A source network and load flexible resource integrated hybrid energy storage optimization method and system

Abstract

The application discloses a kind of source network load flexibility resource integration's hybrid energy storage optimization method and system, it is related to distributed source network load storage planning field, for solving the problem that centralized energy storage planning ignores terrain hydrology and distribution network difference causes site distortion and scale expansion, is by first in terrain and water body information screen out with upper and lower reservoir conditions pumped storage candidate point, again mapping distribution and transmission network and according to energy path and tidal flow sensitivity cutting access position, so that energy storage is located close to load concentration area rather than limited to a few nodes in transmission level, from source to inhibit long-distance transmission loss and invalid expansion demand;On this basis, pumped storage undertakes slow change power movement, electrochemical energy storage undertakes fast change peak valley regulation, and the apportioned result is embedded in power generation, line opening, demand response and other multi-side joint model, so that spatial constructability and operation flexibility synergize in the same optimization framework;Therefore, avoid the expansion of energy storage capacity caused by ignoring flexible resources.

Description

Technical Field

[0001] This invention relates to the field of distributed source-grid-load-storage planning, and more specifically, to a hybrid energy storage optimization method and system that integrates source-grid-load flexibility resources. Background Technology

[0002] In distributed energy scenarios, renewable energy sources such as wind and solar power are clustered, and the load side includes both industrial and residential / commercial electricity consumption, resulting in significant fluctuations in network power flow over time and space. To absorb these fluctuations without expanding power lines, energy storage must be deployed locally according to terrain, hydrology, and distribution network topology: pumped storage relies on elevation differences and water bodies, while electrochemical energy storage should be located near load centers. However, the existing patent "A Hybrid Energy Storage Optimization Configuration Method Integrating Source-Grid-Load Flexibility Resources" (Publication No.: CN115579917A) uses transmission-level discrete nodes as all candidate locations, limiting the deployment of pumped storage and batteries to a limited number of transmission nodes.

[0003] This centralized assumption leads to a series of technical problems: First, the model ignores topographic and hydrological constraints. If pumped storage is located far from elevation differences and water sources, it will inevitably increase earthwork and water diversion projects, increasing construction difficulty. Second, if the batteries are located in substations far from the load, charging and discharging require long-distance power transmission and distribution, leading to amplified line losses and increased risk of local congestion. Third, with limited node selection, the algorithm often compensates for insufficient site selection by increasing capacity, resulting in higher investment. As existing patent documents suggest, without fully exploring flexibility resources, energy storage capacity is easily inflated. The problem arises because existing models treat geographical feasibility, distribution network access capacity, and power flow sensitivity as external prerequisites, without integrating them with capacity planning. The problem manifests in the lack of spatial screening during the site selection stage, and in the operation stage, it is exposed as water resource transportation costs, line losses, and voltage exceeding limits. The ultimate consequence is decreased system efficiency and increased risk of local equipment overload, limiting the realization of the potential for source-grid-load-storage synergy.

[0004] To address the aforementioned problems, a technical solution is provided. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of existing technologies, embodiments of the present invention provide a hybrid energy storage optimization method and system that integrates source-grid-load flexibility resources. This method first screens out pumped storage candidate sites with suitable upper and lower reservoir conditions from topographic and water body information. Then, it maps the distribution and transmission networks and tailors the access locations according to energy paths and power flow sensitivity. This ensures that energy storage is located close to concentrated load areas rather than limited to a few nodes at the transmission level, suppressing long-distance transmission losses and ineffective expansion demands from the source. Based on this, pumped storage is used to handle slow-changing power relocation, while electrochemical energy storage is used for fast-changing peak-valley regulation. The allocation results are embedded into a multi-faceted joint model encompassing power generation, line interruption, and demand response, allowing spatial feasibility and operational flexibility to work synergistically within the same optimization framework to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A hybrid energy storage optimization method integrating source-grid-load flexibility resources includes the following steps:

[0008] S1: Based on topographic elevation grid and water body distribution data, candidate point pairs for pumped storage upper and lower reservoirs and a list of topographic constraints are generated through elevation threshold screening and water body connectivity analysis.

[0009] S2: Map the candidate points of pumped storage upper and lower reservoirs and the candidate locations of electrochemical energy storage to the power distribution and transmission network. Generate a geographic-electrical mapping table based on line impedance, substation voltage level and load level, and mark the access feasibility indicators for each energy storage candidate location.

[0010] S3: Combine the node load time series, use the geographic-electrical mapping table to calculate the energy path level and power flow sensitivity, eliminate energy path level candidates that exceed the corresponding standard or power flow sensitivity that is lower than the corresponding standard, and generate an initial screening set with priority for local absorption.

[0011] S4: In the initial screening, the intraday regulation demand and multi-day regulation demand are evaluated in parallel. The multi-day regulation demand is assigned to pumped storage as a long-cycle power transfer task. The intraday regulation demand is assigned to electrochemical energy storage as a short-cycle peak-valley regulation task. A hybrid energy storage allocation table is generated to record the allocation results.

[0012] S5: Call the hybrid energy storage allocation table and the power generation model, line interruption model and load response model to jointly optimize the location and capacity of energy storage, retain the list of terrain constraints and power flow sensitivity, and output the energy storage layout scheme and the graded charging and discharging sequence.

[0013] In a preferred embodiment, step S1 includes the following:

[0014] Construct a topographic elevation grid, calculate the absolute difference in elevation between the center points of adjacent grid cells to form an elevation grid matrix; traverse the elevation grid matrix, filter grid cell pairs with elevation differences greater than or equal to a preset minimum elevation difference threshold, and generate a preliminary candidate point pair set; use hydrological data to generate a water body distribution layer, mark the water body presence status of each grid cell, check the water body connectivity of grid cell pairs in the preliminary candidate point pair set, retain at least one grid cell containing water and no elevation barriers between the two grid cells, form a candidate point pair set for the upper and lower reservoirs of pumped storage, and generate a topographic constraint list.

[0015] In a preferred embodiment, step S2 includes the following:

[0016] Based on the geographic coordinates of grid cells in the candidate point pair set of pumped storage upper and lower reservoirs and the grid index of load centers in the candidate location set of electrochemical energy storage, the Euclidean distance between the geographic coordinates of grid cells and the geographic coordinates of distribution-transmission network nodes is calculated. The network node with the smallest distance is selected as the mapping node, and a geographic-electrical mapping table is generated to record the grid index and the mapping node number.

[0017] In a preferred embodiment, step S2 further includes the following:

[0018] The equivalent impedance of each line segment is accumulated along the network path from the mapping node to the main load bus to obtain the total impedance value. The total impedance value and the substation voltage level are recorded. The load level label is recorded according to the type of load bus connected to the mapping node. The line impedance, substation voltage level and load level are analyzed to obtain access feasibility indicators and update the geographic-electrical mapping table.

[0019] In a preferred embodiment, step S3 includes the following:

[0020] The node load time sequence is integrated into the geographic-electrical mapping table, and the load power sequence of each mapped node is recorded. The equivalent resistance segment and voltage level crossing number are recorded segment by segment along the power transmission path from the mapped node to the main load bus. The number of high loss segments and voltage steps on the path are accumulated to obtain the energy path level.

