Energy storage configuration method and system for photovoltaic power generation

Abstract

The invention relates to the technical field of energy storage configuration, in particular to an energy storage configuration method and system for photovoltaic power generation. The method comprises the following steps: acquiring information data of a photovoltaic power generation energy station; performing distributed synchronous energy storage network construction on the photovoltaic power generation energy station information data to obtain a distributed synchronous energy storage network; performing real-time power generation data acquisition on the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; performing photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; performing historical electricity load data acquisition on the distributed synchronous energy storage network to obtain historical electricity load data; and performing load curve curvature trend analysis on the historical electricity load data to generate load curve curvature trend data. According to the invention, through real-time photovoltaic power generation prediction, load demand dynamic analysis and adaptive energy storage strategy optimization, the accuracy and flexibility of photovoltaic energy storage configuration are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage configuration, and in particular to an energy storage configuration method and system for photovoltaic power generation. Background Art

[0002] In the early days, the combination of photovoltaic power generation and energy storage systems mainly relied on traditional battery energy storage technology, such as lead-acid batteries and nickel-metal hydride batteries. However, these batteries have disadvantages such as low energy density, short life, and poor charging and discharging efficiency, which limit their application in large-scale photovoltaic power stations. With the advancement of materials science and electrochemical technology, lithium batteries have become an important energy storage solution. They have high energy density, long life, and fast charging and discharging characteristics, and have gradually become a widely used energy storage method in photovoltaic power generation systems. With the development of power electronics technology, intelligent energy storage systems have gradually emerged. The use of advanced current conversion technology, intelligent control algorithms and integrated equipment can effectively optimize the efficiency and reliability of energy storage systems. In addition, new energy storage technologies such as flow batteries and compressed air energy storage are also emerging, providing more options for photovoltaic power generation. At present, the integration of photovoltaic power generation and energy storage technology has gradually entered a systematic and intelligent development stage. In particular, with the application of big data, the Internet of Things and artificial intelligence technologies, the real-time monitoring and optimization management capabilities of photovoltaic energy storage systems have been greatly enhanced, which has promoted the further development of the photovoltaic power generation industry. However, the current traditional photovoltaic energy storage system fails to flexibly adjust according to the real-time energy supply and demand ratio. At the same time, energy storage performance testing and strategy adjustments are often relatively lagging, making it difficult to adapt to rapidly changing electricity consumption and power generation conditions, resulting in low accuracy and flexibility in photovoltaic energy storage configuration. Summary of the invention

[0003] Based on this, it is necessary to provide an energy storage configuration method and system for photovoltaic power generation to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for configuring energy storage for photovoltaic power generation is provided, the method comprising the following steps: Step S1: Acquire photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network for the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data for the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; Step S2: Collect historical power load data of the distributed synchronous energy storage network to obtain historical power load data; perform load curve curvature trend analysis on the historical power load data to generate load curve curvature trend data; classify load demand of the historical power load data based on the load curve curvature trend data to generate power load demand pattern data; Step S3: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to generate an energy storage configuration management scheduling strategy; Step S4: Perform energy storage performance testing on the distributed synchronous energy storage network through energy storage configuration management scheduling strategy to generate energy storage performance test data; perform adaptive energy storage strategy adjustment according to the energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

[0005] The present invention collects real-time data of photovoltaic power generation sites and establishes a distributed synchronous energy storage network, which can ensure that the photovoltaic power generation system can monitor and coordinate each energy storage unit in real time, ensure the real-time and accuracy of the data, improve the response speed and coordination ability of the system, and provide basic data support for subsequent power generation prediction and energy storage optimization. By collecting and analyzing historical power load data, the changing trend of load demand can be accurately understood, and the fluctuation law of load demand can be identified through the curvature trend analysis of the load curve. This process helps to classify the power load more accurately, so as to better predict the future power demand pattern, ensure that the energy storage system can match the load demand, and achieve more efficient energy utilization. By combining real-time photovoltaic power generation prediction data with power load demand pattern data, the energy supply and demand ratio is calculated, thereby providing an optimization basis for the configuration of the energy storage system. According to the changes in the energy supply and demand ratio, the distributed synchronous energy storage network is dynamically configured and optimized, which can achieve a balance between energy supply and demand, improve energy storage efficiency, avoid excessive energy storage or energy waste, and ensure the efficient operation of the photovoltaic power generation system. By testing the performance of the energy storage system and making adaptive adjustments based on the test results, the energy storage strategy can be adjusted in real time to ensure that the energy storage system can operate efficiently under different load and power generation conditions. This step makes the energy storage system more flexible and intelligent, and can adjust the configuration and strategy according to the actual operation data, improve the reliability, stability and life of the system, and optimize the energy management of the photovoltaic power generation system. Therefore, the present invention improves the accuracy and flexibility of photovoltaic energy storage configuration through real-time photovoltaic power generation prediction, dynamic analysis of load demand and adaptive energy storage strategy optimization.

[0006] Preferably, step S1 comprises the following steps: Step S11: Acquire photovoltaic power generation energy site information data; Step S12: performing site distribution topology analysis on the photovoltaic power generation energy site information data to generate energy site distribution topology data; constructing a distributed synchronous energy storage network for the photovoltaic power generation energy site information data based on the energy site distribution topology data to obtain a distributed synchronous energy storage network; Step S13: collecting real-time power generation data of the distributed synchronous energy storage network to obtain real-time photovoltaic power generation data; performing data preprocessing on the real-time photovoltaic power generation data to generate standard real-time photovoltaic power generation data, wherein the data preprocessing includes data cleaning, data denoising, missing value filling and data standardization; Step S14: Perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data.

[0007] The present invention can clearly understand the location of each site and its relationship with other sites by performing a distributed topological analysis on photovoltaic power generation energy sites. This is crucial for building a distributed synchronous energy storage network. Through this network, the transmission and storage of electric energy can be optimized in real time, electric energy waste can be avoided, and energy utilization can be improved. Real-time photovoltaic power generation data acquisition enables the entire system to understand the power generation situation of the photovoltaic site in the first time, which is convenient for rapid response to any fluctuations in power demand. Data preprocessing (such as cleaning, denoising, missing value filling, etc.) helps to improve the quality of data, ensure the accuracy of subsequent analysis and prediction, and avoid decision-making errors caused by data anomalies or errors. Photovoltaic power generation prediction based on standardized real-time photovoltaic power generation data can not only improve the prediction accuracy of future power generation, but also provide a scientific basis for the power dispatching system, make energy distribution and reserves in advance, especially when the weather changes or the demand for electricity increases suddenly, it can more accurately adjust the balance between energy supply and demand. This process realizes the real-time collection, processing and prediction of energy site information, combined with advanced distributed energy storage network management, can improve the intelligence level of the system, so that it can automatically adjust the operating state according to real-time data and prediction results, and maximize the efficiency and stability of energy supply.

[0008] Preferably, constructing a distributed synchronous energy storage network for photovoltaic power generation energy site information data based on energy site distribution topology data includes: Based on the graph theory method, the energy site distribution topology data is topologically modeled to generate energy site topology model data; the energy site topology model data is node-importantly evaluated to generate topology node importance data; the energy site topology model data is node-labeled according to the topology node importance data to generate key nodes and secondary nodes; The energy site topology model data is edge-connected according to the joint points and secondary nodes to generate an initial energy storage network; the initial energy storage network is deployed with distributed nodes to generate a distributed energy storage network; the distributed energy storage network is synchronized with three layers of data using the photovoltaic power generation energy site information data to generate a distributed synchronous energy storage network, where the three layers of data synchronization include time synchronization, power synchronization, and state synchronization.

[0009] The present invention can more accurately describe the relationship between energy sites and reveal the structure and functionality of different sites in the entire network by modeling the energy site distribution topology based on graph theory. Node importance assessment further helps identify sites (key nodes) that are critical to the entire energy storage network, as well as relatively minor sites. This assessment enables the energy storage network to concentrate resources, prioritize key nodes, and avoid system overcomplication or waste of resources. Key nodes and minor nodes are labeled to make the design of the energy storage network more targeted. Edge connections are made according to the roles of nodes to generate an initial energy storage network, laying the foundation for subsequent distributed deployment. Such a design can better ensure the connectivity of the network and the stable flow of energy, and ensure that energy storage resources can be allocated according to actual needs. By deploying distributed nodes in the network, energy storage and release functions can be provided at different locations, reducing the risk of single point failures and improving the robustness and stability of the entire system. The distributed energy storage network enables each photovoltaic power generation site to be flexibly adjusted according to local needs and optimize overall energy management. Ensure that each node in the energy storage network works under a unified time frame to avoid scheduling errors or data inconsistencies caused by time differences. Time synchronization improves the system response speed and accuracy. Through power synchronization, it can ensure that the power output between each energy storage node is coordinated under the same standard, avoid overcharging or discharging, and improve the efficiency and life of the energy storage system. The operating status of each node is synchronized to ensure that the entire network operates under the same operating conditions, maximizing the efficiency and stability of the energy storage network. The distributed synchronous energy storage network design can be scheduled and optimized in real time according to different time, demand, power and state conditions, enhancing the system's adaptability to fluctuations in power demand, failures or unexpected events. The system can flexibly respond to a variety of operating conditions and improve the accuracy and efficiency of overall energy management.

[0010] Preferably, step S14 includes the following steps: Step S141: extracting basic photovoltaic power generation characteristics from standard real-time photovoltaic power generation data to obtain basic photovoltaic power generation characteristic data; performing time series characteristic analysis on the basic photovoltaic power generation characteristic data to generate photovoltaic power generation time series characteristic data; Step S142: dividing the photovoltaic power generation time series characteristic data into data sets to generate a model training set and a model test set; training the model training set using a long short-term memory neural network algorithm to generate a photovoltaic power generation prediction pre-model; optimizing and iterating the photovoltaic power generation prediction pre-model using the model test set to generate a photovoltaic power generation prediction model; Step S143: Importing the standard real-time photovoltaic power generation data into the photovoltaic power generation prediction model to perform photovoltaic power generation prediction and generate real-time photovoltaic power generation prediction data.

