Energy storage capacity optimal configuration method based on shared energy storage mode

Abstract

The invention relates to the technical field of energy storage management, in particular to an energy storage capacity optimal configuration method based on a shared energy storage mode. The method comprises the following steps: carrying out target area load characteristic index extraction by using a power dispatching center, and generating energy storage area load characteristic data; node association strength processing is carried out according to the energy storage area load characteristic data, and power grid node association topological data is generated; acquiring regional power supply plan data; performing space-time coupling supply-demand balance analysis on the regional power supply plan data based on the load characteristic data of the energy storage region to generate complementary potential data between corrected regions; and carrying out shared energy storage node capacity optimization according to the complementary potential data among the correction regions, and generating dynamic energy storage capacity configuration data. Through cross-regional electric energy fine scheduling management, the shared energy storage capacity optimization configuration is realized, the problem of power supply and demand matching is effectively solved, and the acceptance capability of a power grid to renewable energy sources is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage management, and in particular to a method for optimizing energy storage capacity configuration based on a shared energy storage mode. Background Art

[0002] With the large-scale grid connection of renewable energy power generation, its intermittency and volatility have brought great challenges to the stable operation of the power system. As an effective solution, energy storage technology can smooth the fluctuations in renewable energy output, improve the stability and reliability of the power grid, and promote the consumption of renewable energy. Shared energy storage refers to the joint investment, construction and use of energy storage facilities by multiple regions or multiple entities, and the optimal allocation and efficient utilization of energy storage resources through cross-regional power dispatching and market trading mechanisms. Through cross-regional dispatching, shared energy storage can serve the power needs of multiple regions at the same time, improve the utilization rate of energy storage facilities, and reduce the investment cost of unit energy storage capacity. For example, when a region is at peak power consumption, energy storage resources in other regions can be called for peak load regulation; when the output of renewable energy in a region is in excess, the excess electricity can be stored in energy storage facilities in other regions to avoid wind and solar power abandonment. However, traditional energy storage capacity optimization configuration methods are often limited to resource allocation within a single region, and lack overall consideration of the cross-regional power market. This limitation makes it impossible to effectively coordinate and manage power demand and supply between different regions, resulting in increased power transmission losses and unreasonable resource allocation. Summary of the invention

[0003] Based on this, the present invention provides a method for optimizing energy storage capacity configuration based on a shared energy storage mode to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for optimizing energy storage capacity configuration based on a shared energy storage mode comprises the following steps:

[0005] Step S1: using the power dispatching center to divide the shared energy storage area to obtain shared energy storage area data; extracting the target area load characteristic index from the shared energy storage area data to generate regional load characteristic data; performing weighted load characteristic value processing according to the regional load characteristic data to generate energy storage area load characteristic data;

[0006] Step S2: constructing a power grid topology structure for the shared energy storage area data to generate cross-regional power grid topology structure data; performing node association strength processing on the cross-regional power grid topology structure data using the energy storage area load characteristic data to generate power grid node association topology data;

[0007] Step S3: Acquire regional power supply plan data; Based on the energy storage regional load characteristic data, use the regional power supply plan data to perform spatiotemporal coupling supply and demand balance analysis on the grid node associated topology data to generate regional supply and demand gap matrix data;

[0008] Step S4: Evaluate the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data to generate corrected inter-regional complementary potential data; perform inter-regional shared scheduling analysis on the corrected inter-regional complementary potential data using the regional power supply plan data to obtain spatiotemporal power flow scheduling map data;

[0009] Step S5: Perform regional energy storage capacity configuration according to the spatiotemporal power flow dispatching graph data to generate regional energy storage capacity quota data; optimize the capacity of shared energy storage nodes according to the regional energy storage capacity quota data to generate dynamic energy storage capacity configuration data.

[0010] The present invention establishes a shared energy storage area through a professional division method of the power dispatching center, providing a clear geographical scope for the integration and optimal configuration of cross-regional energy storage resources. On this basis, the load characteristic indicators of the target area are extracted and regional load characteristic data are generated, which helps to deeply understand the power consumption patterns and load characteristics of each region. The weighted load characteristic value processing further strengthens the identification of key load characteristics, ensuring that the energy storage configuration can more accurately match the actual needs of the region. Through the node association strength processing, not only the interdependence of nodes within the power grid is revealed, but also the power interaction relationship between different regions is reflected. Combined with the regional power supply plan data and the energy storage area load characteristic data, the grid node association topology data is subjected to a spatiotemporal coupling supply and demand balance analysis. By considering the load changes in the time dimension and the energy distribution in space, the matching of power supply and demand is made more accurate, reducing the situation of surplus or shortage, and accurately reflecting the imbalance of power supply and demand in each region on different time scales. Through the evaluation of the potential for inter-regional power mutual assistance and shared scheduling analysis, the possibility of different regions supporting each other to alleviate their respective power supply and demand contradictions is revealed, which promotes the efficient flow and utilization of resources, reduces the demand for overall spare capacity, and further refines the specific plan for mutual assistance actions to ensure that power flows to the most needed places at the right time, enhancing the collaborative operation capability of the entire power grid system. Regional energy storage capacity is configured according to the spatiotemporal power flow scheduling map data to ensure the efficient utilization of energy storage resources and dynamic adaptation to changes in power system demand, achieve reasonable allocation of energy storage resources, and be able to flexibly respond to changes in power demand and improve the operating efficiency of the energy storage system. Therefore, the energy storage capacity optimization configuration method based on the shared energy storage mode of the present invention extracts and weighted processing of load characteristics, takes into account the time and space dimensions of power supply and demand, accurately identifies the supply and demand gaps of each region at different time periods, and utilizes the complementarity of supply and demand between different regions, so as to achieve dynamic and flexible configuration of energy storage resources and adapt to real-time changes in the power system.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: using the power dispatching center to divide the shared energy storage area and obtain shared energy storage area data;

[0013] Step S12: collecting historical power load data according to the shared energy storage area data to obtain original historical power load data;

[0014] Step S13: performing outlier detection on the original historical power load data and filling in missing values ​​to generate regional historical power load data;

[0015] Step S14: extracting the load characteristic index of the target area according to the historical power load data of the area, and generating regional load characteristic data;

[0016] Step S15: Mapping regional characteristic indicators to the shared energy storage regional data through regional load characteristic data to generate regional multi-dimensional characteristic matrix data;

[0017] Step S16: performing Z-sore standardization on the regional multi-dimensional feature matrix data to obtain a standardized regional feature matrix; performing feature index weight evaluation based on the standardized regional feature matrix to obtain feature index weight vector data;

[0018] Step S17: Use the characteristic index weight vector data to perform weighted load characteristic value processing on the standardized regional characteristic matrix to generate energy storage regional load characteristic data.

[0019] The present invention utilizes the power dispatching center to divide the shared energy storage area, and can reasonably divide the service scope of the energy storage resources. According to the shared energy storage area data, the historical power load data is collected, which can reflect the power consumption characteristics and change trends of each region, and provide a reliable data basis for the optimal configuration of the energy storage capacity. The original historical power load data is detected for abnormal values ​​and filled with missing values, which can eliminate the noise and errors in the data and ensure the accuracy of subsequent analysis. According to the regional historical power load data, the load characteristic index of the target area is extracted, which can deeply understand the characteristics and change laws of the power demand of each region, and reflect the power consumption mode and demand characteristics of each region. The regional characteristic index mapping of the shared energy storage area data by the regional load characteristic data forms a data structure that comprehensively reflects the regional characteristics, which is helpful to comprehensively evaluate the energy storage demand from multiple angles. Through the Z-score standardization processing, the dimensional influence between different indicators is eliminated. The standardized regional characteristic matrix is ​​weighted by the characteristic index weight vector, which provides a precise guide for the optimal configuration of the energy storage capacity and ensures that the resource allocation is highly matched with the actual regional demand.

[0020] Preferably, step S15 comprises the following steps:

[0021] Step S151: Perform dynamic time warping on the regional load characteristic data, and compare the characteristics between regions to generate characteristic difference data between regions;

[0022] Step S152: Calculate the time series feature similarity based on the feature difference data between regions to generate regional load similarity data;

[0023] Step S153: performing load pattern clustering analysis based on regional load similarity data, and performing regional clustering division to generate regional load cluster data;

[0024] Step S154: extracting cluster typical load curves from regional load cluster data to obtain cluster typical load curve data;

[0025] Step S155: performing regional mapping on the shared energy storage regional data through the cluster typical load curve data to generate regional-cluster characteristic mapping data;

[0026] Step S156: performing multi-dimensional feature vector processing according to the region-cluster feature mapping data to obtain multi-dimensional feature vector data;

[0027] Step S157: constructing a regional characteristic matrix for the regional load characteristic data through the multi-dimensional characteristic vector data to generate regional multi-dimensional characteristic matrix data.

[0028] The present invention processes regional load characteristic data through dynamic time warping, which can effectively eliminate the nonlinear time distortion effect in time series data, so that the load characteristics between different regions are comparable on a time scale. By calculating the similarity of the time series characteristics of regional loads, the similarity of load behavior between regions is quantified, and this data helps to identify those regions with similar load patterns. Based on the similarity data, the regions are logically grouped to form load clusters with common characteristics, which can be more targeted to specific types of load demand groups. By extracting the typical load curve data of each cluster, the basic pattern of load behavior within each cluster is represented. The shared energy storage area data is regionally mapped through the cluster typical load curve data, and the corresponding relationship between the region and the load pattern cluster is established. Through multi-dimensional feature vector matrix processing, the power consumption characteristics of each region can be fully described.

