A distributed energy storage cloud-edge collaborative optimization configuration system and method
By introducing information-physical net capacity calculation and intelligent terminal collaborative control in the distribution area, the problem of the unconsidered impact of information transmission reliability in energy storage optimization configuration is solved, realizing the optimized access and operation strategy of the energy storage system and improving the robustness and economy of the system.
Patent Information
- Application Number
- CN202611421566.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-14
- Publication Date
- 2026-11-17
AI Technical Summary
Existing energy storage optimization methods fail to comprehensively consider the distribution network topology, branch impedance, and information transmission reliability, which affects the system's operational safety and economy, especially when information fails and energy storage cannot be effectively regulated.
By introducing information-physical net capacity calculation, the main station-edge collaborative control is realized through the intelligent terminal of the transformer area. It integrates data acquisition and preprocessing, information-physical net capacity calculation and site selection, multi-scenario collaborative optimization modeling and optimization solution and strategy output modules to determine the optimal access point for energy storage and optimize the operation strategy.
It improves the system's robustness to information failures, reduces the load peak-to-valley difference, smooths load fluctuations, enhances the distributed photovoltaic absorption capacity, and optimizes the system's daily comprehensive cost.
Smart Images

Figure CN123121728A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the technical fields of distributed energy storage planning and scheduling, and specifically to a distributed energy storage cloud-edge collaborative optimization configuration system and method. Background Technology
[0002] The high proportion of distributed photovoltaic (PV) power generation connected to the distribution network presents significant challenges to the safe and stable operation of the network due to the volatility and randomness of its output. Energy storage, with its fast response speed and flexible charging / discharging power adjustment, plays a crucial role in promoting large-scale distributed PV grid connection planning. However, the rational configuration of energy storage is a prerequisite for its effective application. Optimal energy storage configuration generally involves comprehensively considering various factors, including the energy storage technology itself, to improve system reliability, safety, and economy, seeking the optimal configuration (optimal location, optimal capacity, and optimal dispatch) solution.
[0003] Currently, the main methods for optimal energy storage location include sensitivity analysis and optimization methods aimed at system reliability and economy. However, these methods do not comprehensively consider factors such as distribution network topology, branch impedance, and power flow distribution. Furthermore, optimization methods aimed at system reliability and economy have long solution times. Because distributed energy storage systems are decentralized and connected to the distribution network, their controllability is poor. Additionally, distributed photovoltaic systems and loads exhibit strong volatility, leading to problems such as increased net load fluctuations and exacerbated peak-to-valley differences.
[0004] Furthermore, existing energy storage site selection methods assume that the information system is completely reliable and do not consider the impact of random information transmission failures (such as communication link interruptions and smart electronic device failures) on the actual control capabilities of energy storage. As the informatization level of active distribution networks deepens, the coupling between information systems and physical power grids becomes increasingly close. Information failures may cause energy storage to be unable to receive or execute dispatch commands, thereby seriously affecting the safety and economy of system operation.