[0011] The present invention provides high-quality input data for subsequent prediction models by extracting key power generation characteristics (such as sunshine intensity, temperature, humidity, etc.) from standard real-time photovoltaic power generation data. This step helps to reduce the interference of irrelevant factors on the prediction results and ensure the accuracy and reliability of the model. By performing time series analysis on the basic photovoltaic power generation characteristic data, the regularity and periodic characteristics of photovoltaic power generation data changing over time (such as seasonal fluctuations, day and night changes, etc.) are extracted. These time series characteristics are crucial factors in photovoltaic power generation prediction. Accurately capturing time series characteristics helps to improve the long-term prediction ability of the prediction model, especially when facing seasonal fluctuations. By dividing the photovoltaic power generation time series characteristic data into a training set and a test set, the data independence in the training process can be guaranteed to avoid overfitting. The training set is used for model learning, and the test set is used to evaluate the generalization ability of the model to ensure the actual application effect of the model. LSTM is a commonly used time series prediction model that can effectively process the time dependency in photovoltaic power generation data and capture long-term time dependency and short-term change trends. Training the training set with the LSTM algorithm allows the model to recognize and learn complex time series patterns in photovoltaic power generation data, thereby improving the prediction accuracy. By optimizing and iterating the prediction pre-model on the model test set, the model can gradually adjust and correct its parameters to reduce the prediction error. This process can effectively improve the accuracy and robustness of the model, especially when dealing with different weather, seasons or sudden changes, so that the model can maintain a high prediction accuracy in different environments. The model can continuously adapt to new input data and environmental changes during the iteration process. As more historical data is accumulated, the prediction accuracy and adaptability of the model will continue to improve. This enables the system to better cope with photovoltaic power generation prediction tasks in different regions and under different climatic conditions. By inputting standard real-time photovoltaic power generation data into the photovoltaic power generation prediction model for prediction, it can provide real-time estimates of current and future power generation. In this way, the photovoltaic power generation system can dynamically adjust power distribution according to the prediction data, optimize energy storage and discharge strategies, thereby improving the stability of the power grid and the balance of power supply and demand.

[0012] Preferably, step S2 comprises the following steps: Step S21: collecting historical power load data of the distributed synchronous energy storage network to obtain historical power load data; Step S22: extracting spatiotemporal load characteristics from historical power load data to obtain spatiotemporal power load characteristic data; analyzing seasonal variation patterns of the spatiotemporal power load characteristic data to generate a load characteristic data set; converting historical power load data into a historical power load curve according to the load characteristic data set to generate a historical power load curve; Step S23: performing load curve curvature trend analysis on the historical power load curve to generate load curve curvature trend data, wherein the load curve curvature trend data includes positive trend curvature, gentle trend curvature and negative trend curvature; Step S24: Segment the historical power load curve based on the positive trend curvature, the gentle trend curvature and the negative trend curvature to generate a high load trend curve, a gentle load trend curve and a low load trend curve; classify the load demand of the historical power load data according to the high load trend curve, the gentle load trend curve and the low load trend curve to generate power load demand pattern data.

[0013] By collecting historical power load data, the present invention can fully understand the change trend and periodic characteristics of the load, and provide basic data for subsequent analysis and decision-making. Historical data provides the model with real power consumption behavior and ensures the reliability of the analysis. By extracting spatiotemporal features from historical load data, the change law of power load over time (such as day, week, season) and space (such as different regions and different user groups) can be revealed. This helps to gain a deeper understanding of the dynamic changes in power demand and supports more accurate load forecasting. By analyzing the seasonal change law of power load, the volatility of load in different seasons can be identified, and the impact of seasonal changes on power load can be accurately captured. For example, air conditioning load increases in summer and heating load increases in winter. Accurately grasping these laws helps to make load scheduling in advance and optimize energy storage and power generation plans. The load characteristic data set generated by seasonal analysis provides the system with a more detailed description of the load pattern, which helps to make more accurate load forecasts in different seasons or weather conditions. By performing curve conversion on historical power load data, the original load data can be converted into a more regular load curve. This conversion helps the system identify load change trends and peak and valley fluctuations, and further supports subsequent trend analysis and classification. By analyzing the curvature trend of the load curve, different trends of load changes can be clearly identified: positive trend (load increase), gentle trend (load stability) and negative trend (load reduction). This provides important guidance information for subsequent load demand forecasting and helps predict future trends in electricity demand. By segmenting the load curve, the load is divided into three trends: high load, gentle load and low load, which helps to analyze different load scenarios independently. This segmentation can help the system adopt different energy storage and power dispatching strategies under different load conditions, improving the flexibility and adaptability of dispatching. Based on the segmentation of the load curve, the system can classify load demand patterns to form clear electricity load demand pattern data. These pattern data provide a basis for power dispatching and energy storage management, and can implement the optimal load forecasting and dispatching scheme according to different modes to avoid power waste and improve energy efficiency.

[0014] Preferably, step S3 comprises the following steps: Step S31: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; wherein the formula for calculating the energy supply and demand ratio is as follows:

[0015] In the formula, Expressed as the energy supply-demand ratio, Expressed as time The photovoltaic power generation power, Expressed as time The power output of the energy storage device, Expressed as time Electricity obtained from the grid, Expressed as time The power demand of the electrical load; Step S32: comparing the energy supply-demand ratio data with a preset standard supply-demand ratio threshold, when the energy supply-demand ratio data is greater than the preset standard supply-demand ratio threshold, a power supply surplus mode is generated; when the energy supply-demand ratio data is less than the preset standard supply-demand ratio threshold, a power supply gap mode is generated; Step S33: Optimizing the energy storage system configuration of the distributed synchronous energy storage network based on the power supply surplus mode and the power supply gap mode, and generating photovoltaic power generation energy storage configuration optimization data; Step S34: Dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data, thereby generating an energy storage configuration management scheduling strategy.

[0016] The present invention calculates the energy supply and demand ratio R(t), and the system can evaluate the supply and demand balance of energy in real time. The calculation formula integrates the photovoltaic power generation power, the output of the energy storage device, the power obtained by the power grid and the power load demand, ensuring a comprehensive consideration of the supply and demand situation. This comprehensive calculation method helps to understand the energy supply situation in real time and provide more accurate basic data for subsequent decision-making. By combining the power output of photovoltaic power generation, energy storage system and power grid, the formula can accurately reflect the contribution of each energy source in meeting the power load demand. This can not only help identify the shortage of energy supply, but also effectively evaluate the role of energy storage and power grid under different load conditions. By comparing the energy supply and demand ratio calculated in real time with the preset standard supply and demand ratio threshold, the system can automatically generate a power supply surplus mode or a power supply gap mode. This automatic identification mechanism helps to quickly respond to the difference between power load demand and energy supply, and provides timely feedback for subsequent energy management and scheduling decisions. When the energy supply and demand ratio is greater than the preset threshold, it indicates that the system has a power supply surplus. At this time, the system can store excess energy to avoid energy waste or over-reliance on the power grid, and enhance the energy self-sufficiency of the system. When the energy supply-demand ratio is lower than the preset threshold, it indicates that the energy supply is insufficient and a power supply gap occurs. At this time, the system will trigger the corresponding adjustment strategy to activate the backup power supply or supplement energy to ensure that the user's power demand is met. Based on the power supply surplus mode and the power supply gap mode, the system can optimize the energy storage system configuration of the distributed synchronous energy storage network. This optimization process can ensure that the energy storage equipment has sufficient power storage capacity during peak load periods and does not waste too much power during low demand periods. Optimized configuration helps to improve the economy and efficiency of the energy storage system and reduce the situation of excessive or insufficient energy storage. By optimizing the configuration of the energy storage system, the system can ensure the efficient storage and release of energy, avoid overcharging of energy storage equipment when there is a power surplus or insufficient energy storage when there is a power supply gap, and improve the overall utilization of the energy storage system. Energy storage configuration optimization can be dynamically adjusted according to different power supply surplus modes and power supply gap modes to ensure that the energy storage system can achieve maximum benefits under different load demands. For example, when the load demand is high, the energy storage equipment can provide additional support to the system; when the load demand is low, the energy storage system can back up future load demand by charging. Through the generated energy storage configuration management and scheduling strategy, the system can dynamically adjust energy storage configuration and energy scheduling based on real-time data. This dynamic adjustment mechanism can flexibly respond to different power demand situations according to real-time changes in load demand, photovoltaic power generation, energy storage status and grid conditions, thereby improving the system's response speed and adjustment capabilities.

[0017] Preferably, step S33 includes the following steps: Step S331: confirming the mode time period of the distributed synchronous energy storage network based on the power supply surplus mode and the power supply gap mode to obtain the power supply surplus time period and the power supply gap time period; performing a buffer time period analysis on the power supply surplus time period and the power supply gap time period to generate a power supply change buffer time period; Step S332: setting energy discharge for the power surplus time period to generate an energy discharge curve; setting energy charging for the power shortage time period to generate an energy charging curve; fitting the energy discharge curve and the energy charging curve according to the power supply change buffer period to generate energy loss data for the buffer period; Step S333: Optimize the configuration of energy storage batteries in the buffer period of the distributed synchronous energy storage network based on the energy loss data in the buffer period, thereby generating photovoltaic power generation energy storage configuration optimization data.

[0018] The present invention confirms the mode time period of the power supply surplus mode and the power supply gap mode, and the system can accurately identify the power supply status of different time periods. The confirmation of the power supply surplus time period and the power supply gap time period provides a clear time frame for the subsequent energy storage scheduling, so that the system can adopt different adjustment strategies in different time periods. This time period confirmation mechanism can respond to the fluctuations of photovoltaic power generation and power load in real time, and flexibly adjust the charging and discharging strategy of the energy storage device. The energy storage device can be charged when the power supply is surplus, and can be discharged when the power supply is short, ensuring the stability and economy of the power supply. There is a certain buffer period between the power supply surplus and the gap. By analyzing these buffer periods, the system can effectively evaluate the transition period of energy and perform fine energy regulation during this period. This process helps alleviate power supply fluctuations, avoids over-reliance on the power grid or energy storage system, and ensures a smooth transition of the energy system. The buffer period analysis can discover potential losses in the process of energy discharge or charging, optimize the energy flow in the discharge and charging process, and reduce unnecessary energy loss. This can further improve the use efficiency of the energy storage device and avoid energy waste during the transition period. By setting energy discharge for power surplus time periods and energy charging for power shortage time periods, the system can generate accurate energy discharge and charging curves. These curves are dynamically adjusted based on real-time load demand and changes in photovoltaic power generation, ensuring that energy storage devices can discharge and charge efficiently. During the buffer period, the energy discharge and charging curves are fitted to generate energy loss data for the buffer period. This process helps the system identify energy losses during the buffer period and minimize such losses through optimization strategies to improve the overall efficiency of the system. By generating energy loss data for the buffer period, the system can quantify the energy losses that occur during the scheduling process. This data can be used to further optimize the charging and discharging strategy, especially to reduce unnecessary energy losses during the adjustment process of the buffer period. By optimizing the energy storage battery configuration based on the energy loss data for the buffer period, the system can achieve more refined allocation of energy storage battery capacity and charging and discharging capabilities. This optimization ensures that energy storage devices can work efficiently under different supply and demand modes, avoids overcharging or discharging, and improves the battery life and energy storage efficiency.