[0029] Preferably, step S2 comprises the following steps:

[0030] Step S21: using the power dispatching center to construct the grid topology structure for the shared energy storage area data, and initialize the line loss weight to generate cross-regional grid topology structure data;

[0031] Step S22: weighting node power loads on cross-regional power grid topology data using energy storage area load characteristic data to generate load-weighted power grid topology data;

[0032] Step S23: performing node betweenness centrality statistics according to the load-weighted power grid topology data to obtain node betweenness centrality data;

[0033] Step S24: Calculate the inter-regional electrical distance according to the load-weighted power grid topology data to generate inter-regional electrical distance data;

[0034] Step S25: Based on the inter-regional electrical distance data, the load-weighted power grid topology data is processed for node association strength through the node betweenness centrality data to generate power grid node association topology data.

[0035] The present invention constructs the power grid topology structure of the shared energy storage area through the professional technology of the power dispatching center, and completes the initialization of the line loss weight, ensuring the reality and efficiency of the resource dispatching configuration. By combining the load characteristic data of the energy storage area to weight the node power load of the power grid topology structure, it essentially gives each node in the power grid an importance level that matches the power load it carries. Statistical node betweenness centrality reveals which nodes play the role of bridges or hubs in the power grid, which helps to give priority to these key nodes in the energy storage configuration to ensure efficient and stable power flow. Electrical distance not only takes into account geographical distance, but also incorporates grid structure and impedance factors. It is an important parameter for evaluating the cost and feasibility of power exchange between regions. Through node association strength processing, it can better reflect the power interaction relationship between different regions and reflect the closeness between nodes in load transfer and power flow.

[0036] Preferably, step S3 comprises the following steps:

[0037] Step S31: performing a multi-step forecast of regional time series load according to the energy storage regional load characteristic data to obtain corrected load forecast curve data;

[0038] Step S32: Acquire regional power supply plan data, including power production plans for each region and renewable energy output estimation data;

[0039] Step S33: using the regional power supply plan data to evaluate the regional power supply capacity of the shared energy storage regional data, and generating regional power supply data;

[0040] Step S34: performing spatiotemporal coupled supply and demand balance analysis on regional power supply data by modifying load forecast curve data to generate regional spatiotemporal supply and demand balance data;

[0041] Step S35: Calculate the supply-demand gap probability based on the regional spatiotemporal supply-demand balance data to generate regional supply-demand gap probability data;

[0042] Step S36: Perform regional supply and demand gap risk assessment based on regional supply and demand gap probability data, and mark regional supply and demand gaps on regional spatiotemporal supply and demand balance data to generate regional supply and demand gap matrix data.

[0043] The present invention provides a power demand forecast for each energy storage area in a certain period of time in the future through a multi-step prediction of time-series loads by regional load characteristic data. The regional power supply plan data obtained, especially the power production plan and renewable energy output estimation of each region, provide detailed information on the supply side for the energy storage area. The power supply capacity of the shared energy storage area is evaluated using the energy supply plan data, and the regional power supply data generated reveals the power supply potential and limitations of each region in a specific period of time. By correcting the load forecast curve data and performing spatiotemporal coupling supply and demand balance analysis on the regional power supply data, the supply and demand matching status of each region in different time periods can be accurately identified, and the specific time and space distribution of the supply and demand imbalance can be revealed. The supply and demand gap probability is calculated based on the regional spatiotemporal supply and demand balance data to quantify the risk degree of supply and demand imbalance. By calculating the supply and demand gap probability, the reliability of power supply in each region can be evaluated. The regional supply and demand gap risk assessment is performed based on the regional supply and demand gap probability data, and the supply and demand risk status of each region is intuitively displayed, which can clearly identify the areas and time periods that need to be focused on.

[0044] Preferably, step S31 includes the following steps:

[0045] Step S311: performing variational modal decomposition on the energy storage regional load characteristic data to generate regional load modal component data;

[0046] Step S312: using a preset time series prediction model to perform intelligent prediction of modal components on regional load modal component data to generate load modal component prediction data;

[0047] Step S313: performing regional component prediction superposition according to the load modal component prediction data, and dynamically reconstructing the load curve to generate regional load prediction curve data;

[0048] Step S314: performing single-region load confidence estimation according to the regional load forecast curve data to generate a single-region load forecast confidence interval;

[0049] Step S315: using the single-region load forecast confidence interval to perform load forecast error correlation analysis on the regional load forecast curve data to generate forecast error correlation data;

[0050] Step S316: Perform curve confidence correction on the regional load forecast curve data using the forecast error correlation data to generate corrected load forecast curve data.

[0051] The present invention effectively decomposes complex load data into several relatively independent and physically meaningful modal components by performing variational mode decomposition (VMD) on the energy storage regional load characteristic data, which helps to deeply understand the inherent laws and driving factors of load changes. The regional load modal component data is predicted using a preset time series prediction model (such as ARIMA, LSTM, etc.), and the generated load modal component prediction data can capture the future change trend of each mode, improve the accuracy and adaptability of the prediction, and better handle the nonlinear, periodic and trend characteristics of the load data by decomposing and predicting each part separately. The prediction results of each modal component are combined to form a complete load prediction curve, which can better reflect the overall trend and local characteristics of load changes through dynamic reconstruction. According to the regional load prediction curve data, a single-region load confidence estimation is performed to quantify the uncertainty of the prediction result. By calculating the confidence interval, the reliability of the prediction can be evaluated. The load prediction error correlation analysis is performed using the single-region load prediction confidence interval, which reveals the relationship between the prediction errors in different regions or at different time points. The regional load forecast curve data is corrected for curve confidence through the prediction error correlation data. Not only the uncertainty of the prediction itself is taken into account, but also the prediction results are corrected through correlation analysis, which improves the robustness and practicality of the prediction.

[0052] Preferably, step S4 comprises the following steps:

[0053] Step S41: evaluating the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data and the grid node association topology data, and generating corrected inter-regional complementary potential data;

[0054] Step S42: constructing typical supply and demand scenarios for the modified inter-regional complementary potential data through regional power supply plan data to obtain multi-time scale supply and demand scenario data;

[0055] Step S43: constructing a power network simulation model using the grid node associated topology data;

[0056] Step S44: using the power network simulation model to perform shared power dispatch simulation on the multi-time scale supply and demand scenario data, and perform dispatch flow calculation to generate multi-scenario power dispatch simulation data;

[0057] Step S45: Perform inter-regional shared scheduling analysis based on the multi-scenario power scheduling simulation data, and perform power flow spatiotemporal graph processing to obtain spatiotemporal power flow scheduling graph data.

[0058] By comprehensively considering the supply-demand gap and network topology, the present invention not only identifies which regions have supply-demand imbalances, but also quantifies the ability of regions to support each other, and can accurately identify the feasibility and potential of inter-regional power mutual assistance. By constructing typical supply and demand scenarios with regional power supply plan data to correct inter-regional complementary potential data, various supply and demand situations can be simulated. By constructing multi-time scale supply and demand scenarios, changes in power supply and demand in different time periods can be fully considered. The power network simulation model constructed using grid node associated topology data can accurately reflect the structural characteristics and operating rules of the power grid. The shared power dispatch simulation of multi-time scale supply and demand scenario data is performed through the power network simulation model, revealing the operating status and power flow pattern of the power grid under different supply and demand situations. The inter-regional shared dispatch analysis based on multi-scenario power dispatch simulation data intuitively displays the flow pattern of power in time and space, and can clearly present the power transmission situation in different times and regions.

[0059] Preferably, step S41 includes the following steps:

[0060] Step S411: performing regional gap time series decomposition on the regional supply-demand gap matrix data to generate regional gap time series data;

[0061] Step S412: performing supply-demand fluctuation correlation analysis based on the regional gap time series data to generate a supply-demand gap fluctuation correlation coefficient;

[0062] Step S413: Calculate the regional supply and demand mutual information of the grid node associated topology data through the supply and demand gap fluctuation correlation coefficient, and evaluate the power complementarity potential to obtain the inter-regional supply and demand complementarity potential data;

[0063] Step S414: performing transmission capacity constraint analysis based on the grid node associated topology data to generate regional transmission capacity constraint data; performing distance attenuation factor processing based on the grid node associated topology data to generate inter-regional distance attenuation factors;

[0064] Step S415: using the inter-regional distance attenuation factor and the regional transmission capacity constraint data to correct the inter-regional supply-demand complementary potential value, and obtain corrected inter-regional complementary potential data.