[0005] In the distribution IoT architecture, the intelligent converged terminal, as the core management unit and key edge computing node of the intelligent distribution area, possesses multi-source data access, edge computing processing, and bidirectional communication capabilities with the main station system. It enables real-time monitoring, status awareness, power quality management, and business order processing of distributed resources within the distribution area, including energy storage devices and photovoltaic power generation equipment. In particular, this terminal simultaneously acts as the edge computing and communication aggregation node between all energy storage devices within the distribution area and the main station. On one hand, the terminal's built-in edge computing strategy requires local parsing and rapid response to energy storage charging and discharging commands; on the other hand, the communication link between the terminal and the main station directly affects the reliable issuance of dispatching commands. Therefore, the availability rate of the intelligent terminal in the distribution area—that is, the probability of its normal operation and uninterrupted communication—directly determines the dispatchability of each energy storage device within the distribution area. If the terminal malfunctions, communication is interrupted, or edge computing services fail, even if the main station issues optimization commands, the energy storage cannot be effectively executed, resulting in the failure of control targets such as peak shaving and valley filling, and distributed photovoltaic power consumption. Cyber-physical net capacity is a comprehensive indicator based on complex network theory, electrophysical characteristics, and information impact factors. A node with high cyber-physical net capacity is not only a crucial point for power transmission in the power grid, but its information nodes are also vital. Its failure can simultaneously trigger physical power flow exceeding limits and information interruption, accurately reflecting the impact of information failure on the actual control capability of energy storage. By considering the importance of nodes in the information network, it quantifies the structured transmission capacity of the power grid, providing a scientific basis for the site selection of distributed energy storage and serving as an important theoretical foundation for achieving synergistic optimization of photovoltaic and energy storage. The power grid is not merely a simple combination of lines and nodes, but a network with a specific topology. The role of net capacity is to quantify the ability of this network to transmit power from the generation end to the load end under physical constraints. However, existing energy storage site selection and scheduling technologies generally assume that the cyber-physical system is completely reliable, only considering the topological impedance and power constraints of the physical power grid, and do not incorporate the availability of key information nodes such as smart terminals in the distribution area into the cyber-physical net capacity evaluation system. Therefore, it is difficult to accurately reflect the impact of information failure on the actual control capability of energy storage.
[0006] Therefore, how to take into account the impact of the reliability of information nodes, including smart terminals in the distribution area, in the energy storage optimization configuration, and to achieve cloud-edge collaborative control, has become a technical problem that urgently needs to be solved in this field. Summary of the Invention
[0007] In view of this, to at least solve some of the aforementioned technical problems, this invention provides a distributed energy storage cloud-edge collaborative optimization configuration system and method, which can be integrated into the distribution network scheduling and operation extension function block of a distribution automation system. This invention achieves optimal energy storage location through information-physical fusion by introducing an information influence factor into net capacity calculation, and realizes master-edge collaborative control through intelligent terminals in distribution areas.
[0008] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0009] In a first aspect, the present invention provides a distributed energy storage cloud-edge collaborative optimization configuration system, the system comprising: The data acquisition and preprocessing module is used to acquire physical layer data and information layer data of the distribution network from external systems through the information interaction bus, and to perform data cleaning, standardization and scenario generation preprocessing; the external systems include, but are not limited to, production management system, upper-level dispatch automation system, power grid GIS platform, marketing management information platform, 95598 system, and intelligent terminals in the distribution area; The information-physical net capacity calculation and location module, whose input is connected to the data acquisition and preprocessing module, is used to determine the information-physical net capacity of the distributed energy storage (DES) system based on the power transmission capacity and equivalent resistance of the distributed energy storage system after it is connected to the grid, as well as the reliability parameters of the information system on which the node of the distributed energy storage system depends. Based on the information-physical net capacity, the optimal access point of the distributed energy storage system is determined. The multi-scenario collaborative optimization modeling module has its input end connected to the information-physical net capacity calculation and location module. Based on the optimal access point of DES, it establishes a distributed energy storage optimization model in an active distribution network with source-grid-load coordination. The objective function of the distributed energy storage multi-scenario collaborative optimization model includes the system's daily comprehensive cost and peak shaving and valley filling. The optimization solution and strategy output module, whose input is connected to the multi-scenario collaborative optimization modeling module, solves the objective function of the distributed energy storage collaborative optimization model in the active distribution network of source-grid-load collaboration, and obtains the operation strategy of the distributed energy storage system. The distributed energy storage system includes energy storage devices deployed on each node of the distribution network, including but not limited to user-side energy storage, feeder-side energy storage and transformer area energy storage. The operation strategy is issued to the transformer area smart terminal through the distribution automation system. The transformer area smart terminal selects to execute the dispatch command or start the local autonomous control strategy based on the local cyber-physical net capacity assessment results.
[0010] Furthermore, the data acquisition and preprocessing module obtains data through the following methods: physical layer data and information layer data from the production management system, the higher-level dispatch automation system, the power grid GIS platform, the marketing management information platform, the 95598 system, and the intelligent terminals in the distribution area. The physical layer data includes real-time data, historical data, electrical single-line diagrams, network topology and equipment information, power outage information, and distributed energy storage data. The information layer data includes the historical mean time between failures (MTBF) and mean time to repair (MTBR) of components, as well as the MTBF and MTBR of the intelligent terminals in the distribution area. The data acquisition frequency can be set manually.