[0019] Preferably, step S34 includes the following steps: Step S341: performing node energy redundancy calculation on the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to obtain node energy redundancy data; performing adjacent node status analysis on the distributed synchronous energy storage network according to the node energy redundancy data to generate adjacent node status data; Step S342: using the adjacent node status data to perform energy scheduling weight allocation on the node energy redundancy data to generate energy scheduling weight allocation data; Step S343: perform emergency energy node screening on adjacent node status data through energy scheduling weight distribution data to obtain emergency energy node data; perform emergency scheduling energy storage path analysis on the emergency energy node data according to node energy redundancy data to generate emergency node scheduling energy storage path data; Step S344: Dispatching and uniformly controlling the energy storage of the distributed synchronous energy storage network based on the emergency node scheduling energy storage path data, thereby generating an energy storage configuration management scheduling strategy.

[0020] The present invention can analyze the remaining energy reserves of each node through node energy redundancy calculation to ensure that the energy supply of each node in the system is not insufficient. This process helps to discover the excess or shortage of resources in the network and adjust the energy storage strategy in time. By analyzing the status of adjacent nodes, the system can evaluate the energy sharing and support relationship with adjacent nodes. The generation of adjacent node status data can help the system understand the energy flow status between nodes, provide a reference for subsequent energy scheduling, and ensure the efficient allocation and utilization of energy in the network. Using the adjacent node status data, the node energy redundancy is dispatched with weight allocation, and the energy scheduling priority between nodes can be dynamically adjusted according to the energy sufficiency of each node and the energy demand in the network. Through this weight allocation, the system can flexibly schedule between different nodes and optimize the discharge and charging strategies of energy storage equipment. Through the energy scheduling weight allocation data and the adjacent node status data, emergency nodes with sudden increase in energy demand and energy shortage can be screened out. These emergency nodes are caused by factors such as insufficient photovoltaic power generation and grid load fluctuations, and need to be scheduled and optimized first. The energy storage path analysis of the emergency energy node is performed through the node energy redundancy data to generate the energy storage scheduling path data of the emergency node. This process helps the system accurately plan the energy dispatch path from different energy storage nodes to emergency nodes, ensuring that emergency nodes can quickly obtain energy support and avoid unstable system operation due to energy shortages. Based on the emergency node dispatch energy storage path data, the distributed synchronous energy storage network can implement energy storage uniform control, that is, balance the allocation of energy storage capacity between different nodes to avoid overloading or excess resources at a single node. This balanced control helps improve the overall efficiency and stability of the system and ensure that energy storage equipment can reach the best state under various supply and demand conditions.

[0021] Preferably, step S4 comprises the following steps: Step S41: executing the distributed synchronous energy storage network strategy through the energy storage configuration management scheduling strategy to generate photovoltaic power generation energy storage strategy execution data; performing energy storage performance test on the photovoltaic power generation energy storage strategy execution data to generate energy storage performance test data; Step S42: Adaptively adjust the photovoltaic power generation energy storage strategy execution data according to the energy storage performance test data to perform the photovoltaic power generation energy storage configuration optimization operation.

[0022] The present invention can dynamically adjust the working mode of the energy storage system according to the real-time status and demand of the distributed synchronous energy storage network by executing the energy storage configuration management scheduling strategy. Through the execution of this strategy, the energy storage and utilization efficiency of the photovoltaic power generation system can be maximized, and the stable energy supply of the system under different load requirements can be ensured. The execution of the energy storage strategy can not only ensure the timely storage and release of photovoltaic power generation power, but also coordinate the energy flow between photovoltaic power generation, energy storage equipment and power grid, optimize the utilization efficiency of energy, thereby reducing energy waste and system operation costs. By testing the energy storage performance, various performance indicators of the energy storage system can be monitored in real time, including energy storage efficiency, discharge capacity, charging speed and system response speed. This testing process ensures that the energy storage system can operate stably under different working conditions and promptly discovers potential performance bottlenecks. The generation of energy storage performance test data can help the system identify weak links in the energy storage process, optimize or adjust the links that do not meet the performance requirements in a timely manner, and improve the stability and reliability of the entire system. Adaptive adjustment of the energy storage strategy based on the energy storage performance test data can ensure that the energy storage system is always in the best operating state. This adaptive adjustment capability can intelligently optimize the energy storage strategy according to the actual performance of the energy storage equipment, the volatility of photovoltaic power generation, and changes in electricity demand, to avoid excessive or insufficient energy reserves. The adjustment of the adaptive energy storage strategy not only improves the operating efficiency of the energy storage equipment, but also ensures that the system can maintain a stable operating state when facing fluctuations in load demand or photovoltaic power generation. Whether it is a change in light intensity or a fluctuation in electricity load, the system can be intelligently dispatched to avoid unstable power supply due to an imbalance in the energy storage system.

[0023] In this specification, a photovoltaic power generation energy storage configuration system is provided, which is used to execute the above-mentioned photovoltaic power generation energy storage configuration method, and the photovoltaic power generation energy storage configuration system includes: The power generation prediction module is used to obtain photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network based on the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data from the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; The power consumption trend analysis module is used to collect historical power consumption load data of the distributed synchronous energy storage network to obtain historical power consumption load data; perform load curve curvature trend analysis on the historical power consumption load data to generate load curve curvature trend data; perform load demand classification on the historical power consumption load data based on the load curve curvature trend data to generate power consumption load demand pattern data; The energy storage configuration control module is used to calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain the energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data, thereby generating an energy storage configuration management scheduling strategy; The energy storage configuration optimization module is used to test the energy storage performance of the distributed synchronous energy storage network through the energy storage configuration management scheduling strategy and generate energy storage performance test data; and to adjust the energy storage strategy adaptively according to the energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

[0024] The beneficial effect of the present invention is that by acquiring the information data of photovoltaic power generation energy sites and constructing a distributed synchronous energy storage network, the power generation data of each site can be collected in real time. This provides high-precision basic data support for subsequent energy scheduling and energy storage management. By predicting real-time photovoltaic power generation data, power generation plans can be planned in advance under unstable weather and power load conditions, energy waste can be reduced, energy storage and power scheduling can be optimized, and stable power supply can be ensured. The spatial and temporal feature extraction and load curve curvature trend analysis of historical power load data can reveal the fluctuation pattern of power load and help predict future power demand, which is especially important in the case of large seasonal fluctuations. By classifying the load curve trend data, different power demand patterns can be identified, which helps optimize the allocation and scheduling of power resources, ensures uninterrupted energy supply under different demand patterns, and avoids energy waste. By combining real-time photovoltaic power generation prediction data with power load demand pattern data, the energy supply and demand ratio is calculated to help timely identify the imbalance between supply and demand in the power system. This can effectively predict the situation of excess or insufficient power and take measures in advance. According to the energy supply and demand ratio data, the energy storage system is optimized in real time, and the charging and discharging strategies of the energy storage equipment are adjusted to ensure the balance between photovoltaic power generation and load demand. This not only improves the efficiency of energy use, but also avoids excessive use or inefficient operation of energy storage equipment. Through energy storage performance testing, the system can monitor the efficiency and status of energy storage equipment in real time to ensure that the energy storage system is always in the best operating state. Through data feedback, potential problems of energy storage equipment can be quickly identified and the probability of failure can be reduced. According to the energy storage performance test data, the energy storage strategy is adaptively adjusted to ensure that the system can flexibly respond to different power loads and power generation conditions and optimize power storage and scheduling strategies. Therefore, the present invention improves the accuracy and flexibility of photovoltaic energy storage configuration through real-time photovoltaic power generation prediction, dynamic analysis of load demand and adaptive energy storage strategy optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of the steps of a method for configuring energy storage for photovoltaic power generation; Figure 2 for Figure 1 Detailed implementation steps of step S2 in the flowchart; Figure 3 for Figure 1 Detailed implementation steps of step S3 in FIG. The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0026] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0027] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0028] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0029] To achieve this, please refer to Figures 1 to 3 , a method for configuring energy storage for photovoltaic power generation, the method comprising the following steps: Step S1: Acquire photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network for the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data for the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; Step S2: Collect historical power load data of the distributed synchronous energy storage network to obtain historical power load data; perform load curve curvature trend analysis on the historical power load data to generate load curve curvature trend data; classify load demand of the historical power load data based on the load curve curvature trend data to generate power load demand pattern data; Step S3: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to generate an energy storage configuration management scheduling strategy; Step S4: Perform energy storage performance testing on the distributed synchronous energy storage network through energy storage configuration management scheduling strategy to generate energy storage performance test data; perform adaptive energy storage strategy adjustment according to the energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

[0030] The present invention collects real-time data of photovoltaic power generation sites and establishes a distributed synchronous energy storage network, which can ensure that the photovoltaic power generation system can monitor and coordinate each energy storage unit in real time, ensure the real-time and accuracy of the data, improve the response speed and coordination ability of the system, and provide basic data support for subsequent power generation prediction and energy storage optimization. By collecting and analyzing historical power load data, the changing trend of load demand can be accurately understood, and the fluctuation law of load demand can be identified through the curvature trend analysis of the load curve. This process helps to classify the power load more accurately, so as to better predict the future power demand pattern, ensure that the energy storage system can match the load demand, and achieve more efficient energy utilization. By combining real-time photovoltaic power generation prediction data with power load demand pattern data, the energy supply and demand ratio is calculated, thereby providing an optimization basis for the configuration of the energy storage system. According to the changes in the energy supply and demand ratio, the distributed synchronous energy storage network is dynamically configured and optimized, which can achieve a balance between energy supply and demand, improve energy storage efficiency, avoid excessive energy storage or energy waste, and ensure the efficient operation of the photovoltaic power generation system. By testing the performance of the energy storage system and making adaptive adjustments based on the test results, the energy storage strategy can be adjusted in real time to ensure that the energy storage system can operate efficiently under different load and power generation conditions. This step makes the energy storage system more flexible and intelligent, and can adjust the configuration and strategy according to the actual operation data, improve the reliability, stability and life of the system, and optimize the energy management of the photovoltaic power generation system. Therefore, the present invention improves the accuracy and flexibility of photovoltaic energy storage configuration through real-time photovoltaic power generation prediction, dynamic analysis of load demand and adaptive energy storage strategy optimization.