[0065] The present invention converts complex supply and demand gap information into a time series form by performing time series decomposition on the regional supply and demand gap matrix data. Through time series decomposition, the time variation characteristics of the supply and demand gaps in each region can be more clearly displayed, which is helpful to identify supply and demand complementary opportunities between different regions. Supply and demand fluctuation correlation analysis is performed based on the regional gap time series data to quantify the correlation degree of supply and demand gap changes between different regions. By calculating the correlation coefficient, regional pairs with complementary potential can be identified. The regional supply and demand mutual information is calculated for the grid node associated topology data through the supply and demand gap fluctuation correlation coefficient, and the network topology and supply and demand fluctuation characteristics are comprehensively considered to evaluate the actual complementary capacity between regions. By analyzing the transmission capacity constraints and distance attenuation factors, the feasibility and efficiency of power transmission can be more truly reflected. The complementary potential value of the regional supply and demand complementary potential data is corrected using the inter-regional distance attenuation factor and the regional transmission capacity constraint data, and the theoretical complementary potential is combined with the actual grid restrictions. By considering the distance attenuation and transmission capacity constraints for correction, a more practical complementary potential evaluation result can be obtained.

[0066] Preferably, step S5 comprises the following steps:

[0067] Step S51: performing regional energy storage capacity configuration on the modified inter-regional complementary potential data through spatiotemporal power flow dispatching graph data to generate regional energy storage capacity quota data;

[0068] Step S52: using the regional energy storage capacity quota data to perform cross-regional power transmission minimum loss path planning on the spatiotemporal power flow scheduling map data, and obtaining energy storage regional scheduling path data;

[0069] Step S53: Optimizing the state of energy storage nodes in a control cycle by using a preset model predictive control algorithm for the energy storage area scheduling path data and the regional energy storage capacity quota data to obtain a dynamic energy storage node control strategy;

[0070] Step S54: optimizing the capacity of the shared energy storage node according to the dynamic energy storage node control strategy to obtain energy storage capacity optimization configuration data;

[0071] Step S55: Perform an operation benefit evaluation based on the energy storage capacity optimization configuration data, and perform rolling adjustment on the capacity configuration to generate dynamic energy storage capacity configuration data.

[0072] The present invention configures regional energy storage capacity by modifying the complementary potential data between regions through spatiotemporal power flow scheduling graph data. It can reasonably regulate the energy storage capacity of each region according to the actual situation of power flow and complementary potential, and can better match the spatiotemporal distribution of power demand and supply. The regional energy storage capacity quota data is used to plan the minimum loss path of cross-regional power transmission for the spatiotemporal power flow scheduling graph data, which can reduce the energy loss in the power transmission process and improve the overall efficiency of the system. The preset model predictive control algorithm is used to comprehensively consider the energy storage regional scheduling path data and the regional energy storage capacity quota data, realize the refined management and dynamic scheduling of energy storage resources, improve the response speed and regulation ability of the energy storage system, and ensure the stable operation of the power system and the real-time balance of energy supply and demand. The capacity of the shared energy storage node is optimized according to the dynamic energy storage node control strategy, ensuring that the capacity setting of each energy storage node can not only meet the power regulation needs within and across regions, but also adapt to the ever-changing power market environment, avoid the problem of excess or insufficient capacity, and improve the utilization efficiency of energy storage resources.

[0073] Preferably, step S51 includes the following steps:

[0074] Step S511: identifying target energy storage nodes according to the spatiotemporal power flow dispatch graph data to obtain target energy storage optimization node data;

[0075] Step S512: collecting energy storage capacity constraint parameters according to the target energy storage optimization node data, and setting energy storage capacity constraint conditions to obtain energy storage capacity constraint condition data;

[0076] Step S513: using the energy storage capacity constraint condition data to perform preliminary configuration of regional energy storage capacity on the modified inter-regional complementary potential data, and generating preliminary regional energy storage capacity configuration data;

[0077] Step S514: performing shared energy storage capacity allocation processing on the modified inter-regional complementary potential data through the spatiotemporal power flow dispatching graph data to obtain shared energy storage capacity allocation problem data;

[0078] Step S515: Markov decision processing is performed on the shared energy storage capacity allocation problem data, and dynamic capacity quota optimization is performed on the preliminary regional energy storage capacity configuration data to generate regional energy storage capacity quota data.

[0079] The present invention identifies the target energy storage node according to the spatiotemporal power flow dispatching graph data, can accurately locate the energy storage node that needs to be optimized, and can identify the energy storage location that is most critical to the system balance and efficiency by analyzing the spatiotemporal characteristics of power flow. According to the target energy storage optimization node data, the energy storage capacity constraint parameter collection is carried out. By setting appropriate constraints, the capacity configuration scheme that exceeds the physical or economic limits can be avoided, ensuring that the final optimization result meets both technical requirements and economic benefits. The energy storage capacity constraint condition data is used to perform preliminary configuration of regional energy storage capacity on the modified inter-regional complementary potential data. Through this preliminary configuration, a reasonable starting point can be provided for the subsequent optimization process, the optimization convergence speed can be accelerated, and the feasibility of the initial scheme can be ensured. The shared energy storage capacity allocation processing is performed on the modified inter-regional complementary potential data through the spatiotemporal power flow dispatching graph data, and the energy storage capacity allocation problem is converted into a quantifiable optimization problem. By considering the spatiotemporal power flow and regional complementary potential, the capacity allocation demand under the shared energy storage mode can be more accurately described. Through Markov decision processing, the randomness and time-varying nature of the power system can be considered to achieve a more flexible and adaptable capacity configuration. Dynamic capacity quota optimization can adjust the energy storage capacity allocation of each area in real time according to changes in system status, thereby maximizing the benefits of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 A schematic diagram of the steps of a method for optimizing energy storage capacity configuration based on a shared energy storage mode according to the present invention;

[0081] Figure 2 for Figure 1 Detailed implementation steps of step S3 in FIG.

[0082] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0083] 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

[0084] 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.

[0085] 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.

[0086] 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.

[0087] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for optimizing energy storage capacity configuration based on a shared energy storage mode, comprising the following steps:

[0088] Step S1: using the power dispatching center to divide the shared energy storage area to obtain shared energy storage area data; extracting the target area load characteristic index from the shared energy storage area data to generate regional load characteristic data; performing weighted load characteristic value processing according to the regional load characteristic data to generate energy storage area load characteristic data;

[0089] Step S2: constructing a power grid topology structure for the shared energy storage area data to generate cross-regional power grid topology structure data; performing node association strength processing on the cross-regional power grid topology structure data using the energy storage area load characteristic data to generate power grid node association topology data;

[0090] Step S3: Acquire regional power supply plan data; Based on the energy storage regional load characteristic data, use the regional power supply plan data to perform spatiotemporal coupling supply and demand balance analysis on the grid node associated topology data to generate regional supply and demand gap matrix data;

[0091] Step S4: Evaluate the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data to generate corrected inter-regional complementary potential data; perform inter-regional shared scheduling analysis on the corrected inter-regional complementary potential data using the regional power supply plan data to obtain spatiotemporal power flow scheduling map data;

[0092] Step S5: Perform regional energy storage capacity configuration according to the spatiotemporal power flow dispatching graph data to generate regional energy storage capacity quota data; optimize the capacity of shared energy storage nodes according to the regional energy storage capacity quota data to generate dynamic energy storage capacity configuration data.

[0093] In the embodiment of the present invention, reference Figure 1 The above is a schematic diagram of the steps of a method for optimizing energy storage capacity configuration based on a shared energy storage mode of the present invention. In this embodiment, the method for optimizing energy storage capacity configuration based on a shared energy storage mode includes the following steps:

[0094] Step S1: using the power dispatching center to divide the shared energy storage area to obtain shared energy storage area data; extracting the target area load characteristic index from the shared energy storage area data to generate regional load characteristic data; performing weighted load characteristic value processing according to the regional load characteristic data to generate energy storage area load characteristic data;

[0095] In an embodiment of the present invention, the regional power data provided by the power dispatching center, such as the load curves, power generation output data, geographical location, power grid structure and other information of each region, is used to divide the entire power system into several shared energy storage areas through cluster analysis or graph theory methods. For example, the K-Means clustering algorithm can be used to divide the power system into three shared energy storage areas: area A, area B, and area C, based on the indicators such as the similarity of regional load curves and the capacity of inter-regional transmission lines as the clustering basis. The historical load data of each shared energy storage area is analyzed to extract the load characteristic indicators of the target area, including peak load, valley load, load rate, load growth rate, load fluctuation rate, etc. For example, the Pandas library in Python is used to read the load curve data of area A, calculate its daily peak load, daily valley load, and calculate the peak-valley difference and daily average load, so as to obtain the load rate and load growth rate of area A. The load characteristic indicator data of all regions are organized into tables to form regional load characteristic data. According to the importance of each indicator, for example, the daily peak-to-valley difference weight is set to 0.5, the load fluctuation weight is set to 0.3, and the average load weight is set to 0.2, and the regional load characteristic data is weighted to obtain the energy storage regional load characteristic data.

[0096] Step S2: constructing a power grid topology structure for the shared energy storage area data to generate cross-regional power grid topology structure data; performing node association strength processing on the cross-regional power grid topology structure data using the energy storage area load characteristic data to generate power grid node association topology data;

[0097] In an embodiment of the present invention, based on the power grid topology data provided by the power dispatching center, such as the voltage level of each node, line parameters, etc., cross-regional power grid topology data is constructed. For example, using power system simulation software such as PowerWorld or Matpower, input power grid line and node information, and construct a cross-regional power grid model including region A, region B, and region C. According to the energy storage regional load characteristic data generated in step S1, such as the maximum load difference between regions, load correlation, etc., the cross-regional power grid topology data is processed for node correlation strength. For example, the PageRank algorithm is used to calculate the importance of each node in the power grid, and combined with the load correlation between regions, weights are assigned to the connections between nodes. The larger the weight, the higher the node correlation strength. Finally, the power grid node correlation topology data is generated, which reflects the power grid connection strength and load correlation degree between different regions. For example, if the load correlation between region A and region B is high, the weights of the nodes and lines connecting the two regions will be high.