[0021] In a preferred embodiment, step S3 further includes the following:

[0022] Under the DC power flow framework, an active power injection disturbance is applied to each mapping node. The power flow sensitivity is obtained by dividing the power flow increment of the bottleneck line by the size of the injection disturbance. Energy storage candidate locations with energy path levels exceeding the upper threshold or power flow sensitivity below the lower threshold are eliminated. The remaining locations are sorted by energy path level from smallest to largest to generate an initial screening set.

[0023] In a preferred embodiment, step S4 includes the following:

[0024] Intraday regulation demand is obtained by using the node load time series of each mapped node in the initial screening set, and multi-day regulation demand is obtained by using the node load time series. The multi-day regulation demand is assigned to the candidate points of pumped storage upper and lower reservoirs as long-cycle power transfer tasks. Intraday regulation demand is assigned to candidate locations of electrochemical energy storage as short-cycle peak-valley regulation tasks. Grouped by energy storage type, the total value of long-cycle tasks is calculated as the sum of multi-day regulation demand for pumped storage locations, and the total value of short-cycle tasks is summed as the intraday regulation demand for electrochemical energy storage locations. A mixed energy storage allocation table is generated.

[0025] In a preferred embodiment, step S4 further includes the following:

[0026] The system iterates through the hybrid energy storage allocation table and extracts the elevation difference, water connectivity, and tidal current sensitivity from the terrain constraint list as weights. It minimizes the sum of the construction cost of all locations multiplied by the location variable, plus the sum of the absolute difference between the tidal current sensitivity and the ideal sensitivity threshold of all locations multiplied by the adjustment task value multiplied by the location variable. The constraints include that the total number of locations does not exceed the budget limit, and that the elevation difference meets the preset threshold and the water connectivity meets the requirements.

[0027] In a preferred embodiment, step S4 further includes the following:

[0028] Minimize the sum of unit capacity cost multiplied by capacity variables at all locations, plus the sum of total regulation demand at all time points minus the sum of charging and discharging energy at all locations; constraints include that capacity variables do not exceed the upper limit of access feasibility indicators, and that power flow under the line interruption model does not exceed the branch capacity.

[0029] A hybrid energy storage optimization system integrating source-grid-load flexibility resources includes:

[0030] Terrain filtering module: Based on terrain elevation grid and water body distribution data, it generates candidate point pairs for pumped storage upper and lower reservoirs and a list of terrain constraints through elevation threshold filtering and water body connectivity analysis;

[0031] Network mapping module: Maps candidate points of pumped storage upper and lower reservoirs and candidate locations of electrochemical energy storage to the power distribution and transmission network. It generates a geographic-electrical mapping table based on line impedance, substation voltage level and load level, and marks access feasibility indicators for each energy storage candidate location.

[0032] Location screening module: Combining node load time series, using geographic-electrical mapping table to calculate energy path level and power flow sensitivity, eliminating energy path level exceeding the corresponding standard or power flow sensitivity below the corresponding standard to generate an initial screening set with priority for local absorption.

[0033] Task assignment module: In the initial screening, the intraday regulation demand and multi-day regulation demand are evaluated in parallel. The multi-day regulation demand is assigned to pumped storage as a long-cycle power transfer task, and the intraday regulation demand is assigned to electrochemical energy storage as a short-cycle peak-valley regulation task. A hybrid energy storage allocation table is generated to record the allocation results.

[0034] Joint optimization module: It calls the hybrid energy storage allocation table and the power generation model, line interruption model and load response model to jointly optimize the location and capacity of energy storage, retain the list of terrain constraints and power flow sensitivity to output the energy storage layout scheme and the graded charging and discharging sequence.

[0035] The technical effects and advantages of the hybrid energy storage optimization method and system integrating source-grid-load flexibility resources of this invention are as follows:

[0036] This invention first screens out candidate pumped storage sites with suitable upper and lower reservoir conditions from topographic and water body information, then maps the power distribution and transmission networks and tailors the access locations according to energy paths and power flow sensitivity. This ensures that energy storage is located close to concentrated load areas rather than limited to a few nodes at the transmission level, suppressing long-distance transmission losses and ineffective expansion demands from the source. Based on this, pumped storage is used to handle slow-changing power relocation, while electrochemical energy storage is used for fast-changing peak-valley regulation. The allocation results are embedded into a multi-sided joint model encompassing power generation, line interruption, and demand response, allowing spatial feasibility and operational flexibility to work synergistically within the same optimization framework. The resulting source-grid-load-storage coordinated planning avoids energy storage capacity expansion caused by neglecting flexible resources, while strengthening the absorption channels for renewable energy output and alleviating investment and operational pressures. The overall operational pattern is more compatible and has greater dispatch flexibility than single-sided or centralized planning. Attached Figure Description

[0037] Figure 1 This is a schematic flowchart of a hybrid energy storage optimization method that integrates source, grid, and load flexibility resources according to the present invention.

[0038] Figure 2 This is a schematic diagram of the structure of a hybrid energy storage optimization system that integrates source, grid, and load flexibility resources according to the present invention. Detailed Implementation

[0039] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0040] Example 1: Figure 1 This invention presents a hybrid energy storage optimization method integrating source-grid-load flexibility resources, comprising:

[0041] S1: Based on topographic elevation grid and water body distribution data, candidate point pairs for pumped storage upper and lower reservoirs and a list of topographic constraints are generated through elevation threshold screening and water body connectivity analysis.

[0042] S2: Map the candidate points of pumped storage upper and lower reservoirs and the candidate locations of electrochemical energy storage to the power distribution and transmission network. Generate a geographic-electrical mapping table based on line impedance, substation voltage level and load level, and mark the access feasibility indicators for each energy storage candidate location.

[0043] S3: Combine the node load time series, use the geographic-electrical mapping table to calculate the energy path level and power flow sensitivity, eliminate energy path level candidates that exceed the corresponding standard or power flow sensitivity that is lower than the corresponding standard, and generate an initial screening set with priority for local absorption.

[0044] S4: In the initial screening, the intraday regulation demand and multi-day regulation demand are evaluated in parallel. The multi-day regulation demand is assigned to pumped storage as a long-cycle power transfer task. The intraday regulation demand is assigned to electrochemical energy storage as a short-cycle peak-valley regulation task. A hybrid energy storage allocation table is generated to record the allocation results.

[0045] S5: Call the hybrid energy storage allocation table and the power generation model, line interruption model and load response model to jointly optimize the location and capacity of energy storage, retain the list of terrain constraints and power flow sensitivity, and output the energy storage layout scheme and the graded charging and discharging sequence.