[0031] In the embodiment of the present invention, reference Figure 1 As shown, it is a schematic diagram of the steps of a photovoltaic power generation energy storage configuration method of the present invention. In this example, the photovoltaic power generation energy storage configuration method includes the following steps: Step S1: Acquire photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network for the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data for the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; In an embodiment of the present invention, basic information of each photovoltaic power generation site, such as geographical location, equipment type, capacity, historical power generation data, weather data, solar radiation, etc., is obtained from a photovoltaic power generation system management platform or a sensor network. According to different site types, a unified data structure is constructed, including site number, installation date, location coordinates, system type, historical power generation curve and other information. A distributed energy storage system is constructed between multiple photovoltaic power generation sites, and energy storage management is performed using real-time data synchronization. Distributed data synchronization protocols (such as NTP, P2P protocols, etc.) are used to ensure that data collected from different sites are synchronized in real time. This process can be completed by establishing a data exchange center or a central control system. The charging and discharging strategies of energy storage devices at each site are dynamically adjusted according to the power generation situation and energy storage demand of the site. The management of the energy storage network needs to consider the load situation of multiple sites, and edge computing or cloud computing platforms are used for data processing and decision-making. A variety of sensors (such as current, voltage, temperature, radiation sensors, etc.) are deployed at each photovoltaic power generation site, and data is transmitted to the central data processing platform via wireless transmission or wired network. The collected data should include power output, solar radiation intensity, temperature, etc., and should be stored in a standard data format (such as JSON, CSV or time series database format) to facilitate subsequent analysis and processing. The power generation of each site is monitored in real time through the central monitoring platform, and the collected data is converted into real-time power output data. The collected real-time photovoltaic power generation data is preprocessed by denoising, filling missing values, standardization and other preprocessing operations to ensure data quality. Machine learning models (such as LSTM, support vector machine, random forest, etc.) are used to predict standard real-time photovoltaic power generation data. The model is trained based on input variables such as historical power generation data, weather forecasts, and solar radiation intensity, and real-time predictions are performed. Since photovoltaic power generation is greatly affected by weather changes, LSTM (long short-term memory network) is an effective method for processing time series data and is suitable for capturing long-term dependencies. The prediction accuracy of the model is evaluated by cross-validation and other methods, and the model parameters are optimized. The trained prediction model generates photovoltaic power generation forecast data for a period of time in the future (such as 1 hour, 6 hours, 24 hours). These data can contain information such as the expected power generation of each photovoltaic power generation site and the impact of expected weather changes on power generation.

[0032] Step S2: Collect historical power load data of the distributed synchronous energy storage network to obtain historical power load data; perform load curve curvature trend analysis on the historical power load data to generate load curve curvature trend data; classify load demand of the historical power load data based on the load curve curvature trend data to generate power load demand pattern data; In an embodiment of the present invention, the historical power load data of each photovoltaic power generation site and its supporting energy storage equipment are collected through the monitoring platform or smart meter system of the distributed synchronous energy storage network. These data include daily, hourly or minute power demand, as well as the charging and discharging status of the energy storage system. The historical power load data should cover a certain time span (such as 1 month, 1 year, etc.) to ensure that the periodic changes and load fluctuations of the power load can be reflected. The collected load data should include information such as current, voltage, power, and energy storage device status at each time point, usually stored in the form of time series data, and recorded in a standard format (such as CSV, JSON, etc.). Noise or outliers in the data are removed by filtering methods (such as mean filtering, Kalman filtering, etc.) to ensure the smoothness and accuracy of the data. Interpolation processing is performed on the missing historical power load data, such as linear interpolation or model-based prediction to fill in the missing values. The historical load data is normalized or standardized so that the data of different dimensions in subsequent analysis are comparable. Based on the historical power load data, a load curve chart is generated to describe the load changes at different time scales (such as day, week, month). Use time series analysis methods (such as STL decomposition method, Holt-Winters seasonal smoothing method, etc.) to decompose the load curve into long-term trends, seasonal changes, periodic fluctuations, and random errors. Use polynomial fitting, B-spline fitting or other smooth curve fitting techniques to fit the historical load curve to obtain a smooth load curve. Calculate the curvature of the load curve, that is, the rate of change of load change. Curvature is an important feature that reflects the trend of curve changes and can help identify the severity of load fluctuations. By calculating the curvature trend of the load curve in different time periods, the volatility pattern of load changes can be identified. For example, the curvature value of some peak load periods is large, indicating that the demand fluctuates violently; while during low load periods, the curvature is small, indicating that the demand is relatively stable. Through curvature analysis, the curvature trend data of the load curve is generated. The data set contains the curvature values ​​of load changes at different time points and the time evolution trend of these curvature values. According to the curvature trend data of the load curve, the historical power load data is divided into different demand modes. For example, it can be divided into peak load mode, stable load mode, and valley load mode according to the volatility of curvature. Use cluster analysis (such as K-means, DBSCAN and other algorithms) to classify the curvature trend data of the load curve and classify different load demand patterns into different categories. Input the curvature trend as a feature into the machine learning classification model (such as decision tree, support vector machine, etc.) to automatically classify the load demand pattern. Evaluate the accuracy of the classification results through methods such as cross-validation, and adjust the model parameters to optimize the classification effect. Generate electricity load demand pattern data based on the classification results. Each historical time period will be marked as a specific demand pattern type (such as peak, stable, trough, etc.).

[0033] Step S3: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to generate an energy storage configuration management scheduling strategy; In an embodiment of the present invention, by setting the ratio of photovoltaic power generation to power demand, energy supply and demand ratio data at each time point is generated. With the goal of maximizing the overall energy efficiency of the system, the purpose of optimizing the configuration of the energy storage system is to reasonably configure the charging and discharging strategy of the energy storage device according to the supply and demand ratio. The goal is to minimize energy loss while ensuring that the energy storage system can meet the load demand when photovoltaic power generation is insufficient; and that the energy storage system can be effectively charged when power generation is in excess. Use linear programming, dynamic programming or other optimization algorithms to optimize the configuration of the energy storage device according to the energy supply and demand ratio data. Based on the photovoltaic power generation energy storage configuration optimization data, the system needs to adjust the charging and discharging strategy of the energy storage device in real time to achieve the optimal energy supply and load demand matching. The goal of dynamic adjustment is to dynamically adjust the operating state of the energy storage device according to the real-time supply and demand ratio and the energy storage optimization configuration data. In the power supply gap mode, the energy storage device charging operation is performed first; in the power supply surplus mode, the energy storage device discharging operation is performed first until the load demand is met. According to the real-time supply and demand ratio, the charging and discharging power of the energy storage device is dynamically adjusted. For example: when the photovoltaic power generation exceeds the load demand, the energy storage device starts to charge. When the photovoltaic power generation is not enough to meet the load demand, the energy storage device discharges. When the photovoltaic power generation and load demand are roughly balanced, the energy storage device is in standby or charge / discharge balance state. According to the above dynamic adjustment model, a detailed energy storage configuration management scheduling strategy is formulated. The scheduling strategy should include the energy storage charging and discharging plan for each time period to ensure the maximum efficiency of the energy storage system. Deploy the scheduling strategy in the real-time system and perform energy storage management operations based on the load demand and power generation forecast data at each time point. Use scheduling control algorithms (such as priority scheduling, time-sharing scheduling, etc.) to achieve automatic adjustment of the energy storage system to ensure supply and demand balance.

[0034] Step S4: Perform energy storage performance testing on the distributed synchronous energy storage network through energy storage configuration management scheduling strategy to generate energy storage performance test data; perform adaptive energy storage strategy adjustment according to the energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

[0035] In the embodiment of the present invention, by evaluating the charging and discharging efficiency of the distributed synchronous energy storage network under the execution of the energy storage management scheduling strategy, it is ensured that the energy storage device can effectively cope with the fluctuation of photovoltaic power generation and meet the load demand in actual operation. Test the response time, charging and discharging rate, charging / discharging depth, loss rate, system stability and sustainable operation capability of the energy storage system. Based on the energy storage configuration management scheduling strategy, a series of simulation tests are carried out to simulate the behavior of energy storage devices in different time periods and different photovoltaic power generation conditions. For the charging stage, the charging speed and charging completion of the energy storage system are tested. For the discharge stage, the responsiveness of the energy storage system at peak load demand is tested. In actual operation, the real-time charging and discharging data of the energy storage device are collected through the installed smart meter and monitoring system. When performing performance testing, the system will continuously record the key parameters such as the charging and discharging power of the energy storage device, the state of the energy storage battery (SOC, DoD) and the efficiency of the system. The energy storage performance test data includes but is not limited to: the charging and discharging power data of the energy storage device at each time point during the test, the charging and discharging efficiency during the energy storage process, the discharge depth of the energy storage battery, the working state of the energy storage system, and the time delay of the energy storage system to start charging and discharging. The data obtained during the test is visualized to generate various charts, such as real-time comparison charts of load demand and photovoltaic power generation, charging and discharging power change curves, efficiency and depth change charts, etc. A variety of chart methods such as line charts, bar charts, and heat maps can be used to analyze the performance of the energy storage system. According to the results of the energy storage performance test, the energy storage configuration management scheduling strategy is optimized and adjusted in real time to improve the responsiveness, charging and discharging efficiency, and load matching effect of the energy storage system. When the charging and discharging efficiency of the energy storage device is low, the charging and discharging rates are optimized to reduce losses. When the load demand fluctuates greatly, the discharge depth of the energy storage device is adjusted to ensure stable supply. When the performance of the energy storage system degrades (such as efficiency below the threshold, reaction time is too long, etc.), the energy storage configuration parameters are adjusted to restore the system efficiency. Machine learning (such as reinforcement learning, genetic algorithms, etc.) is used to achieve adaptive adjustment of strategies. These algorithms can continuously adjust the operating mode of the energy storage device based on real-time energy storage performance data. By training the historical behavior of the energy storage system, the reinforcement learning algorithm can automatically select the optimal charging and discharging strategy. By simulating the process of genetic selection, the energy storage system configuration is optimized and a more efficient charging and discharging scheduling scheme is generated. During the operation of the energy storage system, the real-time operating status of each energy storage device is monitored, including charging and discharging efficiency, energy storage depth, charging time, etc. According to the preset adjustment rules, if the efficiency of the energy storage device is lower than the set threshold or the reaction time exceeds the standard, the adaptive algorithm is automatically started to adjust the energy storage strategy. If the discharge efficiency of the energy storage device is lower than 80%, the duration of the charging phase is increased or the discharge rate is reduced to reduce losses. After the adaptive energy storage strategy is adjusted, the system re-executes the photovoltaic power generation energy storage configuration optimization task. The specific execution process includes: recalculating the charging and discharging strategies of the energy storage system according to the new energy storage configuration adjustment.Execute the optimized strategy to ensure that the energy storage system can be effectively adjusted to maintain stable operation of the system under different photovoltaic power generation levels and power load demands.