[0098] Step S3: Acquire regional power supply plan data; Based on the energy storage regional load characteristic data, use the regional power supply plan data to perform spatiotemporal coupling supply and demand balance analysis on the grid node associated topology data to generate regional supply and demand gap matrix data;

[0099] In an embodiment of the present invention, regional power supply plan data is obtained, such as power generation plans, load forecasts, renewable energy output forecasts, etc. for each region. For example, the power supply plan data for regions A, B, and C in the next week is obtained from the power dispatching center, including the power generation plans, load forecasts, wind and solar power abandonment data, etc. for each region. Based on the energy storage regional load characteristic data and regional power supply plan data, a linear programming or mixed integer programming model is used to perform a spatiotemporal coupling supply and demand balance analysis on the topological data associated with the power grid nodes. For example, an optimization model is constructed, and the objective function is to minimize the system operating cost, and the constraints include node power balance, line flow restrictions, energy storage charging and discharging constraints, etc. By solving the optimization model, the supply and demand gaps of each region in different time periods are obtained, and they are organized into regional supply and demand gap matrix data. For example, the model will calculate the supply and demand gap of region A every hour in the next week and store it in a matrix.

[0100] Step S4: Evaluate the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data to generate corrected inter-regional complementary potential data; perform inter-regional shared scheduling analysis on the corrected inter-regional complementary potential data using the regional power supply plan data to obtain spatiotemporal power flow scheduling map data;

[0101] In an embodiment of the present invention, based on the regional supply and demand gap matrix data, for example, the supply and demand gap situation of different regions in different time periods is analyzed to evaluate the potential for inter-regional power mutual assistance. For example, if region A has a power surplus during a certain period of time, and region B has a power gap, the potential for power transmission from region A to region B can be evaluated. On this basis, combined with factors such as the capacity of inter-regional transmission lines and grid stability, the inter-regional complementary potential is corrected to generate corrected inter-regional complementary potential data. Using the regional power supply plan data and the corrected inter-regional complementary potential data, inter-regional shared scheduling analysis is performed. For example, dynamic programming or genetic algorithms are used to optimize the inter-regional power exchange scheme to maximize system benefits, such as reducing system operating costs and improving new energy absorption capacity. Finally, spatiotemporal power flow scheduling map data is obtained, which reflects the power exchange situation between different regions in different time periods, as well as the scheme for energy storage systems to participate in grid scheduling. For example, through the optimization algorithm, it is possible to obtain the amount of electricity transmitted from area A to area B every hour in the next week, as well as the charging and discharging power of the energy storage system in different time periods, and display this information in the form of a graph. For example, the Matplotlib library in Python can be used to draw a graph showing the direction and size of power flow changing over time.

[0102] Step S5: Perform regional energy storage capacity configuration according to the spatiotemporal power flow dispatching graph data to generate regional energy storage capacity quota data; optimize the capacity of shared energy storage nodes according to the regional energy storage capacity quota data to generate dynamic energy storage capacity configuration data.

[0103] In an embodiment of the present invention, according to the spatiotemporal power flow scheduling graph data, for example, the energy storage charging and discharging requirements of different regions in different time periods are analyzed to configure the regional energy storage capacity. For example, according to the energy storage charging and discharging requirements of region A during the peak period, the energy storage capacity quota of region A is determined. Finally, regional energy storage capacity quota data is generated. According to the regional energy storage capacity quota data, the capacity of the shared energy storage node is optimized. For example, considering the geographical location, grid connection, energy storage technology type and other factors of the shared energy storage node, the capacity configuration scheme of the shared energy storage node is optimized by integer programming. For example, the energy storage capacity of region A is allocated to multiple shared energy storage nodes within region A, and the capacity of each node is adjusted according to the characteristics of the node. Finally, dynamic energy storage capacity configuration data is generated, which reflects the capacity allocation scheme of shared energy storage nodes in different regions and the changes of node capacity over time. For example, the optimization algorithm calculates the optimal capacity of each energy storage node in region A, and formulates the charging and discharging plan of each node based on the load forecast and renewable energy output forecast for the next week to form dynamic energy storage capacity configuration data.

[0104] Preferably, step S1 comprises the following steps:

[0105] Step S11: using the power dispatching center to divide the shared energy storage area and obtain shared energy storage area data;

[0106] Step S12: collecting historical power load data according to the shared energy storage area data to obtain original historical power load data;

[0107] Step S13: performing outlier detection on the original historical power load data and filling in missing values ​​to generate regional historical power load data;

[0108] Step S14: extracting the load characteristic index of the target area according to the historical power load data of the area, and generating regional load characteristic data;

[0109] Step S15: Mapping regional characteristic indicators to the shared energy storage regional data through regional load characteristic data to generate regional multi-dimensional characteristic matrix data;

[0110] Step S16: performing Z-sore standardization processing on the regional multi-dimensional feature matrix data to obtain a standardized regional feature matrix; performing feature index weight evaluation according to the standardized regional feature matrix to obtain feature index weight vector data;

[0111] Step S17: Use the characteristic index weight vector data to perform weighted load characteristic value processing on the standardized regional characteristic matrix to generate energy storage regional load characteristic data.

[0112] In an embodiment of the present invention, the SCADA system of the power dispatching center is used to obtain information such as load data, power generation data, and power grid topology of each region, for example, the power load data, power generation data, and transmission line information of each province and city in East China are obtained. Then, the regional load data is clustered and analyzed based on the K-means clustering algorithm, and the regions with similar load characteristics are divided into a shared energy storage region to obtain shared energy storage regional data. The historical power load data of each region in the region is collected from the historical database of the power dispatching center, for example, the power load data of each hour in the past three years is collected. The collected data includes information such as timestamps and load values. The original historical power load data is detected for outliers, for example, using the 3σ criterion to identify data that exceeds the mean ±3 times the standard deviation as outliers, and remove or correct them. For example, if the power load data of a certain hour suddenly soars to several times the normal value, the data is considered to be an outlier and needs to be removed or corrected. At the same time, the missing values ​​in the data are filled, for example, using linear interpolation or mean filling to complete the missing data. According to the historical power load data of the region, the load characteristic indicators of each region are extracted, such as peak load, valley load, load rate, load growth rate, load fluctuation rate, etc. For example, the historical load data of region A is read using the Pandas library in Python, and its daily peak load, daily valley load, peak-to-valley difference, and daily average load are calculated, thereby obtaining the load rate and load growth rate of region A. The characteristic indicators such as the load rate and load growth rate of region A are associated with the geographical location, grid structure and other information of region A, and finally the regional multi-dimensional characteristic matrix data is generated, for example, a matrix is ​​generated in which each row represents a shared energy storage area and each column represents a load characteristic indicator, such as daily peak-to-valley difference rate, load fluctuation rate, etc. The regional multi-dimensional characteristic matrix data is subjected to Z-score standardization, that is, the value of each characteristic indicator is subtracted from its mean, and then divided by its standard deviation to obtain a standardized regional characteristic matrix. Then, according to the standardized regional characteristic matrix, for example, the entropy weight method or principal component analysis method is used to evaluate the weight of the characteristic indicator to obtain the characteristic indicator weight vector data. For example, the weights of indicators such as the daily peak-to-valley difference rate and the load fluctuation rate are calculated. For example, the weight of the daily peak-to-valley difference rate is 0.5, and the weight of the load fluctuation rate is 0.3. For example, the standardized daily peak-to-valley difference rate of each shared energy storage area is multiplied by its weight 0.5, and the standardized load fluctuation rate is multiplied by its weight 0.3, and then all weighted characteristic values ​​are added together to obtain the weighted load characteristic value of the shared energy storage area. Finally, the energy storage area load characteristic data is generated, for example, a data set containing the weighted load characteristic values ​​of each shared energy storage area is generated.

[0113] Preferably, step S15 comprises the following steps:

[0114] Step S151: Perform dynamic time warping on the regional load characteristic data, and compare the characteristics between regions to generate characteristic difference data between regions;

[0115] Step S152: Calculate the time series feature similarity based on the feature difference data between regions to generate regional load similarity data;

[0116] Step S153: performing load pattern clustering analysis based on regional load similarity data, and performing regional clustering division to generate regional load cluster data;

[0117] Step S154: extracting cluster typical load curves from regional load cluster data to obtain cluster typical load curve data;

[0118] Step S155: performing regional mapping on the shared energy storage regional data through the cluster typical load curve data to generate regional-cluster characteristic mapping data;

[0119] Step S156: performing multi-dimensional feature vector processing according to the region-cluster feature mapping data to obtain multi-dimensional feature vector data;

[0120] Step S157: constructing a regional characteristic matrix for the regional load characteristic data through the multi-dimensional characteristic vector data to generate regional multi-dimensional characteristic matrix data.