[0046] In distributed energy scenarios, renewable energy sources such as wind and solar power exhibit clustered distribution, with loads encompassing industrial, residential, and commercial electricity consumption, leading to significant fluctuations in power grid flow over time and space. To effectively absorb fluctuating power without expanding transmission lines, energy storage facilities need to be deployed locally based on topography, hydrological conditions, and distribution network topology. Pumped storage, as a large-scale energy storage method reliant on topographic elevation differences and water bodies, requires site selection to meet specific elevation difference and water connectivity requirements to ensure construction feasibility and operational efficiency. Existing technologies often limit energy storage site selection to transmission-level nodes, ignoring topographic and hydrological constraints, resulting in increased construction costs and decreased operational efficiency. This invention optimizes energy storage site selection and capacity in stages, emphasizing the synergistic effect of spatial feasibility and operational flexibility. Step S1 focuses on screening candidate sites for upper and lower pumped storage reservoirs. Through analysis of topographic elevation differences and water body distribution, candidate site pairs that meet construction conditions are scientifically determined, providing a geographical constraint basis for subsequent distribution-transmission network mapping.

[0047] Step S1 utilizes topographic elevation data and water distribution information to generate candidate pairs of upper and lower reservoirs for pumped storage and a list of topographic constraints through elevation threshold filtering and water connectivity analysis. This process ensures that the site selection conforms to natural geographical conditions and reduces earthwork and water diversion costs.

[0048] S1.1 Construction of terrain elevation difference grid.

[0049] To address the dependence of pumped storage on topographic elevation differences, it is necessary to first acquire topographic data of the target area to quantify the elevation distribution. Using digital elevation model (DEM) data, the target area is divided into regular grid cells. The elevation of the center point of each grid cell is recorded, denoted as the cell elevation. The elevation difference between adjacent grid cells is calculated as follows: for any two adjacent grid cells, the elevation of their center points is obtained, and the absolute difference between the two elevations is calculated to obtain the elevation difference between the adjacent grid cell pairs. The elevation differences of all adjacent grid cell pairs form an elevation difference grid matrix, where each element represents the elevation difference between a pair of adjacent grid cells. Due to its high accuracy and wide applicability, DEM data can accurately reflect topographic relief characteristics. The generated elevation difference grid matrix provides a unified geographic data framework for subsequent screening, ensuring the systematic nature and consistency of elevation difference calculations.

[0050] S1.2 Height difference threshold screening.

[0051] The power generation efficiency of pumped-storage hydroelectric power stations is directly related to the water head. Therefore, it is necessary to select grid cell pairs that meet the minimum elevation difference requirement to ensure potential energy conversion efficiency. A minimum elevation difference threshold is set, determined based on the technical and economic requirements of the pumped-storage power station. The elevation difference grid matrix is ​​traversed, and the elevation difference value of each pair of adjacent grid cells is checked one by one. Grid cell pairs with an elevation difference value greater than or equal to the minimum elevation difference threshold are retained, forming a preliminary set of candidate point pairs. Each candidate point pair records the indices of the two grid cells and their corresponding elevation difference values. Setting a minimum elevation difference threshold can effectively filter point pairs that do not meet the potential energy conversion requirements, reducing invalid candidates in subsequent analysis. The selected preliminary set of candidate point pairs provides qualified geographical locations for subsequent hydrological analysis, narrowing the candidate range and improving processing efficiency.

[0052] S1.3 Analysis of water body distribution and connectivity.

[0053] Pumped storage power stations rely on water bodies for water circulation; therefore, candidate point pairs need to be further screened based on hydrological data to ensure that the upper and lower reservoirs have water support and connectivity. Hydrological data (such as remote sensing imagery or river system distribution maps) is used to identify the water body distribution within the target area, generating a water body distribution layer. The presence status of water bodies is marked for each grid cell, defined as follows: if the grid cell contains lakes, rivers, or other water bodies, it is marked as having water; if there is no water body, it is marked as not having water. For each pair of grid cells in the initial candidate point pair set, their water connectivity is checked, specifically requiring that at least one grid cell in the point pair contains water, and there is a reachable water flow path between the two grid cells. Connectivity analysis is performed in two steps:

[0054] Path search: Starting from one grid cell in a point pair and ending at the other, the A* path search algorithm is used to find the optimal terrain path between the two points. The elevations of the center points of all grid cells along the path are recorded to form an elevation sequence. Due to its efficiency and heuristic search characteristics, the A* algorithm can quickly find feasible paths in complex terrain.

[0055] Obstacle detection: Examine the elevation sequence along the path to determine if any elevation obstacles exist. An elevation obstacle is defined as a grid cell on the path whose elevation exceeds the maximum elevation of the start and end points plus a preset obstacle elevation difference threshold, in meters. If there are no elevation obstacles on the path, the two grid cells are considered to satisfy water connectivity.

[0056] Grid cell pairs that simultaneously meet both elevation difference thresholds and water connectivity are selected to form a set of candidate sites for the upper and lower reservoirs of the pumped storage project. The application of hydrological data and connectivity analysis ensures that the candidate sites have practical water flow accessibility, reducing the complexity of the water diversion project. The selected set of candidate sites provides geographical locations that meet both topographic and hydrological constraints for subsequent steps, ensuring the engineering feasibility of the site selection.

[0057] S1.4 Terrain constraint list generation.

[0058] To facilitate subsequent distribution-transmission network mapping, the screening results need to be integrated into a structured topographic constraint list. For each pair of grid cells in the candidate point pair set for pumped storage upper and lower reservoirs, the following information is recorded: the indices of the two grid cells, indicating the locations of the upper and lower reservoirs; the elevation difference between the two grid cells, reflecting the potential energy conversion potential; the water presence status of the two grid cells, indicating whether water is present; and a connectivity flag, defined as follows: if the point pair satisfies water connectivity, it is marked as connected; otherwise, it is marked as disconnected. The topographic constraint list is stored in tabular form, with each row containing the index, elevation difference, water presence status, and connectivity flag of a point pair. The topographic constraint list provides directly usable input for the geo-electrical mapping in step S2, maintaining the continuity of the site selection process.

[0059] Step S1 completes the screening of candidate point pairs for pumped storage upper and lower reservoirs by constructing a topographic elevation difference grid, filtering elevation difference thresholds, analyzing water body distribution and connectivity, and generating a topographic constraint list. Starting from the digital elevation model and hydrological data, grid cell pairs that do not meet the elevation difference or hydrological conditions are filtered layer by layer, ultimately forming a set of candidate point pairs for pumped storage upper and lower reservoirs and a topographic constraint list. The output includes the grid index, elevation difference value, water body presence status, and connectivity indicator of the candidate point pairs, providing accurate geographic constraint information for the power distribution-transmission network mapping in step S2.

[0060] Step S1, through topographic elevation grid analysis and water body distribution analysis, has screened candidate pairs of pumped storage upper and lower reservoirs that meet the requirements of elevation difference and water connectivity, and generated a topographic constraint list, providing accurate geographical location information. However, geographical constraints alone are insufficient to ensure the compatibility of energy storage facilities with the power grid. It is necessary to further combine topographic constraints with the characteristics of the distribution-transmission network to assess the electrical access feasibility of the energy storage candidate locations. Step S2 aims to map the candidate pairs of pumped storage upper and lower reservoirs and the candidate locations of electrochemical energy storage to the distribution-transmission network. A geographic-electrical mapping table is formed using line impedance, substation level, and load level, and access feasibility is marked for each energy storage candidate location, providing an electrical constraint basis for subsequent energy path and power flow sensitivity analysis.