[0036] Preferably, step S1 comprises the following steps: Step S11: Acquire photovoltaic power generation energy site information data; Step S12: performing site distribution topology analysis on the photovoltaic power generation energy site information data to generate energy site distribution topology data; constructing a distributed synchronous energy storage network for the photovoltaic power generation energy site information data based on the energy site distribution topology data to obtain a distributed synchronous energy storage network; Step S13: collecting real-time power generation data of the distributed synchronous energy storage network to obtain real-time photovoltaic power generation data; performing data preprocessing on the real-time photovoltaic power generation data to generate standard real-time photovoltaic power generation data, wherein the data preprocessing includes data cleaning, data denoising, missing value filling and data standardization; Step S14: Perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data.

[0037] In an embodiment of the present invention, relevant data are collected from photovoltaic power generation sites through sensor networks, satellite remote sensing technology or ground collection equipment, including the geographical location of the site, battery panel layout, power generation capacity, historical power generation data, meteorological data, etc. The API interface can be connected to the intelligent control system of the photovoltaic equipment to obtain the operating status data of the equipment in real time. The geographical location of the photovoltaic site is topologically analyzed by graph theory or spatial analysis methods to build the grid connection relationship between the sites. This can be visualized by a geographic information system (GIS) to understand the distance, power generation potential and power transmission path between each site. Based on the results of the topological analysis, a distributed synchronous energy storage network is designed. The network works synchronously with photovoltaic power generation equipment through energy storage devices (such as battery packs) to cope with the instability and volatility of photovoltaic power generation. Intelligent scheduling algorithms (such as load forecasting, energy storage capacity optimization, etc.) can be used to ensure the efficient use of energy storage equipment. The output data of photovoltaic power generation is collected in real time through sensors (such as power sensors, light sensors, temperature sensors, etc.) installed at each photovoltaic power generation site. Remote transmission and monitoring of data can be achieved through the Internet of Things (IoT) technology. Duplicate data, outliers and noise are removed. Use filters (such as Kalman filters) to denoise the collected real-time data to ensure data quality. Use interpolation methods (such as linear interpolation or machine learning-based interpolation methods) to fill missing values ​​in the data to ensure data integrity. Standardize data from different sources (such as minimum-maximum standardization or Z-Score standardization) so that the data is on the same scale for subsequent analysis and modeling. Use machine learning-based prediction algorithms (such as LSTM networks, SVR regression, XGBoost, etc.) to model the pre-processed standard real-time photovoltaic power generation data. These models predict future photovoltaic power generation output based on historical data and real-time collected data. Train the prediction model through training data sets, and use cross-validation and other methods to optimize the model to improve the accuracy and generalization of the prediction. Under real-time data input, use the trained prediction model to make real-time predictions of photovoltaic power generation. The prediction results can be used as input for energy storage management, scheduling and load optimization, providing decision support for subsequent grid scheduling and power management.

[0038] Preferably, constructing a distributed synchronous energy storage network for photovoltaic power generation energy site information data based on energy site distribution topology data includes: Based on the graph theory method, the energy site distribution topology data is topologically modeled to generate energy site topology model data; the energy site topology model data is node-importantly evaluated to generate topology node importance data; the energy site topology model data is node-labeled according to the topology node importance data to generate key nodes and secondary nodes; The energy site topology model data is edge-connected according to the joint points and secondary nodes to generate an initial energy storage network; the initial energy storage network is deployed with distributed nodes to generate a distributed energy storage network; the distributed energy storage network is synchronized with three layers of data using the photovoltaic power generation energy site information data to generate a distributed synchronous energy storage network, where the three layers of data synchronization include time synchronization, power synchronization, and state synchronization.

[0039] In an embodiment of the present invention, the geographical distribution of energy sites is constructed by collecting information such as geographical location, power output, grid access point, historical operation data, etc. from each photovoltaic power generation site. Based on a graph theory model (such as an undirected graph or a weighted graph), each energy site is regarded as a node, and the power transmission line (or grid connection line) between the sites is regarded as an edge. By analyzing the geographical location and power transmission capacity of these sites, an energy site topology model is constructed. Each node of the graph represents a photovoltaic power generation site, and the weight of the edge can be set according to information such as the bandwidth and transmission capacity of the grid connection. The importance of each node is evaluated using an importance measurement method in graph theory (such as the degree of the node, betweenness centrality, proximity centrality, etc.). Specific methods may include: degree centrality: the larger the degree of a node, the more it is connected to other nodes, and the node is generally considered to be more important. Betweenness centrality: calculate the number of times a node acts as an intermediary in the network. Nodes with high betweenness centrality play an important bridging role in the network. Proximity centrality: the average distance between a node and other nodes. Nodes with proximity centrality usually have a greater impact on the entire network. According to the above evaluation method, the importance of each node is calculated and data is generated, marking high-importance nodes (key nodes) and low-importance nodes (secondary nodes). According to the importance data of the node, each node is marked as a key node or a secondary node. Key nodes are those sites with large power output, connecting multiple sites or located in the core of the power grid, and secondary nodes are edge nodes or sites with small power demand. A set of topological node data is generated to clarify which nodes are key nodes and which are secondary nodes for subsequent processing. Based on the connection relationship between key nodes and secondary nodes, the initial energy storage network is constructed using methods such as the shortest path algorithm and the minimum spanning tree. More edges need to be connected between key nodes to ensure efficient transmission of power loads and full utilization of energy storage. Each node is paired with its neighboring nodes through a grid connection, and corresponding energy storage devices (such as battery energy storage systems) are set up to ensure that each node in the network has sufficient energy storage resources and can perform charging and discharging operations when needed. According to the needs of the energy storage network, energy storage devices are distributed and deployed between the key nodes and secondary nodes of the network. The capacity of the energy storage equipment should be related to the power generation potential, power demand and network stability of the node to ensure that the energy site can maintain a stable power supply under different loads. In the distributed energy storage network, the energy storage nodes are dynamically loaded according to the real-time power demand and power generation data. Optimization algorithms (such as particle swarm optimization, genetic algorithm, etc.) can be used to determine the optimal deployment location and capacity of the energy storage equipment. In order to ensure the coordinated operation between all energy sites and energy storage equipment, the time of all nodes must be synchronized first. NTP (Network Time Protocol) or GPS synchronization technology can be used to ensure the consistency of time accuracy of all devices. Based on the real-time collected photovoltaic power generation data, the power generation power of each node in the energy storage network is synchronized.Distributed control algorithms (such as consistency algorithms) can be used to ensure that the power generation and storage power of all energy storage nodes match each other to avoid over-storage or under-storage. The status of each node is synchronized through real-time monitoring and feedback mechanisms, including the battery state of charge (SOC), remaining energy storage capacity, grid load, etc. Distributed synchronization algorithms (such as network-based synchronization control or consistency protocols) are used to ensure the coordination and consistency of the state of the energy storage system, reducing scheduling conflicts and efficiency reduction caused by asynchronous states. After three layers of data synchronization, a highly coordinated distributed synchronous energy storage network is formed. The network can dynamically adjust the energy storage capacity according to photovoltaic power generation and load demand, balance the supply and demand of the power grid, and ensure that the energy storage system can respond efficiently when photovoltaic power generation fluctuates or grid load changes.

[0040] Preferably, step S14 includes the following steps: Step S141: extracting basic photovoltaic power generation characteristics from standard real-time photovoltaic power generation data to obtain basic photovoltaic power generation characteristic data; performing time series characteristic analysis on the basic photovoltaic power generation characteristic data to generate photovoltaic power generation time series characteristic data; Step S142: dividing the photovoltaic power generation time series characteristic data into data sets to generate a model training set and a model test set; training the model training set using a long short-term memory neural network algorithm to generate a photovoltaic power generation prediction pre-model; optimizing and iterating the photovoltaic power generation prediction pre-model using the model test set to generate a photovoltaic power generation prediction model; Step S143: Importing the standard real-time photovoltaic power generation data into the photovoltaic power generation prediction model to perform photovoltaic power generation prediction and generate real-time photovoltaic power generation prediction data.

[0041] In an embodiment of the present invention, basic features are extracted from standard real-time photovoltaic power generation data, generally including but not limited to: actual power generation at each time step, environmental factors such as light intensity, temperature, humidity, etc. that affect photovoltaic power generation efficiency, for example, the working state of photovoltaic modules, the charging state of batteries, etc. Data preprocessing methods (such as moving average method, standard deviation, maximum value, minimum value, etc.) are used to extract key statistical features of photovoltaic power generation. The features of seasonal variation and periodic fluctuation are extracted to identify the periodic patterns that affect power generation. Time series analysis is performed on the extracted basic photovoltaic power generation feature data to analyze its time series characteristics, such as seasonal variation, trend change, periodic fluctuation, etc. Time series features in the data are identified by statistical methods such as autocorrelation function (ACF) and partial autocorrelation function (PACF). Methods such as Fourier Transform or Wavelet Transform can be used to further analyze the frequency domain characteristics and change trends of photovoltaic power generation data. Based on the time series characteristic analysis, photovoltaic power generation time series characteristic data is generated. The photovoltaic power generation time series characteristic data is divided into a training set and a test set in chronological order. The partitioning method can be based on a time window (for example, the first 70% of the data is used as a training set and the last 30% of the data is used as a test set), or a cross validation method is used for partitioning. In particular, when partitioning, it should be ensured that the time ranges of the training set and the test set do not overlap to avoid data leakage. The sliding window method can be used to dynamically generate training sets and test sets for different time periods, thereby improving the generalization ability of the model. A long short-term memory neural network (LSTM) is used to perform time series prediction on photovoltaic power generation data. LSTM is a neural network architecture that is particularly suitable for processing sequence data (especially time series data) and can effectively capture long-term dependencies in the data. Input layer: Receives feature data extracted from standard real-time photovoltaic power generation data (such as historical power generation, meteorological data, etc.). Hidden layer: Uses LSTM units to capture the time series characteristics of photovoltaic power generation. Multi-layer LSTM can be set to enhance the expressiveness of the model. Output layer: Outputs the predicted photovoltaic power generation, usually as a continuous value (for example, predicted power generation at the hour or minute level). Using the training set data, the LSTM model is trained by the back propagation algorithm to optimize the weight parameters in the network. Use loss functions (such as mean square error MSE) to measure the prediction error of the model, and iterate and update it through optimization algorithms (such as Adam optimizer). Test the LSTM model through the model test set to evaluate its prediction performance (such as through indicators such as root mean square error (RMSE) and mean absolute error (MAE)). After the model is trained, hyperparameter tuning (such as adjusting the number of LSTM units, learning rate, time step, etc.) can be performed to optimize the model. According to the prediction results of the test set, iteratively optimize the model, continuously adjust the model parameters, and improve the prediction accuracy.Import standard real-time photovoltaic power generation data into the trained photovoltaic power generation prediction model (LSTM model) to predict photovoltaic power generation in the future time period. Input the real-time data collected from photovoltaic power generation sites (such as real-time light intensity, meteorological data, historical power generation data, etc.) into the LSTM model. The LSTM model generates real-time photovoltaic power generation prediction data based on the input data, including photovoltaic power generation predictions for the next few hours or minutes.