[0121] In an embodiment of the present invention, the regional load characteristic data obtained in step S14, such as the peak load, valley load, load rate, etc. of each region, are used to align and compare time series using a dynamic time warping (DTW) algorithm. The DTW algorithm can effectively handle problems such as inconsistent length of time series data and time offset, and calculate the similarity between load curves in different regions. For example, the DTW library in Python can be used to calculate the DTW distance between the historical load curves of region A and region B. Then, inter-regional feature comparison is performed based on the DTW distance, for example, the DTW distance between each region and all other regions is calculated, and inter-regional feature difference data is generated, such as generating a matrix in which each element represents the DTW distance between the two regions. A distance-based similarity calculation method is used, for example, 1 minus the ratio of the DTW distance to the maximum DTW distance can be used as the similarity, and the higher the similarity, the more similar the regional load characteristics are, and finally regional load similarity data is generated, such as generating a matrix in which each element represents the load similarity score between the two regions. Using the K-Means clustering algorithm, all regions are used as initial data points, and the number of clusters is set to 3. Through iterative calculation, the regions are divided into three different clusters to obtain the regional clustering cluster division results. For example, regions A and B are divided into the same cluster, and region C is divided into another cluster. The clustering cluster division results are organized into a table to generate regional load cluster data. For example, the cluster information to which the region belongs is used to extract the cluster typical load curve. For each cluster, the average or median of the historical load curves of all regions in the cluster can be calculated as the typical load curve of the cluster. For example, for cluster 1, the average of the historical load curves of all regions in the cluster can be calculated as the typical load curve of cluster 1. Finally, the cluster typical load curve data is obtained, for example, 3 typical load curves are obtained, representing the load patterns of the 3 clusters respectively. For each shared energy storage area, the DTW distance between its historical load curve and the typical load curve of each cluster is calculated, and the area is mapped to the cluster with the smallest DTW distance. For example, if the DTW distance between the historical load curve of region A and the typical load curve of cluster 2 is the smallest, region A is mapped to cluster 2. Finally, the region-cluster feature mapping data is generated. A multi-dimensional feature vector is extracted for each region. The geographical location, grid structure, load type, cluster number and other information of the region can be used as features to form a multi-dimensional feature vector. For example, the regional load characteristic data (such as daily peak-to-valley difference rate, load fluctuation rate, etc.) is concatenated with the unique hot encoding of the cluster to which the region belongs to generate a new matrix, in which each row represents a shared energy storage area, and each column represents a feature, including load characteristics such as daily peak-to-valley difference rate, load fluctuation rate and the unique hot encoding of the cluster to which the region belongs.

[0122] Preferably, step S2 comprises the following steps:

[0123] Step S21: using the power dispatching center to construct the grid topology structure for the shared energy storage area data, and initialize the line loss weight to generate cross-regional grid topology structure data;

[0124] Step S22: weighting node power loads on cross-regional power grid topology data using energy storage area load characteristic data to generate load-weighted power grid topology data;

[0125] Step S23: performing node betweenness centrality statistics according to the load-weighted power grid topology data to obtain node betweenness centrality data;

[0126] Step S24: Calculate the inter-regional electrical distance according to the load-weighted power grid topology data to generate inter-regional electrical distance data;

[0127] Step S25: Based on the inter-regional electrical distance data, the load-weighted power grid topology data is processed for node association strength through the node betweenness centrality data to generate power grid node association topology data.

[0128] In an embodiment of the present invention, the SCADA system of the power dispatching center is used to obtain the geographical location information, line connection information, line parameters (such as resistance, reactance) and other data of each node in the shared energy storage area. Then, a graph theory library such as NetworkX is used to construct a cross-regional power grid topology structure, in which the node represents a power facility such as a substation or a power plant, and the edge represents a transmission line. Next, the line loss weight of each line is initialized, for example, the resistance and reactance of the line are calculated according to parameters such as line length and conductor model, and the resistance value is used as the line loss weight. The average load value of each node can be used as the weight of the node, and the weight is assigned to the corresponding node in the power grid topology diagram. For example, if the average load of node A is 100MW, the weight of node A in the graph structure is set to 100. Finally, load-weighted power grid topology data is generated, such as a graph structure data containing node information, edge information, line loss weight and node load weight. Network analysis libraries such as NetworkX can be used to take load-weighted power grid topology data as input, calculate the betweenness centrality of each node, and obtain node betweenness centrality data, such as the betweenness centrality of node 1 is 0.5, the betweenness centrality of node 2 is 0.3, etc. The Dijkstra algorithm can be used to calculate the weighted average of the line loss weight on the shortest path between two regions, where the weight is the line loss weight. For example, if the line loss weights on the shortest path between region A and region B are 0.1, 0.2, and 0.3, respectively, then the electrical distance between region A and region B is (0.1+0.2+0.3) / 3=0.2. Finally, inter-regional electrical distance data is generated, such as a matrix in which each element represents the electrical distance between two regions. For example, if the electrical distance between node A and node B is 0.2, the betweenness centrality of node A is 0.5, and the betweenness centrality of node B is 0.3, then the association strength between node A and node B is 1 / 0.2×0.5×0.3=0.75. The calculated association strength is used as the edge weight to update the load-weighted power grid topology data. Finally, the power grid node association topology data is generated, such as a graph structure data including node information, edge information, line loss weight, node load weight and node association strength.

[0129] As an example of the present invention, refer to Figure 2 As shown, Figure 1 Detailed implementation steps of step S3 in the flowchart, in this example, step S3 includes:

[0130] Step S31: performing a multi-step forecast of regional time series load according to the energy storage regional load characteristic data to obtain corrected load forecast curve data;

[0131] In an embodiment of the present invention, for example, historical load data and meteorological data of each region, time series prediction models such as LSTM and Prophet are used to perform multi-step prediction of regional time series load. For example, the statsmodels library or prophet library in Python can be used to build a time series prediction model, and historical load data and meteorological data can be used as input to predict the power load of each region in the future (for example, the next 24 hours). The prediction result can be the load prediction value for each time step and the confidence interval of the prediction value. Since there are errors in the prediction results, the prediction results need to be corrected, for example, the prediction results are calibrated according to the historical prediction errors. Finally, the corrected load forecast curve data is obtained, such as a data set containing the load forecast values ​​of each region in the next 24 hours.

[0132] Step S32: Acquire regional power supply plan data, including power production plans for each region and renewable energy output estimation data;

[0133] In an embodiment of the present invention, regional power supply plan data is obtained from a power dispatching center or relevant departments, including power production plans of each region, such as power generation plans of thermal power, hydropower, nuclear power, etc., and renewable energy output estimation data, such as forecast data of power generation of solar energy, wind power, etc. For example, the planned power generation of thermal power, hydropower, nuclear power, etc., and the forecast power generation of solar energy, wind power, etc. in each region within the next 24 hours can be obtained. For example, the power supply plan data of regions A, B, and C in the next week is obtained from the power dispatching center, including power generation plans, load forecasts, wind and solar power abandonment data, etc. in each region.

[0134] Step S33: using the regional power supply plan data to evaluate the regional power supply capacity of the shared energy storage regional data, and generating regional power supply data;

[0135] In an embodiment of the present invention, regional power supply planning data, such as the planned power generation of each region and the predicted power generation of renewable energy, are used to evaluate the regional power supply capacity of the shared energy storage area data. For example, the planned power generation and the predicted power generation of renewable energy in each region can be added to obtain the total power supply capacity of the region. Taking into account the volatility and uncertainty of renewable energy output, the regional power supply capacity can be estimated based on the confidence interval of the predicted power generation of renewable energy. Finally, regional power supply data is generated, such as data containing the power supply capacity of each region in the next 24 hours (such as maximum value, minimum value, average value).

[0136] Step S34: performing spatiotemporal coupled supply and demand balance analysis on regional power supply data by modifying load forecast curve data to generate regional spatiotemporal supply and demand balance data;

[0137] In an embodiment of the present invention, by correcting the load forecast curve data, such as the load forecast value for the next 24 hours, a spatiotemporal coupled supply and demand balance analysis is performed on the regional power supply data. For example, the load forecast value of each region can be compared with the power supply capacity of the region to determine whether the region is in a state of oversupply, undersupply, or supply and demand balance at each time step. For example, if the load forecast value of region A at a certain time step is 100MW and the power supply capacity is 120MW, then the region is in a state of oversupply at the time step. Finally, regional spatiotemporal supply and demand balance data is generated, such as a data set containing the supply and demand balance state (e.g., oversupply, undersupply, supply and demand balance) of each region at each time step in the next 24 hours.

[0138] Step S35: Calculate the supply-demand gap probability based on the regional spatiotemporal supply-demand balance data to generate regional supply-demand gap probability data;

[0139] In an embodiment of the present invention, the probability of supply-demand gap is calculated based on the regional spatiotemporal supply-demand balance data, such as the supply-demand balance state of each time step in each region in the next 24 hours. For example, the number of time steps in which each region is in a state of less supply than demand in the next 24 hours can be counted, and the number can be divided by the total number of time steps (e.g., 24) to obtain the probability of supply-demand gap in the region. For example, if region A has 4 time steps in a state of less supply than demand in the next 24 hours, the probability of supply-demand gap in region A is 4 / 24=1 / 6, and the regional supply-demand gap probability data is finally generated.