[0061] S2.1 Terrain constraints are mapped to the power distribution-transmission network.

[0062] To ensure that energy storage site selection meets both geographical and electrical requirements, the candidate sites for pumped storage upper and lower reservoirs and the candidate sites for electrochemical energy storage need to be associated with the power distribution and transmission network.

[0063] The candidate point pair set for the upper and lower reservoirs of pumped storage includes grid cell pairs that meet the requirements of elevation difference and water connectivity. Each point pair records the indices of two grid cells.

[0064] The candidate locations for electrochemical energy storage are determined by the geographical coordinates of the load center, which refers to the central location of an industrial park, residential area, or commercial area. These load centers constitute a set of candidate locations for electrochemical energy storage, and each element records the grid index of the load center.

[0065] The power distribution-transmission network is represented by a graph model, where nodes are substations or load buses, and edges are transmission lines or distribution lines. The impedance value of each line is recorded. For each pair of points in the candidate set of pumped storage upper and lower reservoirs, the geographical coordinates of its two grid cells are obtained. The distance from each grid cell to all nodes in the power distribution-transmission network is calculated, defined as the Euclidean distance between the two points' geographical coordinates. The network node with the smallest distance is selected as the mapping node for that pair, and the grid index of the pair and the corresponding mapping node number are recorded. For each load center in the candidate set of electrochemical energy storage locations, the distance from its grid index to the nearest network node is also calculated to determine the corresponding mapping node.

[0066] All mapping relationships are stored in a geographic-electrical mapping table, which contains grid indexes of pumped storage point pairs or electrochemical energy storage locations and their mapping node numbers.

[0067] S2.2 Analysis of line impedance and substation voltage level.

[0068] To quantify the electrical access characteristics of candidate energy storage locations, it is necessary to analyze their line impedance and substation voltage levels within the distribution-transmission network. Each candidate energy storage location in the geographic-electrical mapping table is associated with a specific mapping node; further extraction of the lines and substations connected to these nodes is required. Each line in the distribution-transmission network records its equivalent impedance, which consists of resistance and reactance. Resistance reflects the energy loss of the line, while reactance reflects the inductive effect of the line.

[0069] For each mapping node, determine its network path to the main load bus, which is defined as the load bus with the highest power demand in the region. Accumulate the equivalent impedance of the line segment by segment along the path to obtain the total impedance from the mapping node to the main load bus.

[0070] The path search employs Dijkstra's algorithm, using line impedance values ​​as path weights and prioritizing the path with the lowest impedance. Dijkstra's algorithm is a classic graph theory algorithm proposed by Dutch computer scientist Ezra Dijkstra in 1956. It is used to calculate the shortest path from a single source node to all other nodes in a directed or undirected graph with non-negative weights. The algorithm maintains nodes to be visited through a priority queue, expands paths according to their cumulative weights from smallest to largest, and ensures that each node with the currently known shortest path is selected for relaxation operations until all nodes have been processed. In the path search application, the distribution-transmission network is modeled as a weighted graph, where nodes represent substations or load buses, edges represent lines with weights set to line impedance values, and the mapped node of the energy storage candidate location is used as the starting point. The cumulative impedance distances of adjacent nodes are updated successively, and the path with the lowest total impedance is selected to reach the main load bus, thereby maximizing energy transfer efficiency.

[0071] The total impedance is synthesized by summing the resistance and reactance of each segment of the line along the path, using the square root of the sum of the squares of the resistance and reactance. Simultaneously, the voltage level of the substation where the mapping node is located is recorded, categorized into high voltage, medium voltage, and low voltage, corresponding to the transmission network, the main distribution network, and the end of the distribution network, respectively. High-voltage substations support high-power transmission, while low-voltage substations are suitable for small-scale distributed access. The analysis results are updated to the geographic-electrical mapping table, with new fields including the total impedance value from the mapping node to the main load bus and the voltage level of the substation where the mapping node is located.

[0072] Impedance and voltage level analysis combines the network's transmission characteristics and capacity constraints to ensure the compatibility of energy storage access locations with the network's electrical characteristics. The generated total impedance and voltage level information provides a quantitative electrical basis for subsequent access feasibility assessments.

[0073] S2.3 Load tier assessment.

[0074] To refine the matching degree between energy storage candidate locations and load demand, the mapping nodes need to be classified according to the load level.

[0075] Load buses are categorized by power consumption type into industrial loads, residential loads, and commercial loads, each exhibiting characteristics of continuous high power demand, periodic peak-valley demand, and daytime peak demand, respectively. The geographic-electrical mapping table is traversed, and for each mapping node, the type of its directly connected load bus is checked and recorded as a load level label.

[0076] Load level labels are defined as follows:

[0077] If the mapping node is connected to the industrial load bus, it is marked as an industrial load;

[0078] If connected to the residential load bus, it is marked as a residential load;

[0079] If connected to a commercial load bus, it is marked as a commercial load.

[0080] If a mapping node connects to multiple load buses, the load type with the highest power demand is selected as the label. The load hierarchy assessment reflects the suitability of energy storage candidate locations for different load demands; for example, industrial loads are suitable for the long-term power regulation of pumped hydro storage, while residential and commercial loads are suitable for the rapid response characteristics of electrochemical energy storage. The assessment results are updated in the geographic-electrical mapping table, and a new load hierarchy label field is added.

[0081] S2.4 access feasibility labeling.

[0082] To comprehensively consider geographical and electrical constraints, the access feasibility of each energy storage candidate location needs to be quantified. Access feasibility indicators are calculated based on total impedance and substation voltage level, reflecting the transmission efficiency and capacity adaptability of the energy storage access network. For each mapped node in the geographical-electrical mapping table, its total impedance value to the main load bus is obtained, and the modulus of the total impedance is calculated, i.e., the square root of the sum of the squares of the resistance and the reactance.

[0083] The substation voltage level is quantified into a numerical value, with high-voltage substations assigned the highest value, medium-voltage substations assigned the second highest value, and low-voltage substations assigned the lowest value. The access feasibility index is defined as the reciprocal of the total impedance modulus multiplied by the quantified voltage level value. The larger the value, the lower the transmission loss and the higher the electrical capacity of the energy storage connection to that node.

[0084] The geographic-electrical mapping table is traversed, and the access feasibility index is calculated for the mapping nodes of each energy storage candidate location and recorded in the table. The access feasibility index integrates the effects of impedance and voltage level to ensure that the electrical access efficiency of the energy storage location matches the network capacity. The generated geographic-electrical mapping table includes grid index, mapping node number, total impedance value, substation voltage level, load level label, and access feasibility index, providing complete electrical constraint information for the energy path length and power flow sensitivity calculation in step S3.

[0085] Step S2 maps the candidate point pairs for pumped storage upper and lower reservoirs and the candidate location set for electrochemical energy storage to the distribution-transmission network, analyzes line impedance, substation voltage level, and load level, and calculates access feasibility indicators, generating a geographic-electrical mapping table containing geographic and electrical information. The processing takes the candidate point pairs for pumped storage upper and lower reservoirs from step S1 and the newly added candidate location set for electrochemical energy storage as input, and combines network topology and load characteristics to form structured electrical constraint data.