[0042] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes: Step S21: collecting historical power load data of the distributed synchronous energy storage network to obtain historical power load data; Step S22: extracting spatiotemporal load characteristics from historical power load data to obtain spatiotemporal power load characteristic data; analyzing seasonal variation patterns of the spatiotemporal power load characteristic data to generate a load characteristic data set; converting historical power load data into a historical power load curve according to the load characteristic data set to generate a historical power load curve; Step S23: performing load curve curvature trend analysis on the historical power load curve to generate load curve curvature trend data, wherein the load curve curvature trend data includes positive trend curvature, gentle trend curvature and negative trend curvature; Step S24: Segment the historical power load curve based on the positive trend curvature, the gentle trend curvature and the negative trend curvature to generate a high load trend curve, a gentle load trend curve and a low load trend curve; classify the load demand of the historical power load data according to the high load trend curve, the gentle load trend curve and the low load trend curve to generate power load demand pattern data.

[0043] In an embodiment of the present invention, historical power load data is collected from each power user in a distributed synchronous energy storage network. These data are usually power consumption data recorded based on time (e.g., hour, day, or month). Based on the historical power load data, the following spatiotemporal characteristics are extracted: the change trend of the load within a day, usually day and night changes, manifested as the load difference between day and night. Weekly power consumption fluctuations, such as the difference in power consumption between weekdays and weekends. Seasonal fluctuations, especially the impact of temperature changes on power load, such as increased air conditioning load in summer. Analyze the seasonal load changes in historical data to generate power consumption patterns for summer and winter, such as increased heating demand in winter and increased air conditioning load in summer. Use statistical methods (such as periodic analysis, Fourier transform) to analyze historical load data to reveal the change pattern of power demand with seasons, climate, time and other factors. Seasonal model: Establish a seasonal fluctuation model, such as using seasonal difference method to eliminate the impact of seasonal fluctuations, and generate a load characteristic data set to describe the load change characteristics in different seasons, weather and time periods. Based on the load characteristic data set, the original historical power load data is smoothed by interpolation method or curve fitting to generate a more regular historical power load curve. The generated historical power load curve is subjected to curvature trend analysis, that is, the change trend is identified by calculating the derivative or second-order derivative of the load curve: Positive trend curvature: The load curve gradually rises, indicating that the load demand is increasing. Gentle trend curvature: The change amplitude of the load curve is small, indicating that the load demand changes steadily. Negative trend curvature: The load curve gradually decreases, indicating that the load demand is decreasing. The trend of load change is evaluated by calculating the first-order derivative (i.e., slope) and second-order derivative (i.e., curvature) of the load curve. The curvature is calculated by numerical difference method or spline interpolation, and the load change trend of each period is marked. The obtained load curve curvature trend data includes: Positive trend curvature: If the load rises sharply, it is related to high temperature weather or special events (such as holidays). Gentle trend curvature: If the load changes steadily, it indicates a regular power consumption mode. Negative trend curvature: If the load decreases, it is related to factors such as equipment failure. The load curve segmentation of the historical electricity load curve is performed according to the positive trend curvature, gentle trend curvature and negative trend curvature: When the load curve rises faster, it belongs to the high load demand period, which is related to load peaks such as high temperature in summer or heating in winter. The load change is relatively smooth, indicating the normal working day electricity load, which usually occurs in the alternating period between day and night. The load curve drops, indicating a decrease in electricity demand, such as at night, holidays or seasonal air conditioning load reduction periods. According to the load curve segmentation results, the load demand is divided into three main modes: High load demand mode: The period represented by the high load trend curve is used for load scheduling optimization and energy storage scheduling. Gentle load demand mode: Suitable for conventional load management and demand response management. Low load demand mode: Suitable for load peak shaving and valley filling to reduce peak load.Integrate high load, flat load and low load trend data to generate electricity load demand pattern data, providing a basis for future load forecasting, energy scheduling and energy storage management.

[0044] As an example of the present invention, refer to Figure 3 As shown, in this example, step S3 includes: Step S31: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; wherein the formula for calculating the energy supply and demand ratio is as follows:

[0045] In the formula, Expressed as the energy supply-demand ratio, Expressed as time The photovoltaic power generation power, Expressed as time The power output of the energy storage device, Expressed as time Electricity obtained from the grid, Expressed as time The power demand of the electrical load; Step S32: comparing the energy supply-demand ratio data with a preset standard supply-demand ratio threshold, when the energy supply-demand ratio data is greater than the preset standard supply-demand ratio threshold, a power supply surplus mode is generated; when the energy supply-demand ratio data is less than the preset standard supply-demand ratio threshold, a power supply gap mode is generated; Step S33: Optimizing the energy storage system configuration of the distributed synchronous energy storage network based on the power supply surplus mode and the power supply gap mode, and generating photovoltaic power generation energy storage configuration optimization data; Step S34: Dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data, thereby generating an energy storage configuration management scheduling strategy.

[0046] In the embodiment of the present invention, the energy supply and demand ratio is calculated based on the real-time photovoltaic power generation forecast data, the output power of the energy storage device, the power obtained from the power grid, and the power load demand to evaluate the energy supply and demand balance state. The formula for calculating the energy supply and demand ratio is as follows: In the formula, Expressed as the energy supply-demand ratio, Expressed as time The photovoltaic power generation power, Expressed as time The power output of the energy storage device, Expressed as time Electricity obtained from the grid, Expressed as time The power demand of the power load. The current power supply mode (surplus mode or gap mode) is identified by comparing the calculated energy supply and demand ratio R(t) with the preset standard supply and demand ratio threshold. Through historical data analysis and experience, a reasonable supply and demand ratio threshold is set, which can be adjusted according to different power consumption scenarios and seasons. For example, during the peak power consumption in summer, the threshold is set to 1.2, while during the low peak in winter, the threshold is set to 1.5. When the energy supply and demand ratio data is greater than the preset standard supply and demand ratio threshold, a power supply surplus mode is generated; when the energy supply and demand ratio data is less than the preset standard supply and demand ratio threshold, a power supply gap mode is generated; in the power supply surplus mode, the goal is to store excess energy for future use in the power supply gap mode. The system should store excess power from photovoltaic power generation, energy storage equipment and power grid in the energy storage equipment. According to the capacity and charging and discharging efficiency of the existing energy storage equipment, the configuration of the energy storage equipment is optimized to ensure that the excess power can be absorbed and stored efficiently. In the power supply gap mode, the task of the energy storage system is to provide the necessary power to the power grid to ensure that the load demand is met. At this time, the energy storage device should have a high discharge capacity and provide supplementary power when the power grid is insufficient. According to the discharge capacity and storage capacity of the energy storage device, adjust the configuration of the energy storage system so that it can provide the required power when needed. Use heuristic methods such as genetic algorithms and particle swarm optimization algorithms to globally optimize the configuration of the energy storage system to ensure that the best energy allocation effect can be achieved under different power supply modes. In the optimization process, it is necessary to consider the capacity of the energy storage device, the charging and discharging efficiency, the system safety and other constraints to ensure that the optimization scheme meets the actual application requirements. According to the changes in real-time photovoltaic power generation, energy storage device status, power supply of the power grid and load demand, dynamically adjust the charging and discharging strategy of the energy storage device. For example, if the system enters the power supply gap mode, it will give priority to discharge from the energy storage device; if it enters the power supply surplus mode, it will give priority to storing the electric energy. Use optimization algorithms (such as linear programming, dynamic programming, etc.) to optimize the scheduling of the energy storage system to ensure that under different supply and demand conditions, the energy storage device can maximize the efficiency of power use and reduce energy waste. Use a real-time monitoring system to monitor the status of the energy storage device. The dispatching system can adjust the energy allocation strategy in real time to avoid overcharging or discharging of the energy storage device and ensure the stability of the power supply. Through intelligent scheduling algorithms, energy storage resources are rationally allocated during system load peaks to ensure uninterrupted power supply.

[0047] Preferably, step S33 includes the following steps: Step S331: confirming the mode time period of the distributed synchronous energy storage network based on the power supply surplus mode and the power supply gap mode to obtain the power supply surplus time period and the power supply gap time period; performing a buffer time period analysis on the power supply surplus time period and the power supply gap time period to generate a power supply change buffer time period; Step S332: setting energy discharge for the power surplus time period to generate an energy discharge curve; setting energy charging for the power shortage time period to generate an energy charging curve; fitting the energy discharge curve and the energy charging curve according to the power supply change buffer period to generate energy loss data for the buffer period; Step S333: Optimize the configuration of energy storage batteries in the buffer period of the distributed synchronous energy storage network based on the energy loss data in the buffer period, thereby generating photovoltaic power generation energy storage configuration optimization data.