[0140] Step S36: Perform regional supply and demand gap risk assessment based on regional supply and demand gap probability data, and mark regional supply and demand gaps on regional spatiotemporal supply and demand balance data to generate regional supply and demand gap matrix data.

[0141] In an embodiment of the present invention, a regional supply and demand gap risk assessment is performed based on regional supply and demand gap probability data, such as the supply and demand gap probability of each region. For example, based on a pre-set risk threshold, an area where the supply and demand gap probability is higher than the threshold can be identified as a high-risk area. For example, if the risk threshold is set to 0.2, an area where the supply and demand gap probability is higher than 0.2 is identified as a high-risk area. Then, the regional spatiotemporal supply and demand balance data is annotated with regional supply and demand gaps, for example, the supply and demand balance state of a high-risk area in a time step where supply is less than demand is annotated as "gap", and the supply and demand balance state of other areas and time steps is annotated as "balance". Finally, regional supply and demand gap matrix data is generated, such as a matrix in which each row represents a region, each column represents a time step, and the elements in the matrix are "gap" or "balance", indicating whether there is a supply and demand gap in the region at that time step.

[0142] Preferably, step S31 includes the following steps:

[0143] Step S311: performing variational modal decomposition on the energy storage regional load characteristic data to generate regional load modal component data;

[0144] Step S312: using a preset time series prediction model to perform intelligent prediction of modal components on regional load modal component data to generate load modal component prediction data;

[0145] Step S313: performing regional component prediction superposition according to the load modal component prediction data, and dynamically reconstructing the load curve to generate regional load prediction curve data;

[0146] Step S314: performing single-region load confidence estimation according to the regional load forecast curve data to generate a single-region load forecast confidence interval;

[0147] Step S315: using the single-region load forecast confidence interval to perform load forecast error correlation analysis on the regional load forecast curve data to generate forecast error correlation data;

[0148] Step S316: Perform curve confidence correction on the regional load forecast curve data using the forecast error correlation data to generate corrected load forecast curve data.

[0149] In an embodiment of the present invention, the PyVMD ​​library in Python is used, the historical load curve data of region A is taken as input, the modal number of VMD decomposition is set, for example, 5, and VMD decomposition is performed to obtain 5 modal components of the load curve of region A, each modal component represents the information of different frequency components in the load curve, for example, the low-frequency component represents the trend change of the load, and the high-frequency component represents the fluctuation change of the load. Each modal component is stored as a separate file, such as a CSV file, to generate regional load modal component data. Using the LSTM model, the historical data of each modal component of region A is used as training data to train the LSTM model, and the trained model is used to predict the change trend of each modal component in the future (for example, the next week) to generate load modal component prediction data. According to the load modal component prediction data, regional component prediction superposition is performed. For example, the multiple IMF component prediction values ​​of each region are added together to obtain the total load prediction value of the region. Then, the load curve is dynamically reconstructed, for example, the total load prediction value of each time step is connected to form a complete load prediction curve, and finally the regional load prediction curve data is generated. The Bootstrap method can be used to resample the load forecast curve of each region, and the confidence interval of the load forecast value of each time step can be calculated. For example, the historical load data of each region can be resampled 1000 times, and after each resampling, a load forecast curve is generated using the method of steps S311-S313, and then the confidence interval of 1000 load forecast curves of each time step is calculated. Finally, the single-region load forecast confidence interval is generated. The correlation coefficient of the load forecast error between each two regions can be calculated. For example, the Pearson correlation coefficient can be used to calculate the correlation coefficient between the load forecast errors of region A and region B at each time step in the next 24 hours, and finally generate the forecast error correlation data. According to the correlation coefficient of the load forecast error between regions, the load forecast confidence interval of each region is adjusted. For example, if the load forecast errors of region A and region B are highly positively correlated, the load forecast confidence interval of region A can be expanded to reflect the impact of the load forecast error of region B on region A, and finally generate the modified load forecast curve data.

[0150] As an example of the present invention, refer to Figure 3 As shown, Figure 1 Detailed implementation steps of step S4 in the embodiment are shown in the flowchart. In this embodiment, step S4 includes:

[0151] Step S41: evaluating the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data and the grid node association topology data, and generating corrected inter-regional complementary potential data;

[0152] In an embodiment of the present invention, an evaluation of the potential for inter-regional power mutual assistance is performed based on regional supply and demand gap matrix data, such as whether there is a supply and demand gap in each region at each time step within the next 24 hours, and grid node association topology data, such as graph structure data containing node information, line information, line loss weight, node load weight, and node association strength. For example, the maximum power support that each region can obtain from other regions at each time step can be evaluated based on information such as the electrical distance between regions, line transmission capacity, and node betweenness centrality. Taking into account the transmission loss and stability limitations of the power grid, the initial evaluation results are corrected, for example, the maximum power support is reduced according to the line transmission capacity and the node betweenness centrality, and finally the corrected inter-regional complementary potential data is generated.

[0153] Step S42: constructing typical supply and demand scenarios for the modified inter-regional complementary potential data through regional power supply plan data to obtain multi-time scale supply and demand scenario data;

[0154] In an embodiment of the present invention, a typical supply and demand scenario is constructed for the corrected inter-regional complementary potential data by acquiring regional power supply plan data, such as the planned power generation of each region and the predicted power generation of renewable energy. For example, different supply and demand scenarios can be constructed based on the output forecasts of different power generation types (such as wind power, photovoltaic power, and thermal power) in the regional power supply plan data, as well as the load forecasts, such as scenarios with high wind power output and low load, scenarios with low wind power output and high load, etc. Each scenario corresponds to a set of regional supply and demand data and covers different time scales, such as days, weeks, months, etc. All supply and demand scenario data are organized into data sets to obtain multi-time scale supply and demand scenario data.

[0155] Step S43: constructing a power network simulation model using the grid node associated topology data;

[0156] In an embodiment of the present invention, a power network simulation model is constructed using grid node associated topological data, such as graph structure data including node information, line information, line loss weight, node load weight, and node association strength. For example, a power network simulation model can be constructed using power system simulation software such as Matpower and PSS / E, and the node information, line information, line parameters, etc. in the grid node associated topological data are imported into the simulation software.

[0157] Step S44: using the power network simulation model to perform shared power dispatch simulation on the multi-time scale supply and demand scenario data, and perform dispatch flow calculation to generate multi-scenario power dispatch simulation data;

[0158] In the embodiment of the present invention, in each supply and demand scenario, PowerWorld or Matpower software is used to perform power system flow calculation and optimal scheduling, for example, control strategies such as generator output and load distribution are set, the process of energy storage system participating in power grid scheduling is simulated, and the simulation results are recorded. The load data, power generation data and inter-regional power exchange scheme in each supply and demand scenario are input into the power network simulation model, and scheduling flow calculation is performed, for example, the voltage and current of each node and the flow distribution on the line are calculated using the Newton-Raphson method. Finally, multi-scenario power scheduling simulation data is generated.

[0159] Step S45: Perform inter-regional shared scheduling analysis based on the multi-scenario power scheduling simulation data, and perform power flow spatiotemporal graph processing to obtain spatiotemporal power flow scheduling graph data.

[0160] In an embodiment of the present invention, inter-regional shared scheduling analysis is performed based on multi-scenario power dispatch simulation data, such as data such as voltage, current, and line flow distribution of each node under each supply and demand scenario. For example, indicators such as the amount of power exchange between regions under each supply and demand scenario, line transmission loss, and node voltage stability can be analyzed. Then, the power flow spatiotemporal map processing is performed, such as displaying the flow distribution on each line under each supply and demand scenario in a graphical manner, such as using arrows to indicate the flow direction, and the thickness of the arrows to indicate the flow size, and finally obtaining the spatiotemporal power flow dispatch map data, such as how much power is transmitted from region A to region B, and in which time period the energy storage system is charged and discharged.

[0161] Preferably, step S41 includes the following steps:

[0162] Step S411: performing regional gap time series decomposition on the regional supply-demand gap matrix data to generate regional gap time series data;

[0163] Step S412: performing supply-demand fluctuation correlation analysis based on the regional gap time series data to generate a supply-demand gap fluctuation correlation coefficient;

[0164] Step S413: Calculate the regional supply and demand mutual information of the grid node associated topology data through the supply and demand gap fluctuation correlation coefficient, and evaluate the power complementarity potential to obtain the inter-regional supply and demand complementarity potential data;

[0165] Step S414: performing transmission capacity constraint analysis based on the grid node associated topology data to generate regional transmission capacity constraint data; performing distance attenuation factor processing based on the grid node associated topology data to generate inter-regional distance attenuation factors;

[0166] Step S415: using the inter-regional distance attenuation factor and the regional transmission capacity constraint data to correct the inter-regional supply-demand complementary potential value, and obtain corrected inter-regional complementary potential data.