[0086] Step S2 maps candidate pairs of pumped storage reservoirs and candidate locations for electrochemical energy storage to the distribution-transmission network, forming a geographic-electrical mapping table. However, static electrical constraints cannot fully reflect the network's dynamic response; therefore, it is necessary to introduce node load time series to calculate energy transfer attenuation and power flow impact, in order to eliminate locations unfavorable for local absorption. Step S3 aims to combine node load time series with the geographic-electrical mapping table to quantify energy path length and power flow sensitivity, generating an initial screening set to provide optimal energy storage locations for regulation task allocation.

[0087] S3.1 node load timing integration.

[0088] To capture the time-varying characteristics of power networks, the nodal load time series needs to be correlated with the geographic-electrical mapping table to reflect the dynamic load environment of energy storage candidate locations. The nodal load time series records the load power value of each mapped node in the distribution-transmission network, collected at fixed time intervals, covering the continuous high power characteristics of industrial loads, the peak-valley fluctuation characteristics of residential loads, and the daytime peak characteristics of commercial loads.

[0089] Traverse the geographic-electrical mapping table, and for each energy storage candidate location's mapping node number, extract the load power sequence of the corresponding node. The main load bus is defined as the bus in the geographic-electrical mapping table with the highest access feasibility index and a load level label of industrial load, and its load power sequence is used to identify the energy absorption target point.

[0090] The integration of node load time series ensures that the calculation process incorporates actual power flow fluctuations, facilitating the assessment of the adaptability of energy storage locations to renewable energy fluctuations. The integrated geographic-electrical mapping table adds a node load time series field.

[0091] S3.2 Calculation of energy path length.

[0092] To assess the energy transfer attenuation from candidate energy storage locations to the main load bus, it is necessary to quantify loss segments and voltage conversion complexity along the network path. Based on the mapping node numbers and total impedance values ​​in the geographic-electrical mapping table, the energy transfer path from each mapping node to the main load bus is determined, and the Dijkstra algorithm is used to search using the total impedance value as the path weight. The equivalent resistance segments of the line are recorded segment by segment along the path, defined as the product of the conductor resistance and the line length. High-loss segments and low-loss segments are divided according to a preset resistance threshold; high-loss segments refer to segments where the resistance value exceeds the threshold. Simultaneously, the number of voltage level crossings is recorded; each transition from high voltage to medium voltage or from medium voltage to low voltage at a substation is counted as a voltage step. The number of high-loss segments and voltage steps on the path are accumulated to calculate the energy path level. Specifically, this is calculated by multiplying the number of high-loss segments by a constant greater than the maximum number of voltage steps on the path, plus the number of voltage steps. This is dimensionless; the smaller the value, the lower the path level and the smaller the energy attenuation.

[0093] The energy path length classification is primarily based on loss segments, supplemented by voltage steps, to ensure priority identification of low-attenuation paths. The calculation results are updated to the geographic-electrical mapping table, with a new energy path level field added to provide a transmission efficiency indicator for subsequent elimination processes.

[0094] S3.3 Power flow sensitivity calculation.

[0095] To quantify the response of candidate energy storage locations to network bottlenecks, power injection needs to be simulated and bottleneck changes measured within a DC power flow framework. This DC power flow framework focuses on active power balance, using nodal load time series to establish the ground-state power flow distribution, while ignoring reactive power to simplify calculations.

[0096] A bottleneck line is defined as the line in the distribution-transmission network with the highest load rate and close to its capacity limit, whose base-state power flow is the power value under undisturbed conditions. For each mapped node, a uniform small-amplitude active power injection disturbance is applied, with the disturbance size being a fixed proportion of the node's average load. Power flow calculations are performed after the disturbance, and the power flow increment of the bottleneck line is recorded. Power flow sensitivity is calculated, specifically by dividing the bottleneck line's power flow increment by the injection disturbance size; this is dimensionless, and a larger absolute value indicates a more significant impact on the bottleneck.

[0097] The perturbation simulation of power flow sensitivity ensures that the actual potential of nodes to alleviate congestion is reflected, facilitating the retention of high-response locations. The calculation results are updated to the geographic-electrical mapping table, and a new power flow sensitivity field is added, providing a bottleneck impact indicator for the subsequent removal process.

[0098] S3.4 Candidate point elimination and initial screening set generation.

[0099] To focus on suitable locations for local energy absorption, candidate energy storage locations need to be screened based on calculated indicators. An upper limit for energy path level thresholds and a lower limit for power flow sensitivity thresholds are set, determined based on network topology and load fluctuation intensity. The geographic-electrical mapping table is traversed, checking the energy path level and power flow sensitivity of each candidate energy storage location, eliminating locations with energy path levels exceeding the upper limit or power flow sensitivity below the lower limit. The remaining candidate energy storage locations constitute the initial screening set, including pumped storage upper and lower reservoir candidate point pairs and electrochemical energy storage candidate locations, sorted by energy path level from smallest to largest. The threshold application during the elimination process ensures that the initial screening set emphasizes low attenuation and high sensitivity characteristics, facilitating matching with fluctuating power demand. The generated initial screening set records the grid index, mapping node number, energy path level, and power flow sensitivity of the remaining locations, providing an optimal set for the regulation demand assessment in step S4.

[0100] Step S3 integrates the node load time series into the geographic-electrical mapping table, calculates the energy path level and power flow sensitivity, and eliminates unsuitable locations based on thresholds, forming a preliminary screening set prioritizing on-site absorption. The processing takes the geographic-electrical mapping table and node load time series from step S2 as input, quantifies energy transfer and power flow response, and finally outputs the preliminary screening set containing grid indexes, mapped node numbers, energy path levels, and power flow sensitivity, ensuring that step S4 can directly reference this information for adjustment task allocation.

[0101] In distributed energy scenarios, the clustered distribution of renewable energy and diversified electricity consumption on the load side lead to drastic fluctuations in power grid flow. To achieve optimal configuration of hybrid energy storage, it is necessary to assess regulation demand after location screening to match energy storage types. Step S1 screens candidate point pairs for pumped storage upper and lower reservoirs; step S2 generates a geographic-electrical mapping table; and step S3 calculates energy path levels and power flow sensitivity to form an initial screening set. However, the initial screening set only selects preferred locations and cannot reflect differences in regulation cycles. It is necessary to assess intraday and multi-day demand in parallel, allocating pumped storage for long-cycle needs and electrochemical energy storage for short-cycle needs. Step S4 aims to assess regulation demand based on the initial screening set, assign tasks, and generate a hybrid energy storage allocation table to provide allocation information for optimization.

[0102] S4.1 Intraday adjustment demand assessment.

[0103] To capture short-cycle peak-valley energy gaps, intraday regulation demand needs to be quantified using the node load time series from the initial screening set. The initial screening set includes grid indices, mapped node numbers, energy path levels, and power flow sensitivity for energy storage candidate locations, with each mapped node associated with a node load time series.

[0104] Select a typical intraday period and calculate the difference between the intraday highest load power and the lowest load power for each mapping node. The intraday highest load power is the maximum value in the time series, and the intraday lowest load power is the minimum value.