[0048] In the embodiment of the present invention, by comparing the energy supply and demand ratio data calculated previously with the set standard supply and demand ratio threshold, a power supply surplus time period and a power supply gap time period are generated, and these time periods will be used for further energy storage configuration and scheduling. In order to avoid unnecessary losses caused by frequent charging and discharging of the energy storage system, it is necessary to analyze the buffer period between the power supply surplus and the power supply gap. The buffer period refers to the transition stage between the power supply surplus and the power supply gap, and there is a lag in energy conversion (for example, photovoltaic power generation gradually weakens after being blocked by clouds, or grid scheduling lags, etc.). Through time series analysis, the transition period between the power supply surplus and power supply gap modes is determined, and a reasonable buffer period is set. The buffer period will be used as a key time window for energy storage system scheduling to ensure that appropriate energy charging and discharging adjustments are made during this period. The length of the buffer period should be dynamically adjusted according to factors such as the volatility of historical data and the load change rate. In the power supply surplus period, the energy storage system should discharge the excess electric energy to avoid energy waste. In the power supply surplus period, the energy storage system starts the discharge process based on the excess photovoltaic power generation and the excess power of the grid. The energy discharge curve reflects the discharge process of the energy storage system during the power surplus period. Linear, nonlinear or piecewise linear models can be used to fit the change of discharge over time. If the energy storage device fails to fully release its stored energy during the discharge process, the discharge curve shows a gentle decline. The energy storage discharge curve should be dynamically adjusted according to different energy storage technologies (such as batteries, supercapacitors, etc.) to avoid excessive discharge of the equipment. During the power shortage period, the energy storage system should be charged to replenish the energy in the battery to ensure that future electricity demand is met. During the power shortage period, the energy storage system should give priority to using grid power for charging based on the grid power and the current remaining power of the energy storage device. The charging curve reflects the charging process of the energy storage system during the power shortage period. The charging curve usually shows an upward trend and is adjusted according to the battery charging efficiency and the power supply capacity of the grid. During the charging process, the charging curve should be adjusted according to factors such as battery charging capacity limit, charging efficiency and battery power limit. Based on the buffer period between the power surplus and power shortage modes, the energy discharge curve and charging curve are curve fitted, considering factors such as energy loss, equipment response time and efficiency loss. The energy loss in the buffer period is calculated by fitting the curve. These losses are usually related to factors such as efficiency loss and response delay of the equipment during the charging and discharging process. Input the energy loss data of the buffer period, combined with the technical characteristics of the energy storage system (such as battery capacity, charging and discharging efficiency, response time, etc.). Use optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms or linear programming, etc.) to globally optimize the configuration of the energy storage system, with the goal of minimizing energy losses during the buffer period. Based on the results of the optimization algorithm, generate photovoltaic power generation energy storage configuration optimization data to guide the energy storage system's storage and discharge strategies in different time periods.

[0049] Preferably, step S34 includes the following steps: Step S341: performing node energy redundancy calculation on the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to obtain node energy redundancy data; performing adjacent node status analysis on the distributed synchronous energy storage network according to the node energy redundancy data to generate adjacent node status data; Step S342: using the adjacent node status data to perform energy scheduling weight allocation on the node energy redundancy data to generate energy scheduling weight allocation data; Step S343: perform emergency energy node screening on adjacent node status data through energy scheduling weight distribution data to obtain emergency energy node data; perform emergency scheduling energy storage path analysis on the emergency energy node data according to node energy redundancy data to generate emergency node scheduling energy storage path data; Step S344: Dispatching and uniformly controlling the energy storage of the distributed synchronous energy storage network based on the emergency node scheduling energy storage path data, thereby generating an energy storage configuration management scheduling strategy.

[0050] In an embodiment of the present invention, the energy redundancy level of each node is calculated by optimizing the data according to the photovoltaic power generation energy storage configuration. The energy redundancy of a node refers to the additional energy available for scheduling in the current energy storage device of the node, which is usually determined by the following factors: the current energy storage capacity of the node, the energy input and output status of the node (such as photovoltaic power generation, energy storage discharge, charging, etc.), the load demand of the node and other network parameters. Through redundancy calculation, the node energy redundancy data of each node is obtained, which represents the additional power that the node can provide or consume in the future. Through the node energy redundancy data, the state of the adjacent nodes in the distributed synchronous energy storage network is analyzed. Specifically, the energy redundancy level of the adjacent nodes is analyzed to identify potential load overload or energy shortage areas. The adjacent node status data includes the energy redundancy status, load demand, energy storage status, etc. of each node. These data help to determine which nodes need more energy support and which nodes have excess energy to provide, and generate adjacent node status data for subsequent energy scheduling and emergency handling. Energy scheduling weight allocation: The node energy redundancy data of each node is weighted and allocated according to the adjacent node status data to generate energy scheduling weight allocation data. The specific method is as follows: For each node, the weight of energy dispatch is assigned according to the energy redundancy status and importance of its adjacent nodes. For example, if a node has high energy redundancy and is close to a node with large load demand, the dispatch weight of the node is high and is dispatched first. Considering the charging and discharging capacity, geographical location, and power flow of the node, the dispatch weight of each node is adjusted to make the energy dispatch decision more intelligent and flexible, and generate energy dispatch weight allocation data, which will be used for subsequent emergency energy node screening and dispatch decisions. Based on the energy dispatch weight allocation data, the adjacent node status data is screened to identify emergency energy nodes. Emergency energy nodes are those nodes where: load demand increases sharply and the currently dispatchable energy is insufficient to meet the demand. Nodes where energy storage equipment is about to run out, or nodes where power shortages are expected in a short period of time. The screened nodes will be marked as emergency nodes and need to be dispatched first. Based on the emergency energy node data and node energy redundancy data, emergency dispatch energy storage path analysis is performed on the emergency energy nodes. The goal of this analysis is to determine the best path to dispatch power from other nodes with higher energy redundancy. Path analysis considers factors including: charging and discharging efficiency of energy storage equipment, physical path of power flow, scheduling priority and constraints, and generates emergency node scheduling energy storage path data, which describes the power scheduling path from nodes with higher energy redundancy to emergency energy nodes. Based on the emergency node scheduling energy storage path data, the distributed synchronous energy storage network is uniformly controlled. Uniform control aims to ensure the balance and efficiency of energy scheduling and avoid excessive discharge or charging of a single node, which leads to system imbalance. Energy storage uniform control strategies include: dynamically adjusting the charging and discharging behavior of energy storage equipment according to the node energy redundancy level and load demand.Prioritize the dispatch of emergency energy nodes, and consider the power demand of other nodes at the same time to avoid wasting resources by over-dispatching a certain node. Based on energy storage uniform control and path analysis, generate energy storage configuration management scheduling strategy, which includes: charging and discharging timing and power of each node; selection and scheduling priority of energy storage path; priority dispatching plan for emergency energy nodes; energy balance management of the overall energy storage system.

[0051] Preferably, step S4 comprises the following steps: Step S41: executing the distributed synchronous energy storage network strategy through the energy storage configuration management scheduling strategy to generate photovoltaic power generation energy storage strategy execution data; performing energy storage performance test on the photovoltaic power generation energy storage strategy execution data to generate energy storage performance test data; Step S42: Adaptively adjust the photovoltaic power generation energy storage strategy execution data according to the energy storage performance test data to perform the photovoltaic power generation energy storage configuration optimization operation.

[0052] In the embodiment of the present invention, the distributed synchronous energy storage network is executed according to the energy storage configuration management scheduling strategy generated in the previous step. During execution, the charging and discharging behavior of each node is scheduled according to the energy storage strategy. Specifically, it includes: adjusting the discharge and charging behavior of the energy storage device according to the energy redundancy, load demand and scheduling weight of each node. The charging and discharging time of the energy storage device is accurately controlled to ensure that the energy storage resources are optimally allocated. By executing the energy storage strategy, the charging and discharging status of each node, the remaining battery power, the energy flow and other data are recorded in real time to form the photovoltaic power generation energy storage strategy execution data. The energy storage performance test is performed on the data after the energy storage strategy is executed. The purpose of the performance test is to evaluate the actual operation performance of the energy storage system and ensure the effectiveness and efficiency of the strategy execution. The test content includes: the energy utilization efficiency of the energy storage system, the charging and discharging efficiency of the energy storage device, the response speed and accuracy of the energy storage strategy, the system stability and load adaptability. Through the performance test, the test results of each node, each energy storage device and the entire system are obtained, and the energy storage performance test data is generated. According to the energy storage performance test data, the photovoltaic power generation energy storage strategy execution data is adaptively adjusted. The core goal of adaptive adjustment is to optimize the energy storage strategy based on real-time test results to better meet future energy needs and system changes. If the system experiences large fluctuations in load demand during certain periods, the charging and discharging rate or time of the energy storage device can be adjusted to ensure the rapid response capability of the energy storage device. If the test finds that the efficiency of the energy storage device is insufficient or the charging and discharging process is unbalanced, the use priority of the energy storage device can be adjusted to optimize the charging and discharging behavior of the battery and improve energy efficiency. According to the test results, the scheduling weights of each node are dynamically adjusted, and nodes with high energy redundancy or emergency nodes are scheduled first to reduce the response delay of the energy storage system when sudden load changes occur. Based on the above adjustments, the energy storage configuration optimization operation is performed. This includes reallocating energy storage, discharging and charging tasks, optimizing power flow, adjusting the working status of energy storage devices, and generating adaptive energy storage strategy adjustment data as the basis for the next scheduling cycle to ensure that the energy storage system can continue to optimize and adjust under changing conditions.

[0053] In this specification, a photovoltaic power generation energy storage configuration system is provided, which is used to execute the above-mentioned photovoltaic power generation energy storage configuration method, and the photovoltaic power generation energy storage configuration system includes: The power generation prediction module is used to obtain photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network based on the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data from the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; The power consumption trend analysis module is used to collect historical power consumption load data of the distributed synchronous energy storage network to obtain historical power consumption load data; perform load curve curvature trend analysis on the historical power consumption load data to generate load curve curvature trend data; perform load demand classification on the historical power consumption load data based on the load curve curvature trend data to generate power consumption load demand pattern data; The energy storage configuration control module is used to calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain the energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data, thereby generating an energy storage configuration management scheduling strategy; The energy storage configuration optimization module is used to test the energy storage performance of the distributed synchronous energy storage network through the energy storage configuration management scheduling strategy and generate energy storage performance test data; and to adjust the energy storage strategy adaptively according to the energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

[0054] The beneficial effect of the present invention is that by acquiring the information data of photovoltaic power generation energy sites and constructing a distributed synchronous energy storage network, the power generation data of each site can be collected in real time. This provides high-precision basic data support for subsequent energy scheduling and energy storage management. By predicting real-time photovoltaic power generation data, power generation plans can be planned in advance under unstable weather and power load conditions, energy waste can be reduced, energy storage and power scheduling can be optimized, and stable power supply can be ensured. The spatial and temporal feature extraction and load curve curvature trend analysis of historical power load data can reveal the fluctuation pattern of power load and help predict future power demand, which is especially important in the case of large seasonal fluctuations. By classifying the load curve trend data, different power demand patterns can be identified, which helps optimize the allocation and scheduling of power resources, ensures uninterrupted energy supply under different demand patterns, and avoids energy waste. By combining real-time photovoltaic power generation prediction data with power load demand pattern data, the energy supply and demand ratio is calculated to help timely identify the imbalance between supply and demand in the power system. This can effectively predict the situation of excess or insufficient power and take measures in advance. According to the energy supply and demand ratio data, the energy storage system is optimized in real time, and the charging and discharging strategies of the energy storage equipment are adjusted to ensure the balance between photovoltaic power generation and load demand. This not only improves the efficiency of energy use, but also avoids excessive use or inefficient operation of energy storage equipment. Through energy storage performance testing, the system can monitor the efficiency and status of energy storage equipment in real time to ensure that the energy storage system is always in the best operating state. Through data feedback, potential problems of energy storage equipment can be quickly identified and the probability of failure can be reduced. According to the energy storage performance test data, the energy storage strategy is adaptively adjusted to ensure that the system can flexibly respond to different power loads and power generation conditions and optimize power storage and scheduling strategies. Therefore, the present invention improves the accuracy and flexibility of photovoltaic energy storage configuration through real-time photovoltaic power generation prediction, dynamic analysis of load demand and adaptive energy storage strategy optimization.