[0167] In an embodiment of the present invention, regional supply and demand gap matrix data, such as whether there is a supply and demand gap in each time step in each region within the next 24 hours, is subjected to regional gap time series decomposition. The supply and demand gap time series data of each region can be decomposed into several modal components with different frequency characteristics using methods such as empirical mode decomposition (EMD) and variational mode decomposition (VMD). The Pearson correlation coefficient of the corresponding IMF component between each two regions can be calculated, or the cross-correlation function between the supply and demand gap time series data of the two regions can be calculated. For example, the Pearson correlation coefficient between the IMF components of the two regions can be calculated using the numpy library in Python, or the cross-correlation function between the supply and demand gap time series data of the two regions can be calculated using the scipy library. The supply and demand gap fluctuation correlation coefficient of region A and region B, as well as the grid connection strength between region A and region B can be calculated, and the supply and demand complementarity potential between region A and region B can be calculated using methods such as mutual information, for example, to evaluate whether the power surplus of region A can effectively make up for the power gap of region B. The supply and demand complementarity potential between all regions is organized into a matrix form to obtain inter-regional supply and demand complementarity potential data. Analyze the capacity limitations of transmission lines between different regions and generate regional transmission capacity constraint data. For example, the capacity of the transmission line from region A to region B is 100MW. According to the grid node association topology data, calculate the distance between different regions, and use a distance attenuation factor model, such as an exponential attenuation model, to calculate the inter-regional distance attenuation factor. For example, the distance between region A and region B is 100km, and the distance attenuation factor is 0.8. Multiply the supply and demand complementary potential between region A and region B by the inter-regional distance attenuation factor, and consider the capacity limitations of the transmission line from region A to region B to obtain the corrected inter-regional complementary potential. For example, the corrected complementary potential from region A to region B is 80MW. Arrange the corrected complementary potential between all regions in a matrix form to obtain the corrected inter-regional complementary potential data.

[0168] Preferably, step S5 comprises the following steps:

[0169] Step S51: performing regional energy storage capacity configuration on the modified inter-regional complementary potential data through spatiotemporal power flow dispatching graph data to generate regional energy storage capacity quota data;

[0170] Step S52: using the regional energy storage capacity quota data to perform cross-regional power transmission minimum loss path planning on the spatiotemporal power flow scheduling map data, and obtaining energy storage regional scheduling path data;

[0171] Step S53: Optimizing the state of energy storage nodes in a control cycle by using a preset model predictive control algorithm for the energy storage area scheduling path data and the regional energy storage capacity quota data to obtain a dynamic energy storage node control strategy;

[0172] Step S54: optimizing the capacity of the shared energy storage node according to the dynamic energy storage node control strategy to obtain energy storage capacity optimization configuration data;

[0173] Step S55: Perform an operation benefit evaluation based on the energy storage capacity optimization configuration data, and perform rolling adjustment on the capacity configuration to generate dynamic energy storage capacity configuration data.

[0174] In an embodiment of the present invention, according to the spatiotemporal power flow scheduling map data, it is analyzed that region A needs the energy storage system to discharge during the peak period and the energy storage system needs to charge during the valley period, and combined with the correction of the complementary potential data between regions, for example, region A can provide energy storage services to region C, and determine the energy storage capacity quota of region A, for example, the energy storage capacity quota of region A is 100MWh. Use the shortest path algorithm such as the Dijkstra algorithm or the Floyd algorithm to calculate the minimum loss path for achieving inter-regional power mutual assistance under the condition of satisfying the energy storage capacity constraint. For example, if region A needs to transmit 100MWh of electricity to region B, the Dijkstra algorithm can be used to calculate the minimum loss path from region A to region B, and the actual power that can be transmitted can be determined based on the line loss and energy storage capacity constraints on the path. Set the control cycle to 1 hour, and use the MPC algorithm to optimize the charging and discharging strategy of the energy storage system in region A. The objective function can be set to minimize the operating cost of the energy storage system or maximize the benefits of the energy storage system. The constraints may include the charging and discharging power limit of the energy storage system, the energy storage capacity limit, and the grid voltage stability limit. According to the optimization results, the charge and discharge power and state of each energy storage node in each time period are generated to obtain the dynamic energy storage node control strategy. According to the charge and discharge state of each energy storage node at different time steps and the line loss on the power transmission path, the capacity of each energy storage node is optimized to minimize the total cost of the energy storage system or maximize the benefit of the energy storage system. The charge and discharge power and state of the energy storage system in area A in different time periods are analyzed, and the capacity of the energy storage node is optimized according to the charge and discharge frequency, charge and discharge depth and other indicators of the energy storage node, for example, the capacity of a certain node is increased and the capacity of another node is reduced to meet the energy storage demand of area A and reduce the total cost of the energy storage system. According to the information such as the charge and discharge times, charge and discharge amount, power transmission loss and other information of each energy storage node, the operating cost of the energy storage system, the reduced carbon emissions, the improved power system reliability and other indicators are calculated. Then, the capacity configuration is adjusted in a rolling manner, for example, the capacity of the energy storage node is adjusted according to the results of the operation benefit evaluation, for example, the capacity of the energy storage node with higher benefits is increased and the capacity of the energy storage node with lower benefits is reduced. According to the evaluation results, the energy storage capacity configuration is adjusted on a rolling basis. For example, the capacity of the energy storage node is adjusted according to the operation of the energy storage system to adapt to the changes in the grid load and the fluctuations in the output of renewable energy. Finally, dynamic energy storage capacity configuration data is generated, which contains information such as the capacity and charging and discharging strategy of each energy storage node, and is updated over time.

[0175] Preferably, step S51 includes the following steps:

[0176] Step S511: identifying target energy storage nodes according to the spatiotemporal power flow dispatch graph data to obtain target energy storage optimization node data;

[0177] Step S512: collecting energy storage capacity constraint parameters according to the target energy storage optimization node data, and setting energy storage capacity constraint conditions to obtain energy storage capacity constraint condition data;

[0178] Step S513: using the energy storage capacity constraint condition data to perform preliminary configuration of regional energy storage capacity on the modified inter-regional complementary potential data, and generating preliminary regional energy storage capacity configuration data;

[0179] Step S514: performing shared energy storage capacity allocation processing on the modified inter-regional complementary potential data through the spatiotemporal power flow dispatching graph data to obtain shared energy storage capacity allocation problem data;

[0180] Step S515: Markov decision processing is performed on the shared energy storage capacity allocation problem data, and dynamic capacity quota optimization is performed on the preliminary regional energy storage capacity configuration data to generate regional energy storage capacity quota data.

[0181] In an embodiment of the present invention, according to the power supply and demand balance state, power flow direction, node importance and other information of each node at different time steps, nodes that need to be configured with energy storage are identified, such as nodes with large power supply and demand gaps, nodes with frequent power flow changes, nodes with high betweenness centrality, etc. In the spatiotemporal power flow scheduling graph data, which nodes play an important role in the cross-regional power transmission process, for example, which nodes have large power flow changes, which nodes are connected to multiple regions, and which nodes are close to renewable energy power plants or load centers. These key nodes are extracted, and the geographical location, voltage level and other information of the nodes are recorded to obtain the target energy storage optimization node data. According to the geographical location, environmental conditions and other information of the target energy storage optimization node, the technical parameters of the energy storage system are queried, such as the type of energy storage system, rated power, energy storage capacity, charging and discharging efficiency, etc. In combination with the operation requirements of the power grid, such as the voltage stability and frequency stability of the power grid, the energy storage capacity constraints are set, such as the energy storage capacity cannot exceed a certain upper limit, and the charging and discharging power cannot exceed a certain upper limit. According to the power supply and demand gap of each region at different time steps, as well as the potential for inter-regional power mutual assistance, the energy storage capacity required for each region is preliminarily determined on the premise of meeting the energy storage capacity constraint. Optimization methods such as linear programming can be used to solve the preliminary configuration plan of regional energy storage capacity that meets the constraint conditions, and finally generate preliminary regional energy storage capacity configuration data. Analyze the power exchange between different regions in the spatiotemporal power flow scheduling map data, as well as the process of energy storage system participating in power grid scheduling, and combine the correction of inter-regional complementary potential data to transform the energy storage capacity allocation problem into an optimization problem. For example, how to allocate the 100MWh energy storage capacity of region A to maximize the inter-regional power complementary benefits and meet the energy storage capacity constraint conditions. Model the shared energy storage capacity allocation problem as a Markov decision process. The state space can be the energy storage capacity state of different regions, the action space can be the energy storage capacity allocation strategy between different regions, and the reward function can be the operating benefit of the energy storage system, such as reducing the system operating cost and increasing the new energy consumption rate. Reinforcement learning algorithms, such as the Q-learning algorithm, are used to solve the Markov decision process and obtain the optimal energy storage capacity allocation strategy. The preliminary regional energy storage capacity configuration data is optimized based on the strategy to obtain the final energy storage capacity quota for each region. For example, the energy storage capacity quota for region A is adjusted to 80MWh.

[0182] The beneficial effect of this application is that, through fine regional division and load characteristic analysis, the power demand characteristics of each region are accurately grasped, laying a data foundation for cross-regional collaboration. Next, the constructed cross-regional power grid topology and node correlation strength analysis provide an optimized path from a network perspective for the efficient flow of power resources, reducing transmission losses. Through supply and demand balance analysis and gap assessment, as well as the precise quantification of power mutual assistance potential, the supply and demand mismatch problems between different regions are effectively identified and coordinated, promoting the optimal allocation of resources. Furthermore, by utilizing the spatiotemporal power flow scheduling map and dynamic energy storage capacity configuration, the flexible scheduling and optimized layout of energy storage resources are realized, ensuring the real-time matching of power demand and supply, and greatly improving the overall response speed and adaptability of the system. The optimized configuration of cross-regional energy storage capacity is achieved, the new energy absorption capacity is improved, the system operating cost is reduced, and the stability and reliability of power system operation are improved.