[0011] Furthermore, in the cyber-physical net capacity calculation and location module, after the distributed energy storage system is connected to the grid, the expression for calculating the grid's cyber-physical net capacity is:
[0012] In the formula, Represents information - net physical capacity; , These represent the number of generators and the number of load nodes, respectively. , These are the generator and the load node, respectively. , These are the sets of generators and load nodes, respectively. This indicates the branch transmission capacity of the generator-load pair after distributed energy storage is connected to the grid. This represents the equivalent resistance of distributed energy storage after it is connected to the power grid. For load nodes The information impact factor characterizes the reliability of the information system that the node depends on in its control capabilities.
[0013] Furthermore, information impact factor The calculation method is as follows:
[0014] in, For nodes The set of key information components along the information transmission path on which it depends, including but not limited to intelligent electronic devices (IEDs), communication links, optical line terminals (OLTs), optical network units (ONUs), passive optical splitters (POSs), servers, and smart terminals in the distribution area. For the first The availability of each information element is calculated using the following formula:
[0015] in, For components Mean time between failures (MTBF) For components The average repair time.
[0016] Furthermore, the information transmission path is from the server of the power distribution information system to the node. The set of communication links between corresponding intelligent electronic devices (IEDs) is such that when a node has a substation intelligent terminal, the terminal is included in the path as a necessary node, and its availability is equal to the product of the availability of all information elements on the path.
[0017] This invention calculates equivalent impedance, equivalent capacity, power transfer distribution factor, and information element reliability separately using the net capacity equation. By comparing the net cyber-physical capacity value of the system at each node, nodes with larger net cyber-physical capacity values are more beneficial to the system and are therefore selected as the optimal access locations for energy storage. This achieves decoupling of the topology from branch impedance, branch transmission power limits, etc.
[0018] Furthermore, and They are obtained using the following formulas respectively:
[0019] in, For the transmission capacity of the branch of the energy storage access system; , These are the charging and discharging times for energy storage, respectively. Represents a node and nodes between the capacities, Represents a node To the node The capacity between;
[0020]
[0021] In the formula, For the system impedance matrix OK Column elements; For the system impedance matrix OK Column elements; For the system impedance matrix OK Column elements; Nodes represented by equivalent impedance and nodes Electrical distance between Nodes represented by equivalent impedance and nodes Electrical distance between For energy storage access nodes After that, node ,node ,node The equivalent impedance between.
[0022] Furthermore, in the multi-scenario collaborative optimization modeling module, the objective function expression of the source-grid-load-storage multi-scenario collaborative optimization model is:
[0023] In the formula: These are the weighting coefficients, and , The total daily cost of the system, This is an indicator for peak shaving and valley filling.
[0024] This module uses fuzzy hierarchical analysis and entropy weighting method for comprehensive evaluation to determine the optimal weight coefficients of each sub-objective in the multi-objective function. This approach can take into account expert experience while reducing the influence of subjective factors on the results, and can more reasonably determine the weight of each evaluation indicator.
[0025] Based on the above, the weights obtained by fuzzy analytic hierarchy process are: The weights obtained by the entropy weight method are Combined weights The calculation formula is as follows:
[0026] In the formula: This represents the number of evaluation indicators.
[0027] The system's daily comprehensive cost expression is:
[0028] in, The objective function for the system's daily comprehensive cost; , These are the investment costs and operation and maintenance costs of Distributed Energy Storage (DES), respectively. These are the investment cost and operation and maintenance cost of the smart soft switch (SoftOpen Point, SOP). The cost of purchasing electricity from the main grid.