[0055] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0056] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for configuring energy storage for photovoltaic power generation, characterized in that: The following steps are involved: Step S1: Acquire photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network for the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data for the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; Step S2: Collect historical power load data of the distributed synchronous energy storage network to obtain historical power load data; perform load curve curvature trend analysis on the historical power load data to generate load curve curvature trend data; classify load demand of the historical power load data based on the load curve curvature trend data to generate power load demand pattern data; Step S3: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to generate an energy storage configuration management scheduling strategy; Step S4: Performing energy storage performance test on the distributed synchronous energy storage network through energy storage configuration management scheduling strategy to generate energy storage performance test data; Adaptive energy storage strategy adjustments are made based on energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

2. The energy storage configuration method for photovoltaic power generation according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Acquire photovoltaic power generation energy site information data; Step S12: performing site distribution topology analysis on the photovoltaic power generation energy site information data to generate energy site distribution topology data; constructing a distributed synchronous energy storage network for the photovoltaic power generation energy site information data based on the energy site distribution topology data to obtain a distributed synchronous energy storage network; Step S13: collecting real-time power generation data of the distributed synchronous energy storage network to obtain real-time photovoltaic power generation data; performing data preprocessing on the real-time photovoltaic power generation data to generate standard real-time photovoltaic power generation data, wherein the data preprocessing includes data cleaning, data denoising, missing value filling and data standardization; Step S14: Perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data.

3. The energy storage configuration method for photovoltaic power generation according to claim 2, characterized in that: The distributed synchronous energy storage network construction based on the photovoltaic power generation energy site information data based on the energy site distribution topology data includes: Based on the graph theory method, the energy site distribution topology data is topologically modeled to generate energy site topology model data; the energy site topology model data is node-importantly evaluated to generate topology node importance data; the energy site topology model data is node-labeled according to the topology node importance data to generate key nodes and secondary nodes; The energy site topology model data is edge-connected according to the joint points and secondary nodes to generate an initial energy storage network; the initial energy storage network is deployed with distributed nodes to generate a distributed energy storage network; the distributed energy storage network is synchronized with three layers of data using the photovoltaic power generation energy site information data to generate a distributed synchronous energy storage network, where the three layers of data synchronization include time synchronization, power synchronization, and state synchronization.

4. The energy storage configuration method for photovoltaic power generation according to claim 2, characterized in that: Step S14 includes the following steps: Step S141: extracting basic photovoltaic power generation characteristics from standard real-time photovoltaic power generation data to obtain basic photovoltaic power generation characteristic data; performing time series characteristic analysis on the basic photovoltaic power generation characteristic data to generate photovoltaic power generation time series characteristic data; Step S142: dividing the photovoltaic power generation time series characteristic data into data sets to generate a model training set and a model test set; training the model training set using a long short-term memory neural network algorithm to generate a photovoltaic power generation prediction pre-model; optimizing and iterating the photovoltaic power generation prediction pre-model using the model test set to generate a photovoltaic power generation prediction model; Step S143: Importing the standard real-time photovoltaic power generation data into the photovoltaic power generation prediction model to perform photovoltaic power generation prediction and generate real-time photovoltaic power generation prediction data.

5. The energy storage configuration method for photovoltaic power generation according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: collecting historical power load data of the distributed synchronous energy storage network to obtain historical power load data; Step S22: extracting spatiotemporal load characteristics from historical power load data to obtain spatiotemporal power load characteristic data; analyzing seasonal variation patterns of the spatiotemporal power load characteristic data to generate a load characteristic data set; converting historical power load data into a historical power load curve according to the load characteristic data set to generate a historical power load curve; Step S23: performing load curve curvature trend analysis on the historical power load curve to generate load curve curvature trend data, wherein the load curve curvature trend data includes positive trend curvature, gentle trend curvature and negative trend curvature; Step S24: Segment the historical power load curve based on the positive trend curvature, the gentle trend curvature and the negative trend curvature to generate a high load trend curve, a gentle load trend curve and a low load trend curve; classify the load demand of the historical power load data according to the high load trend curve, the gentle load trend curve and the low load trend curve to generate power load demand pattern data.

6. The energy storage configuration method for photovoltaic power generation according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: Calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain energy supply and demand ratio data; wherein the formula for calculating the energy supply and demand ratio is as follows: In the formula, Expressed as the energy supply-demand ratio, Expressed as time The photovoltaic power generation power, Expressed as time The power output of the energy storage device, Expressed as time Electricity obtained from the grid, Expressed as time The power demand of the electrical load; Step S32: comparing the energy supply-demand ratio data with a preset standard supply-demand ratio threshold, when the energy supply-demand ratio data is greater than the preset standard supply-demand ratio threshold, a power supply surplus mode is generated; when the energy supply-demand ratio data is less than the preset standard supply-demand ratio threshold, a power supply gap mode is generated; Step S33: Optimizing the energy storage system configuration of the distributed synchronous energy storage network based on the power supply surplus mode and the power supply gap mode, and generating photovoltaic power generation energy storage configuration optimization data; Step S34: Dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data, thereby generating an energy storage configuration management scheduling strategy.

7. The energy storage configuration method for photovoltaic power generation according to claim 6, characterized in that: Step S33 includes the following steps: Step S331: confirming the mode time period of the distributed synchronous energy storage network based on the power supply surplus mode and the power supply gap mode to obtain the power supply surplus time period and the power supply gap time period; performing a buffer time period analysis on the power supply surplus time period and the power supply gap time period to generate a power supply change buffer time period; Step S332: setting energy discharge for the power surplus time period to generate an energy discharge curve; setting energy charging for the power shortage time period to generate an energy charging curve; fitting the energy discharge curve and the energy charging curve according to the power supply change buffer period to generate energy loss data for the buffer period; Step S333: Optimize the configuration of energy storage batteries in the buffer period of the distributed synchronous energy storage network based on the energy loss data in the buffer period, thereby generating photovoltaic power generation energy storage configuration optimization data.

8. The energy storage configuration method for photovoltaic power generation according to claim 6, characterized in that: Step S34 includes the following steps: Step S341: performing node energy redundancy calculation on the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data to obtain node energy redundancy data; performing adjacent node status analysis on the distributed synchronous energy storage network according to the node energy redundancy data to generate adjacent node status data; Step S342: using the adjacent node status data to perform energy scheduling weight allocation on the node energy redundancy data to generate energy scheduling weight allocation data; Step S343: perform emergency energy node screening on adjacent node status data through energy scheduling weight distribution data to obtain emergency energy node data; perform emergency scheduling energy storage path analysis on the emergency energy node data according to node energy redundancy data to generate emergency node scheduling energy storage path data; Step S344: Dispatching and uniformly controlling the energy storage of the distributed synchronous energy storage network based on the emergency node scheduling energy storage path data, thereby generating an energy storage configuration management scheduling strategy.

9. The energy storage configuration method for photovoltaic power generation according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: executing the distributed synchronous energy storage network strategy through the energy storage configuration management scheduling strategy to generate photovoltaic power generation energy storage strategy execution data; performing energy storage performance test on the photovoltaic power generation energy storage strategy execution data to generate energy storage performance test data; Step S42: Adaptively adjust the photovoltaic power generation energy storage strategy execution data according to the energy storage performance test data to perform the photovoltaic power generation energy storage configuration optimization operation.

10. A photovoltaic power generation energy storage configuration system, characterized in that: The method for configuring energy storage for photovoltaic power generation according to claim 1 is used to implement the energy storage configuration method for photovoltaic power generation, and the energy storage configuration system for photovoltaic power generation comprises: The power generation prediction module is used to obtain photovoltaic power generation energy site information data; construct a distributed synchronous energy storage network based on the photovoltaic power generation energy site information data to obtain a distributed synchronous energy storage network; collect real-time power generation data from the distributed synchronous energy storage network to obtain standard real-time photovoltaic power generation data; perform photovoltaic power generation prediction on the standard real-time photovoltaic power generation data to generate real-time photovoltaic power generation prediction data; The power consumption trend analysis module is used to collect historical power consumption load data of the distributed synchronous energy storage network to obtain historical power consumption load data; perform load curve curvature trend analysis on the historical power consumption load data to generate load curve curvature trend data; perform load demand classification on the historical power consumption load data based on the load curve curvature trend data to generate power consumption load demand pattern data; The energy storage configuration control module is used to calculate the energy supply and demand ratio of the power load demand pattern data according to the real-time photovoltaic power generation forecast data to obtain the energy supply and demand ratio data; optimize the energy storage system configuration of the distributed synchronous energy storage network through the energy supply and demand ratio data to generate photovoltaic power generation energy storage configuration optimization data; dynamically adjust the energy of the distributed synchronous energy storage network according to the photovoltaic power generation energy storage configuration optimization data, thereby generating an energy storage configuration management scheduling strategy; The energy storage configuration optimization module is used to test the energy storage performance of the distributed synchronous energy storage network through the energy storage configuration management scheduling strategy and generate energy storage performance test data; and to adjust the energy storage strategy adaptively according to the energy storage performance test data to perform energy storage configuration optimization operations for photovoltaic power generation.

Images