[0183] 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.

[0184] 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 optimizing energy storage capacity configuration based on a shared energy storage mode, characterized in that: The following steps are involved: Step S1: using the power dispatching center to divide the shared energy storage area and obtain the shared energy storage area data; Extract the target area load characteristic index from the shared energy storage area data to generate regional load characteristic data; process the weighted load characteristic value according to the regional load characteristic data to generate energy storage area load characteristic data; Step S2: constructing a power grid topology structure for the shared energy storage area data to generate cross-regional power grid topology structure data; performing node association strength processing on the cross-regional power grid topology structure data using the energy storage area load characteristic data to generate power grid node association topology data; Step S3: Acquire regional power supply plan data; Based on the energy storage regional load characteristic data, use the regional power supply plan data to perform spatiotemporal coupling supply and demand balance analysis on the grid node associated topology data to generate regional supply and demand gap matrix data; Step S4: Evaluate the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data to generate corrected inter-regional complementary potential data; perform inter-regional shared scheduling analysis on the corrected inter-regional complementary potential data using the regional power supply plan data to obtain spatiotemporal power flow scheduling map data; Step S5: configuring regional energy storage capacity according to the spatiotemporal power flow dispatching graph data, and generating regional energy storage capacity quota data; The shared energy storage node capacity is optimized according to the regional energy storage capacity quota data, and dynamic energy storage capacity configuration data is generated.

2. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: using the power dispatching center to divide the shared energy storage area and obtain shared energy storage area data; Step S12: collecting historical power load data according to the shared energy storage area data to obtain original historical power load data; Step S13: performing outlier detection on the original historical power load data and filling in missing values ​​to generate regional historical power load data; Step S14: extracting the load characteristic index of the target area according to the historical power load data of the area, and generating regional load characteristic data; Step S15: Mapping regional characteristic indicators to the shared energy storage regional data through regional load characteristic data to generate regional multi-dimensional characteristic matrix data; Step S16: performing Z-sore standardization on the regional multi-dimensional feature matrix data to obtain a standardized regional feature matrix; performing feature index weight evaluation based on the standardized regional feature matrix to obtain feature index weight vector data; Step S17: Use the characteristic index weight vector data to perform weighted load characteristic value processing on the standardized regional characteristic matrix to generate energy storage regional load characteristic data.

3. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 2 is characterized in that: Step S15 includes the following steps: Step S151: Perform dynamic time warping on the regional load characteristic data, and compare the characteristics between regions to generate characteristic difference data between regions; Step S152: Calculate the time series feature similarity based on the feature difference data between regions to generate regional load similarity data; Step S153: performing load pattern clustering analysis based on regional load similarity data, and performing regional clustering division to generate regional load cluster data; Step S154: extracting cluster typical load curves from regional load cluster data to obtain cluster typical load curve data; Step S155: performing regional mapping on the shared energy storage regional data through the cluster typical load curve data to generate regional-cluster characteristic mapping data; Step S156: performing multi-dimensional feature vector processing according to the region-cluster feature mapping data to obtain multi-dimensional feature vector data; Step S157: constructing a regional characteristic matrix for the regional load characteristic data through the multi-dimensional characteristic vector data to generate regional multi-dimensional characteristic matrix data.

4. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: using the power dispatching center to construct the grid topology structure for the shared energy storage area data, and initialize the line loss weight to generate cross-regional grid topology structure data; Step S22: weighting node power loads on cross-regional power grid topology data using energy storage area load characteristic data to generate load-weighted power grid topology data; Step S23: performing node betweenness centrality statistics according to the load-weighted power grid topology data to obtain node betweenness centrality data; Step S24: Calculate the inter-regional electrical distance according to the load-weighted power grid topology data to generate inter-regional electrical distance data; Step S25: Based on the inter-regional electrical distance data, the load-weighted power grid topology data is processed for node association strength through the node betweenness centrality data to generate power grid node association topology data.

5. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: performing a multi-step forecast of regional time series load according to the energy storage regional load characteristic data to obtain corrected load forecast curve data; Step S32: Acquire regional power supply plan data, including power production plans for each region and renewable energy output estimation data; Step S33: using the regional power supply plan data to evaluate the regional power supply capacity of the shared energy storage regional data, and generating regional power supply data; Step S34: performing spatiotemporal coupled supply and demand balance analysis on regional power supply data by modifying load forecast curve data to generate regional spatiotemporal supply and demand balance data; Step S35: Calculate the supply-demand gap probability based on the regional spatiotemporal supply-demand balance data to generate regional supply-demand gap probability data; Step S36: Perform regional supply and demand gap risk assessment based on regional supply and demand gap probability data, and mark regional supply and demand gaps on regional spatiotemporal supply and demand balance data to generate regional supply and demand gap matrix data.

6. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 5 is characterized in that: Step S31 includes the following steps: Step S311: performing variational modal decomposition on the energy storage regional load characteristic data to generate regional load modal component data; Step S312: using a preset time series prediction model to perform intelligent prediction of modal components on regional load modal component data to generate load modal component prediction data; Step S313: performing regional component prediction superposition according to the load modal component prediction data, and dynamically reconstructing the load curve to generate regional load prediction curve data; Step S314: performing single-region load confidence estimation according to the regional load forecast curve data to generate a single-region load forecast confidence interval; Step S315: using the single-region load forecast confidence interval to perform load forecast error correlation analysis on the regional load forecast curve data to generate forecast error correlation data; Step S316: Perform curve confidence correction on the regional load forecast curve data using the forecast error correlation data to generate corrected load forecast curve data.

7. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: evaluating the inter-regional power mutual assistance potential according to the regional supply-demand gap matrix data and the grid node association topology data, and generating corrected inter-regional complementary potential data; Step S42: constructing typical supply and demand scenarios for the modified inter-regional complementary potential data through regional power supply plan data to obtain multi-time scale supply and demand scenario data; Step S43: constructing a power network simulation model using the grid node associated topology data; Step S44: using the power network simulation model to perform shared power dispatch simulation on the multi-time scale supply and demand scenario data, and perform dispatch flow calculation to generate multi-scenario power dispatch simulation data; Step S45: Perform inter-regional shared scheduling analysis based on the multi-scenario power scheduling simulation data, and perform power flow spatiotemporal graph processing to obtain spatiotemporal power flow scheduling graph data.

8. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 7 is characterized in that: Step S41 includes the following steps: Step S411: performing regional gap time series decomposition on the regional supply-demand gap matrix data to generate regional gap time series data; Step S412: performing supply-demand fluctuation correlation analysis based on the regional gap time series data to generate a supply-demand gap fluctuation correlation coefficient; Step S413: Calculate the regional supply and demand mutual information of the grid node associated topology data through the supply and demand gap fluctuation correlation coefficient, and evaluate the power complementarity potential to obtain the inter-regional supply and demand complementarity potential data; Step S414: performing transmission capacity constraint analysis based on the grid node associated topology data to generate regional transmission capacity constraint data; performing distance attenuation factor processing based on the grid node associated topology data to generate inter-regional distance attenuation factors; Step S415: using the inter-regional distance attenuation factor and the regional transmission capacity constraint data to correct the inter-regional supply-demand complementary potential value, and obtain corrected inter-regional complementary potential data.

9. The method for optimizing energy storage capacity configuration based on a shared energy storage mode according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: performing regional energy storage capacity configuration on the modified inter-regional complementary potential data through spatiotemporal power flow dispatching graph data to generate regional energy storage capacity quota data; Step S52: using the regional energy storage capacity quota data to perform cross-regional power transmission minimum loss path planning on the spatiotemporal power flow scheduling map data, and obtaining energy storage regional scheduling path data; Step S53: Optimizing the state of energy storage nodes in a control cycle by using a preset model predictive control algorithm for the energy storage area scheduling path data and the regional energy storage capacity quota data to obtain a dynamic energy storage node control strategy; Step S54: optimizing the capacity of the shared energy storage node according to the dynamic energy storage node control strategy to obtain energy storage capacity optimization configuration data; Step S55: Perform an operation benefit evaluation based on the energy storage capacity optimization configuration data, and perform rolling adjustment on the capacity configuration to generate dynamic energy storage capacity configuration data.

10. The energy storage capacity optimization configuration method based on the shared energy storage mode according to claim 1 is characterized in that: Step S51 includes the following steps: Step S511: identifying target energy storage nodes according to the spatiotemporal power flow dispatch graph data to obtain target energy storage optimization node data; Step S512: collecting energy storage capacity constraint parameters according to the target energy storage optimization node data, and setting energy storage capacity constraint conditions to obtain energy storage capacity constraint condition data; Step S513: using the energy storage capacity constraint condition data to perform preliminary configuration of regional energy storage capacity on the modified inter-regional complementary potential data, and generating preliminary regional energy storage capacity configuration data; Step S514: performing shared energy storage capacity allocation processing on the modified inter-regional complementary potential data through the spatiotemporal power flow dispatching graph data to obtain shared energy storage capacity allocation problem data; Step S515: Markov decision processing is performed on the shared energy storage capacity allocation problem data, and dynamic capacity quota optimization is performed on the preliminary regional energy storage capacity configuration data to generate regional energy storage capacity quota data.

Images