[0029] To fully account for the impact of the uncertainty in distributed photovoltaic (DPV) output on system operation, the costs under various typical scenarios are calculated using a weighted average. The cost calculation formulas for each part are as follows:
[0030] In the formula: For the scene The probability of occurrence; These are respectively a set of typical scenarios, a set of SOP ports, and a set of lines; For the discount rate, These refer to the investment periods for DES and SOP, respectively. These represent the number of DES and DPV connections, respectively. These are the unit capacity cost of DES, unit power cost of DES, unit operation and maintenance cost of DES, unit capacity cost of SOP, time-of-use electricity price, and unit curtailment cost of solar power. The first The installed capacity and power of the DES unit; They are respectively DES charging and discharging power at all times; For port capacity; This is the cost coefficient for SOP operation and maintenance; branch road The resistance; for Branch roads during time periods The current; The first Taiwan's predicted DPV output and actual dispatched output; for Active power flowing from the main grid to this grid during a given time period; The scheduling time is in units.
[0031] Furthermore, in the multi-scenario collaborative optimization modeling module, multiple scenarios are obtained through the following steps: (1) Perform Monte Carlo sampling on the random variable of photovoltaic output to generate N scenarios, and calculate the output of each scenario according to the following formula. 2-norm ,
[0032] In the formula: It is a 2-norm; For the first The first scenario One value, For the first The first scenario There are several values, where L is the length of the scene sequence.
[0033] (2) Obtain each pair of scenes in all scenes L2 norm and scenarios probability of occurrence The set of minimum products, denoted by set D, is represented as:
[0034] (3) Reduce set D according to the following formula to remove scenarios that are not representative and have a low occurrence rate;
[0035] In the formula: To reduce the number of scenes The probability of occurrence; For the scene The probability of occurrence.
[0036] (4) For the elements in set D, sort them in ascending order according to their probability distance; (5) Determine the scenario with the smallest distance from the reduced scenario probability, and update the scenario probability and the number of scenarios in set D.
[0037] (6) Repeat steps (2)-(5) until the required number of scenarios is met, then stop updating.
[0038] Furthermore, peak shaving and valley filling The expression is:
[0039]
[0040] In the formula: A collection of scenes; For the scene The probability of occurrence; T is the scheduling period; The net load at time t; Let t be the load. , These are the energy storage charging and discharging power, respectively. This indicates the actual dispatch output of distributed photovoltaic power.
[0041] Furthermore, the constraints of the multi-scenario collaborative optimization model of source-grid-load-storage include: second-order cone relaxation power flow constraints, photovoltaic power output constraints, on-load tap changer (OLTC) regulation, reactive power compensation device, optimized operation of smart soft switch (SOP), grid connection optimization of distributed energy storage (DES), as well as communication bandwidth constraints and edge computing latency constraints of intelligent terminals in the distribution area.
[0042] Furthermore, second-order cone relaxation is used to process the soft open point (SOP) constraints, and the Big-M method is used to process the constraints containing binary variables. When the information influence factor of the node where the smart terminal in the distribution area is located is lower than the preset threshold, the terminal automatically switches to the local autonomous control mode and independently regulates the distributed energy storage in the distribution area based on local measurement information.
[0043] Furthermore, the expressions for the communication bandwidth constraints and edge computing latency constraints of the intelligent terminals in the distribution area are as follows: Communication bandwidth constraints:
[0044] In the formula, , For each of the smart terminals in the distribution area Uplink and downlink communication rates at any given time (unit: Mbps); , These represent the maximum allowed uplink and downlink bandwidth (unit: Mbps).
[0045] Edge computing latency constraints:
[0046] In the formula: The total latency (in milliseconds) from receiving instructions from the main station to completing local autonomous control decisions and sending them to the energy storage device. This represents the maximum allowable delay.