[0105] Intraday adjustment demand is defined as the sum of the total number of time points from time one to time point one. Each term is the load power of the mapped node at that time point minus the lowest load power of the day, multiplied by the time interval. The integral calculation of intraday adjustment demand approximates the total peak-valley fluctuation, ensuring the identification of energy demand suitable for rapid response. The evaluation results are added to the initial screening set, and an intraday adjustment demand field is added to provide short-cycle indicators for subsequent allocation.

[0106] S4.2 Multi-day adjustment demand assessment.

[0107] To reflect long-term energy imbalances, it is necessary to extend the assessment to multi-day regulation demand, using the node load time series from the initial screening set.

[0108] By selecting a continuous multi-day period, the multi-day overall average load power of each mapping node is calculated as the arithmetic mean of the time series, and the daily average load power is calculated as the arithmetic mean of the daily series.

[0109] Multi-day adjustment demand is defined as the sum of the totals from day one to day two. Each term represents the absolute difference between the average load power of the mapped node on that day and the overall average load power over the multi-day period, multiplied by the daily interval. The cumulative absolute value calculation of multi-day adjustment demand captures cross-day power deviations, ensuring the identification of energy demand suitable for continuous relocation. The evaluation results are appended to the initial screening set, and a new multi-day adjustment demand field is added, providing a long-term indicator for subsequent allocation.

[0110] S4.3 Adjust task assignment.

[0111] To accommodate the characteristics of different energy storage types, the assessment needs should be allocated to pumped hydro storage and electrochemical energy storage.

[0112] The initial screening set distinguishes between candidate pumped-storage reservoir pairs and candidate electrochemical storage locations. The former is suitable for long-cycle regulation, while the latter is suitable for short-cycle regulation. Iterating through the initial screening set, for each candidate storage location, if it belongs to a pumped-storage reservoir pair, the multi-day regulation demand is recorded as a long-cycle power transfer task, with a value equal to the multi-day regulation demand. If it belongs to an electrochemical storage location, the intraday regulation demand is recorded as a short-cycle peak-valley regulation task, with a value equal to the intraday regulation demand. The allocation method for regulation tasks ensures that long-cycle tasks correspond to the durability characteristics of pumped-storage, and short-cycle tasks correspond to the agility characteristics of electrochemical storage. The allocation results are updated to the initial screening set, adding regulation task type and value fields to provide allocation details for subsequent records.

[0113] S4.4 Hybrid Energy Storage Allocation Table Generation.

[0114] To aggregate task allocation information, a hybrid energy storage allocation table needs to be generated from the initial screening data. The hybrid energy storage allocation table is presented in tabular form, with each row corresponding to a candidate energy storage location. Columns include grid index, mapping node number, energy path level, power flow sensitivity, intraday regulation demand, multi-day regulation demand, regulation task type, and value. The initial screening data is traversed, grouped by energy storage type. For pumped storage, the total value of long-cycle tasks is calculated as the sum of multi-day regulation demands for all relevant locations. For electrochemical energy storage, the total value of short-cycle tasks is calculated as the sum of intraday regulation demands for all relevant locations. The type of allocation is indicated in the table header. The structured organization of the hybrid energy storage allocation table ensures that the allocation results are complete and traceable, facilitating optimization. The generated hybrid energy storage allocation table contains all evaluation and allocation information, providing input data for the joint optimization in step S5.

[0115] Step S4 assesses intraday and multi-day regulation demands in the initial screening set, allocating multi-day regulation demands to pumped storage as long-cycle power transfer tasks and intraday regulation demands to electrochemical energy storage as short-cycle peak-valley regulation tasks, generating a hybrid energy storage allocation table. The processing takes the initial screening set from step S3 and the node load time series as input, quantifies the demand, matches types, and finally outputs a hybrid energy storage allocation table containing grid indexes, mapped node numbers, regulation demands, task types, and values, ensuring that step S5 can directly reference this information for energy storage location and capacity optimization.

[0116] In distributed energy scenarios, the clustered distribution of renewable energy and diversified electricity consumption on the load side lead to drastic fluctuations in power grid flow. To achieve coordinated optimization of power generation, grid, load, and storage, multiple models need to be jointly solved after task assignment to determine the final layout. Step S1 filters candidate point pairs for pumped storage upper and lower reservoirs; step S2 generates a geographic-electrical mapping table; step S3 forms an initial screening set; and step S4 generates a hybrid energy storage allocation table. However, the hybrid energy storage allocation table only assigns tasks and cannot solve for location and capacity. It is necessary to integrate the generation model, line interruption model, and load response model for optimization, while retaining the terrain constraint list and power flow sensitivity. Step S5 aims to call the hybrid energy storage allocation table to optimize the location and capacity of energy storage, outputting the energy storage layout scheme and tiered charging and discharging sequence to ensure overall efficiency.

[0117] S5.1 Model Integration and Constraint Preservation.

[0118] To achieve multi-side resource synergy, the power generation model, line interruption model, load response model, and hybrid energy storage allocation table need to be combined, while retaining the topographic constraint list and power flow sensitivity as optimization inputs. The power generation model simulates the time-series power output of wind and solar power. The line interruption model calculates the branch capacity limits after topological changes. The load response model estimates the demand-side adjustable proportion. The hybrid energy storage allocation table includes the grid index, mapping node number, adjustment task type, and value of the initial screening locations. Traversing the hybrid energy storage allocation table, for each energy storage candidate location, the elevation difference and water connectivity are extracted from the topographic constraint list in step S1, and the power flow sensitivity is extracted as the adjustment weight in step S3. The constraint retention integration method ensures that the optimization considers geographical feasibility and network response, with the overall objective being to minimize the sum of construction costs and operating losses. The model framework is constructed by solving mixed integer programming or linear programming to ensure that the results conform to actual engineering conditions.

[0119] S5.2 Energy storage location optimization.

[0120] To balance cost and network stability, the location variable needs to be solved in the integrated model, incorporating a list of terrain constraints and power flow sensitivity weights. The location variable is a binary value, denoted as 1 for selected locations and 0 for unselected locations.

[0121] The objective function is optimized to minimize the following sum: the sum of construction costs at all locations multiplied by location variables, plus the sum of the absolute difference between the power flow sensitivity and the ideal sensitivity threshold at all locations multiplied by the adjustment task value multiplied by location variables.

[0122] The construction cost is determined based on the elevation difference and water connectivity in the topographic constraint list. For example, for a location in a candidate pair of upper and lower reservoirs for pumped storage, if the elevation difference recorded in the topographic constraint list is a specific height and the water connectivity indicator is connected, the construction cost can be calculated by adding the elevation difference compensation fee and the water diversion fee to the basic cost. The elevation difference compensation fee is proportional to the elevation difference threshold minus the actual elevation difference, reflecting the additional expenditure on earthwork. The water diversion fee adds a fixed proportion of the engineering cost when the connectivity indicator is not connected, thus ensuring that the cost reflects the actual impact of geographical conditions.

[0123] The adjustment task value is obtained from the hybrid energy storage allocation table.

