level data topology nodes.

[0104] Furthermore, the system is also used to perform the following steps:

[0105] Traversing the valid monitoring nodes, and configuring node monitoring devices based on node monitoring types, wherein each valid monitoring node corresponds to a monitoring device of at least one monitoring type;

[0106] Based on the node monitoring device, it is assembled in the distributed photovoltaic system, and upper management integration is performed to determine the data acquisition array.

[0107] Furthermore, the system is also used to perform the following steps:

[0108] Transmitting the photovoltaic data to the auxiliary transfer module, performing data preprocessing, and determining effective photovoltaic data;

[0109] The valid photovoltaic data are traversed to perform timestamp constraints on data transmission.

[0110] Furthermore, the system is also used to perform the following steps:

[0111] Determine data management requirements and mine sample evaluation records based on the operational records;

[0112] Traversing the data management requirements, performing clustering processing on the sample evaluation records based on the requirements similarity, and determining multiple groups of samples;

[0113] Based on the multiple groups of samples and the data management requirements, data-driven modeling is performed to generate the photovoltaic evaluation model.

[0114] Furthermore, the system is also used to perform the following steps:

[0115] Based on the multiple groups of samples, extract a group of sample data and divide it into a training set and a validation set;

[0116] Training a first evaluation branch based on the training set, and verifying the convergence of the first evaluation branch based on the validation set;

[0117] If the convergence degree does not meet the preset convergence degree, re-dividing and training the set of sample data;

[0118] Traversing the multiple groups of sample data to complete the training of the Nth evaluation branch;

[0119] The first evaluation branches up to the Nth evaluation branch are integrated in parallel to obtain the photovoltaic evaluation model.

[0120] Furthermore, the system is also used to perform the following steps:

[0121] Traversing the photovoltaic analysis results to determine a pre-control point;

[0122] Performing clustering processing on the pre-control points based on the control type, and configuring control point plans;

[0123] Performing control collision analysis and priority sorting on the pre-control points to determine the control order;

[0124] Based on the control sequence and the control point plan, the photovoltaic system regulation plan is generated.

[0125] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0126] Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and / or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A distributed photovoltaic new energy data management method based on cloud computing, characterized in that: The method comprises: Traverse the distributed photovoltaic system and build the photovoltaic topology architecture; Reading the operation records of the preset time zone of the distributed photovoltaic system, mining multi-level data topology nodes, and configuring a data acquisition array, wherein the data acquisition array includes multiple types of sensor devices; Controlling the data acquisition array to perform sensor monitoring of the distributed photovoltaic system and determine photovoltaic data, wherein the photovoltaic data carries a data topology node identifier and a time-space identifier; An auxiliary transfer module transmits the photovoltaic data to a cloud computing platform, performs modular data storage and photovoltaic operation and maintenance evaluation, and determines photovoltaic analysis results, wherein the cloud computing platform has a built-in photovoltaic evaluation model, and a communication connection is established between the cloud computing platform and the data acquisition array; Based on the photovoltaic analysis results, determine a photovoltaic system adjustment plan; The photovoltaic analysis results and the photovoltaic system adjustment plan are displayed visually on the terminal, and the photovoltaic system adjustment plan is decentralized to perform optimized regulation and management of the distributed photovoltaic system.

2. The distributed photovoltaic new energy data management method based on cloud computing according to claim 1, characterized in that: The mining of multi-level data topology nodes includes: Traversing the operation records to determine valid monitoring nodes; Based on the data criticality level, traverse the valid monitoring nodes and divide the data nodes into multiple levels; Based on data relevance, traverse the valid monitoring nodes and divide the data batches, wherein the data relevance includes synchronization relevance and time relevance; The multi-level data nodes are fitted with the data batches to determine the multi-level data topology nodes.

3. The distributed photovoltaic new energy data management method based on cloud computing according to claim 2, characterized in that: The configuration data acquisition array comprises: Traversing the valid monitoring nodes, and configuring node monitoring devices based on node monitoring types, wherein each valid monitoring node corresponds to a monitoring device of at least one monitoring type; Based on the node monitoring device, it is assembled in the distributed photovoltaic system, and upper management integration is performed to determine the data acquisition array.

4. The distributed photovoltaic new energy data management method based on cloud computing according to claim 1, characterized in that: The auxiliary transfer module comprises: Transmitting the photovoltaic data to the auxiliary transfer module, performing data preprocessing, and determining effective photovoltaic data; The valid photovoltaic data are traversed to perform timestamp constraints on data transmission.

5. The distributed photovoltaic new energy data management method based on cloud computing according to claim 1, characterized in that: The cloud computing platform has a built-in photovoltaic evaluation model, including: Determine data management requirements and mine sample evaluation records based on the operational records; Traversing the data management requirements, performing clustering processing on the sample evaluation records based on the requirements similarity, and determining multiple groups of samples; Based on the multiple groups of samples and the data management requirements, data-driven modeling is performed to generate the photovoltaic evaluation model.

6. The distributed photovoltaic new energy data management method based on cloud computing according to claim 5, characterized in that: The generating the photovoltaic evaluation model comprises: Based on the multiple groups of samples, extract a group of sample data and divide it into a training set and a validation set; Training a first evaluation branch based on the training set, and verifying the convergence of the first evaluation branch based on the validation set; If the convergence degree does not meet the preset convergence degree, re-dividing and training the set of sample data; Traversing the multiple groups of sample data to complete the training of the Nth evaluation branch; The first evaluation branches up to the Nth evaluation branch are integrated in parallel to obtain the photovoltaic evaluation model.

7. The distributed photovoltaic new energy data management method based on cloud computing according to claim 1, characterized in that: Based on the photovoltaic analysis results, a photovoltaic system adjustment plan is determined, including: Traversing the photovoltaic analysis results to determine a pre-control point; Performing clustering processing on the pre-control points based on the control type, and configuring control point plans; Performing control collision analysis and priority sorting on the pre-control points to determine the control order; Based on the control sequence and the control point plan, the photovoltaic system regulation plan is generated.

8. A distributed photovoltaic new energy data management system based on cloud computing, characterized in that: For executing the method according to any one of claims 1 to 7, the system comprises: Build photovoltaic topology architecture module: used to traverse the distributed photovoltaic system and build photovoltaic topology architecture; Configuring a data acquisition array module: used to read the operation records of the preset time zone of the distributed photovoltaic system, mine multi-level data topology nodes, and configure a data acquisition array, wherein the data acquisition array includes multiple types of sensor devices; A photovoltaic data determination module is used to control the data acquisition array, perform sensor monitoring of the distributed photovoltaic system, and determine photovoltaic data, wherein the photovoltaic data carries a data topology node identifier and a time-space identifier; A photovoltaic analysis result determination module is used to assist the transfer module, transmit the photovoltaic data to the cloud computing platform, perform data modular storage and photovoltaic operation and maintenance evaluation, and determine the photovoltaic analysis results, wherein the cloud computing platform has a built-in photovoltaic evaluation model, and the cloud computing platform has a communication connection with the data acquisition array; A module for determining a photovoltaic system adjustment plan: used to determine a photovoltaic system adjustment plan based on the photovoltaic analysis result; Terminal visualization display module: used for terminal visualization display of the photovoltaic analysis results and the photovoltaic system adjustment plan, and for delegating the photovoltaic system adjustment plan to perform optimization, regulation and management of the distributed photovoltaic system.

Images