[0047] Secondly, embodiments of the present invention also provide a distributed energy storage cloud-edge collaborative optimization configuration method. This method applies the above-described system to achieve distributed energy storage cloud-edge collaborative optimization configuration, and includes the following steps: S1. Obtain physical layer data and information layer data of the distribution network from external systems, and preprocess the obtained data; S2. Based on the preprocessed data, calculate the net information-physical capacity of the distributed energy storage system, and determine the optimal access point of the distributed energy storage system according to the net information-physical capacity; wherein the calculation of the net information-physical capacity introduces an information impact factor, and the information impact factor incorporates the availability rate of the smart terminals in the distribution area; S3. Based on the optimal access point of the distributed energy storage system, establish a multi-scenario collaborative optimization model of source-grid-load-storage with the goal of system daily comprehensive cost and peak shaving and valley filling; S4. Solve the objective function of the source-grid-load-storage multi-scenario collaborative optimization model to obtain the operation strategy of the distributed energy storage system, and send the operation strategy to the intelligent terminal of the distribution area. The intelligent terminal of the distribution area selects to execute the scheduling command or start the local autonomous control based on the evaluation result of the local information impact factor.
[0048] As can be seen from the above technical solutions, the solutions disclosed in this invention have at least the following beneficial effects compared with the prior art: This invention provides a distributed energy storage cloud-edge collaborative optimization configuration system and method, which is integrated into the distribution network dispatching and operation extension function block of the distribution automation system. The system includes: a data acquisition and preprocessing module, which acquires multi-source data through an information interaction bus; an information-physical net capacity calculation and location module, used to calculate the net capacity taking into account information impact factors and determine the optimal access point for energy storage, wherein the information impact factor incorporates the availability rate of the smart terminals in the distribution area; a multi-scenario collaborative optimization modeling module, used to establish a source-grid-load-storage multi-scenario collaborative optimization model with the goals of system daily comprehensive cost and peak shaving and valley filling; and an optimization solution and strategy output module, used to solve the model and distribute the operation strategy to the smart terminals in the distribution area through the distribution automation system, for execution by the terminals or edge autonomous correction. This invention quantifies the impact of information node reliability on the dispatchability of energy storage by introducing information impact factors, and utilizes smart terminals in the distribution area to achieve cloud-edge collaborative control, thereby reducing load peak-valley differences and daily comprehensive costs, smoothing load fluctuations, improving the distributed photovoltaic absorption capacity, and enhancing the system's robustness to information failures.
[0049] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.
[0050] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0051] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0052] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.
[0053] Figure 1 A schematic diagram of the distributed energy storage cloud-edge collaborative optimization configuration system architecture based on cyber-physical net capacity provided by the present invention; Figure 2 This is a schematic diagram of the power transfer distribution factor calculation process provided by the present invention; Figure 3 A schematic diagram of the equivalent impedance and capacity of the generator-energy storage-load provided by the present invention; Figure 4This is a schematic diagram of the system information-physical net capacity calculation process provided by the present invention; Figure 5 A schematic diagram of the net capacity of an IEEE 33-node distribution network system provided by this invention; Figure 6 A schematic diagram of the scene generation and reduction process provided by the present invention; Figure 7 This is a schematic diagram of a photovoltaic power output scenario provided by the present invention; Figure 8 This is a schematic diagram of a typical photovoltaic power output scenario provided by the present invention. Detailed Implementation
[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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 some embodiments of the present invention, but not all embodiments.
[0055] In the description of this invention, it should be noted that some processes described in this application specification and drawings include multiple operations that appear in a specific order. However, it should be clearly understood that these operations may be performed in any order or in parallel. Furthermore, various numbers are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0056] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0057] Reference Figure 1 As shown, this invention discloses a distributed energy storage cloud-edge collaborative optimization configuration system based on cyber-physical net capacity. This system is integrated into the distribution network dispatching and operation functional block of the distribution automation system. The system connects to external systems via an information interaction bus. These external systems include a production management system, a higher-level dispatch automation system, a power grid GIS platform, a marketing management information platform, the 95598 system, and intelligent terminals for distribution transformer areas. The system includes the following modules: 1. A data acquisition and preprocessing module, used to acquire physical layer data and information layer data of the distribution network from the external system via an information exchange bus. This module cleans, normalizes, and imputes missing values in the acquired data to form a standardized input dataset.