[0124] Constraints include the total number of locations not exceeding the budget limit, the elevation difference meeting a preset threshold, and water connectivity meeting requirements. The deviation minimization calculation for location optimization ensures optimal tidal flow distribution through location selection, using a mixed-integer programming algorithm for iterative solution. The optimization results update the hybrid energy storage allocation table, adding a location variable field to provide selected locations for capacity calculation.

[0125] S5.3 Energy storage capacity optimization and charge / discharge sequence generation.

[0126] To cover the regulation requirements, the capacity variables at the selected locations need to be optimized and a graded charge-discharge sequence needs to be generated.

[0127] The capacity variable is a continuous value, representing the energy storage capacity at each location.

[0128] The objective function is optimized to minimize the following sum: the sum of the unit capacity cost multiplied by the capacity variable at all locations, plus the total regulation demand at all time points minus the sum of the charging and discharging energy at all locations. The unit capacity cost is determined based on the energy storage type in the hybrid energy storage allocation table, the total regulation demand is aggregated from the generation model and the load response model, and the charging and discharging energy is the output of the location at the time point.

[0129] For example, regarding the determination of unit capacity cost, if the energy storage type in the hybrid energy storage allocation table is pumped storage, then the unit capacity cost can be calculated as the fixed foundation cost per megawatt-hour plus the additional costs of terrain excavation and reservoir construction, reflecting the investment intensity of large-scale projects. If the energy storage type is electrochemical energy storage, then the unit capacity cost is determined based on battery materials and installation costs, and the cost per megawatt-hour adjusts as technology maturity decreases. Regarding the aggregation of total regulation demand, if the power generation model simulates wind and solar output time series showing a specific daily power deficit, and the load response model estimates a different demand-side adjustable amount, then the total regulation demand is obtained by summing the two, reflecting the overall energy balance requirements of the network.

[0130] Constraints include that the capacity variable does not exceed the upper limit of the access feasibility index, and that the power flow under the line interruption model does not exceed the branch capacity. The tiered charge-discharge sequences are classified into long-period sequences for pumped hydro storage and short-period sequences for electrochemical energy storage, with the charge-discharge power recorded at each time point in each sequence. Unmet demand minimization calculations for capacity optimization ensure energy balance, and a linear programming algorithm is used for solution. The optimization results form an energy storage deployment scheme, including location, capacity values, and sequence details.

[0131] In power systems, power flow specifically refers to the steady-state distribution and flow of power in a power network, including the flow of active and reactive power along each branch. This term describes the power distribution of the system under given operating conditions, and when related to branch capacity constraints, it indicates that the power flow must not exceed the transmission capacity limit of the line.

[0132] S5.4 output scheme.

[0133] To present the complete results, the energy storage deployment scheme and tiered charge-discharge sequences need to be compiled. The energy storage deployment scheme organizes the grid index, mapped node number, energy storage type, and capacity value of the selected locations in tabular form, and verifies the elevation difference and water connectivity of the terrain constraint list. The tiered charge-discharge sequences are represented as time-series matrices, with long-period sequences handling multi-day power shifts and short-period sequences handling intraday peak-valley regulation. Each column corresponds to a time point, and each row corresponds to the power at a location. The output process checks whether the power flow sensitivity weights alleviate bottlenecks to ensure scheme consistency. The tabular and matrix formats of the scheme output facilitate implementation and enable collaborative planning.

[0134] The following is a simplified hypothetical example, assuming that the optimization results selected two energy storage locations: a pumped storage location (grid index (10,15), mapping node number N1, type pumped storage, capacity 500 MWh, elevation difference 150 meters, water connectivity verified) and an electrochemical energy storage location (grid index (20,25), mapping node number N2, type electrochemical energy storage, capacity 200 MWh, elevation difference not applicable, water connectivity not applicable). The energy storage layout scheme is shown in Table 1.

[0135] Table 1: Energy Storage Deployment Scheme

[0136]

[0137] The graded charge-discharge sequence is represented by a time series matrix. The long-period sequence (pumped storage, handling multi-day power transfer, assuming a 3-day period, simplified to one power value per day, in megawatts) is shown in Table 2.

[0138] Table 2: Long-Period Sequence Staged Charge-Discharge Sequences

[0139]

[0140] The short-period sequence (electrochemical energy storage, handling intraday peak-valley regulation, assuming a 4-hour period, unit is megawatt) is shown in Table 3.

[0141] Table 3: Short-Period Sequence Graded Charge-Discharge Sequence Table

[0142]

[0143] After verifying that the power flow sensitivity weights (e.g., N1 is 0.8 and N2 is 0.6) alleviate the bottleneck in the output process, the consistency of the scheme is ensured.

[0144] Step S5 integrates the generation model, line interruption model, load response model, and hybrid energy storage allocation table to optimize energy storage location and capacity, retaining the terrain constraint list and power flow sensitivity, and generating energy storage layout schemes and tiered charge / discharge sequences. The processing starts from the hybrid energy storage allocation table in step S4, solves for location and capacity variables, and finally outputs a scheme containing grid indexes, capacity values, and time-series power, ensuring optimization consistency.

[0145] Example 2: Figure 2 This invention presents a hybrid energy storage optimization system integrating source-grid-load flexibility resources, comprising:

[0146] Terrain filtering module: Based on terrain elevation grid and water body distribution data, it generates candidate point pairs for pumped storage upper and lower reservoirs and a list of terrain constraints through elevation threshold filtering and water body connectivity analysis.

[0147] Network mapping module: Maps candidate points of pumped storage upper and lower reservoirs and candidate locations of electrochemical energy storage to the power distribution and transmission network. It generates a geographic-electrical mapping table based on line impedance, substation voltage level and load level, and marks access feasibility indicators for each energy storage candidate location.

[0148] Location screening module: Combining node load time series, using geographic-electrical mapping table to calculate energy path level and power flow sensitivity, and eliminating energy path level exceeding the corresponding standard or power flow sensitivity below the corresponding standard to generate an initial screening set with priority for local absorption.

[0149] Task assignment module: In the initial screening, the intraday regulation demand and multi-day regulation demand are evaluated in parallel. The multi-day regulation demand is assigned to pumped storage as a long-cycle power transfer task, and the intraday regulation demand is assigned to electrochemical energy storage as a short-cycle peak-valley regulation task. A hybrid energy storage allocation table is generated to record the assignment results.

[0150] Joint optimization module: It calls the hybrid energy storage allocation table and the power generation model, line interruption model and load response model to jointly optimize the location and capacity of energy storage, retain the list of terrain constraints and power flow sensitivity to output the energy storage layout scheme and the graded charging and discharging sequence.

[0151] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0152] It should be noted that the system of the present invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting a variety of hardware environments and usage requirements.

[0153] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0154] It should be noted that, in this document, the use of relational terms such as "first" and "second" is merely to distinguish one entity or operation from another, and does not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0155] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A hybrid energy storage optimization method of source-network-payload flexibility resource integration, characterized in that, The method comprises the steps of: S1: Based on the terrain elevation grid and water body distribution data, the upper and lower reservoir candidate point pairs of pumped storage power station and the terrain constraint list are generated through elevation threshold screening and water body connectivity analysis; S2: The upper and lower reservoir candidate point pairs of pumped storage power station and the electrochemical energy storage candidate position are mapped to the power distribution-transmission network, a geographical-electrical mapping table is generated based on line impedance, substation voltage grade and load level, and an access feasibility index is marked for each energy storage candidate position; S3: Combined with the node load time sequence, the geographical-electrical mapping table is used to calculate the energy path level and the power flow sensitivity, and the energy storage candidate positions with energy path level exceeding the corresponding standard or power flow sensitivity lower than the corresponding standard are removed, and the initial screening set of local absorption priority is generated; S4: In the initial screening set, the daily regulation demand and the multi-day regulation demand are evaluated in parallel, the multi-day regulation demand is assigned to the pumped storage power station as a long-period power shifting task, and the daily regulation demand is assigned to the electrochemical energy storage as a short-period peak-valley regulation task, and a mixed energy storage allocation table is generated to record the assignment results; S5: The mixed energy storage allocation table is called to jointly optimize the energy storage location and capacity with the power generation model, line opening model and load response model, the terrain constraint list and power flow sensitivity are retained, and the energy storage arrangement scheme and hierarchical charging and discharging sequence are output. Step S5 includes the following contents: The elevation values, water body connectivity and power flow sensitivity in the terrain constraint list are extracted as weights by traversing the mixed energy storage allocation table, and the sum of the sum of the construction cost of all positions multiplied by the location variable and the absolute difference between the power flow sensitivity of all positions and the ideal sensitivity threshold multiplied by the regulation task value multiplied by the location variable is minimized; The constraint conditions include that the total number of location positions does not exceed the upper limit of the budget, and the elevation value meets the preset threshold and the water body connectivity meets the requirement; The sum of the sum of the unit capacity cost of all positions multiplied by the capacity variable and the sum of all time points of the total regulation demand minus the sum of all position charging and discharging energy is minimized; The constraint conditions include that the capacity variable does not exceed the upper limit of the access feasibility index, and the power flow under the line opening model does not exceed the branch capacity. 2.The source-network-payload flexible resource integrated hybrid energy storage optimization method of claim 1, wherein, Step S1 includes the following contents: A terrain elevation grid is constructed, the absolute difference of the altitude of the center points of adjacent grid cells is calculated, and an elevation grid matrix is formed; The elevation grid matrix is traversed, and grid cell pairs with an elevation value greater than or equal to a preset minimum elevation threshold are screened to generate a preliminary candidate point pair set; A water body distribution layer is generated using hydrological data, the water body existence state of each grid cell is marked, the water body connectivity of the grid cell pairs in the preliminary candidate point pair set is checked, and the point pairs with at least one grid cell containing water and no elevation barrier between the two grid cells are retained to form a pumped storage power station upper and lower reservoir candidate point pair set, and a terrain constraint list is generated. 3.The source-network-payload flexible resource integrated hybrid energy storage optimization method of claim 2, wherein, Step S2 includes the following contents: Based on the geographical coordinates of the grid cells in the pumped storage power station upper and lower reservoir candidate point pair set and the load center grid index in the electrochemical energy storage candidate position set, the Euclidean distance between the geographical coordinates of the grid cells and the geographical coordinates of the power distribution-transmission network nodes is calculated, the network node with the smallest distance is selected as the mapping node, and a geographical-electrical mapping table is generated to record the grid index and the mapping node number. 4.The source-network-payload flexible resource integrated hybrid energy storage optimization method of claim 3, wherein, Step S2 further includes the following contents: Accumulate the equivalent impedance of each section of line along the network path from the mapping node to the main load bus to obtain the total impedance value, and record the total impedance value and the voltage level of the substation; According to the load level label recorded by the mapping node connected to the load bus type, analyze the line impedance, substation voltage level and load level to obtain the access feasibility index, and update the geographical-electricity mapping table.

5. The hybrid energy storage optimization method of source, network and payload flexibility integration according to claim 4, characterized in that, Step S3 includes the following contents: Integrate the node load time sequence into the geographical-electricity mapping table, and record the load power sequence of each mapping node; Record the line equivalent resistance section and voltage level crossing times along the energy transmission path from the mapping node to the main load bus, accumulate the number of high-loss section and voltage step number on the path, and obtain the energy path level.

6. The hybrid energy storage optimization method of source, network and payload flexibility integration according to claim 5, characterized in that, Step S3 also includes the following contents: Under the framework of DC power flow, apply active power injection disturbance to each mapping node, calculate the power flow increment of the bottleneck line divided by the injection disturbance size to obtain the power flow sensitivity; Remove the energy path level exceeding the upper threshold or the power flow sensitivity below the lower threshold of the energy storage candidate position, and sort the remaining positions in the initial screening set according to the energy path level from small to large.

7. The hybrid energy storage optimization method of source, network and payload flexibility integration according to claim 6, characterized in that, Step S4 includes the following contents: Use the node load time sequence of each mapping node in the initial screening set to obtain the intraday regulation demand, and use the node load time sequence to obtain the multi-day regulation demand; Assign the multi-day regulation demand to the upper and lower reservoir candidate point pair of pumped storage as a long-period power shifting task; Assign the intraday regulation demand to the electrochemical energy storage candidate position as a short-period peak-valley regulation task; Group by storage type, calculate the total value of long-period task as the sum of multi-day regulation demand of pumped storage position, and the total value of short-period task as the sum of intraday regulation demand of electrochemical energy storage position; Generate a mixed energy storage allocation table.

8. A hybrid energy storage optimization system for source, network, and load flexibility resource integration, used to implement the hybrid energy storage optimization method for source, network, and load flexibility resource integration according to any one of claims 1-7, characterized in that, It includes: Terrain screening module: Based on the terrain elevation grid and water body distribution data, the upper and lower reservoir candidate point pair of pumped storage and the terrain constraint list are generated through elevation threshold screening and water body connectivity analysis; Network mapping module: Map the upper and lower reservoir candidate point pair of pumped storage and the electrochemical energy storage candidate position to the distribution-transmission network, generate the geographical-electricity mapping table based on the line impedance, substation voltage level and load level, and mark the access feasibility index for each energy storage candidate position; Position screening module: Combine the node load time sequence, use the geographical-electricity mapping table to calculate the energy path level and the power flow sensitivity, and remove the energy storage candidate position with energy path level exceeding the corresponding standard or power flow sensitivity below the corresponding standard to generate the initial screening set with priority of local absorption; Task assignment module: In the initial screening set, evaluate the intraday regulation demand and the multi-day regulation demand in parallel, assign the multi-day regulation demand to the pumped storage as a long-period power shifting task, and assign the intraday regulation demand to the electrochemical energy storage as a short-period peak-valley regulation task to generate a mixed energy storage allocation table to record the assignment results; Joint optimization module: Call the mixed energy storage allocation table and the power generation model, line opening model and load response model to jointly optimize the energy storage location and capacity, retain the terrain constraint list and power flow sensitivity, and output the energy storage layout scheme and hierarchical charge-discharge sequence.

Images