[0058] 2. The information-physical net capacity calculation and location module is connected to the data acquisition and preprocessing module at its input end. It is used to determine the information-physical net capacity of the distributed energy storage system based on the power transmission capacity and equivalent resistance after the distributed energy storage is connected to the grid, as well as the reliability parameters of the information system, and to determine the optimal access point of the distributed energy storage system based on the determined information-physical net capacity. 3. A multi-scenario collaborative optimization modeling module, whose input end is connected to the information-physical net capacity calculation and site selection module, establishes a multi-scenario collaborative optimization model of source-grid-load-storage based on the optimal access point of distributed energy storage. The objective function of the model includes the daily comprehensive cost of the system and peak shaving and valley filling. 4. The optimization solution and strategy output module is connected to the multi-scenario collaborative optimization modeling module at its input end. It solves the objective function of the source-grid-load-storage multi-scenario collaborative optimization model of the cyber-physical coupling to obtain the operation strategy of the distributed energy storage system. The operation strategy is then sent to the intelligent terminal of the distribution area through the information interaction bus. The intelligent terminal of the distribution area selects to execute the scheduling command or start the local autonomous control strategy based on the local cyber-physical net capacity assessment results.
[0059] The following sections will provide a detailed explanation of each of the above modules.
[0060] In one specific embodiment, the data acquisition and preprocessing module acquires physical layer data and information layer data from the production management system, the higher-level dispatch automation system, the power grid GIS platform, the marketing management information platform, and the 95598 system. The physical layer data includes real-time data, historical data, electrical single-line diagrams, network topology and equipment information, power outage information, and distributed energy storage data. The information layer data includes the historical mean time between failures (MTBF) and mean time to repair (MTBR) of components, as well as the MTBF and MTBR of the smart terminals in the distribution area. The data acquisition frequency can be set manually.
[0061] In one specific embodiment, the cyber-physical net capacity calculation and location module calculates the cyber-physical net capacity of the power grid after the distributed energy storage system is connected, and determines the optimal access location of the distributed energy storage system based on its net capacity.
[0062] The power transfer distribution factor (PTDF) defines the change in branch power caused by changes in the power transmitted between node pairs. Branch power flow and node power flow are defined as follows:
[0063]
[0064] In the formula: , These are the susceptance matrix and the correlation matrix, respectively. , These are branch flow and node flow, respectively.
[0065] The power transfer distribution factor (PTDF) is expressed as: .
[0066] Since the branch power flow and node power flow mentioned above are linearly independent, the inverse of the matrix cannot be calculated. Remove the column containing the reference node, for Matrix. To make and Matrix multiplication also requires removing... The columns and rows of the reference nodes in the matrix are: Matrix. The calculated PTDF is: The reference node is added as a zero column to the PTDF matrix. The PTDF calculation method is described in [link to PTDF calculation process]. Figure 2 Elements in the matrix Represents a node The change in unit power makes the first The power change of each branch corresponds to the node. and nodes The line can be written as and ,element Represents a node The change in unit power makes the first The power change of the branch, the first Branch distribution factor The calculation is as follows:
[0067] The distribution factor describes the power changes in each branch during power injection and removal for each generator-load node pair. The branch power transmission limit capacity for each node pair can be calculated; the branch that first reaches its power transmission capacity limit determines the overall system capacity. Node and nodes between capacity It can be represented as:
[0068] In the formula: branch road Maximum transmission power.
[0069] Furthermore, the net cyber-physical capabilities of the distribution network It can be represented as:
[0070] in,
[0071] In the formula: Nodes represented by equivalent impedance and nodes Electrical distance between; For the system impedance matrix OK List items. , These represent the number of generator nodes and the number of load nodes, respectively. For load nodes The information impact factor characterizes the reliability of the information system that the node depends on in its control capabilities.
[0072] For a power grid containing energy storage, i.e., a distributed energy storage system, calculating net capacity will apply the power transfer capacity and equivalent impedance of each generator-storage pair and energy storage-load pair. Each time the net capacity contribution of a generator-load pair is calculated, there will be two equivalent impedance and capacity values respectively. For example... Figure 3 As shown, to simplify the model, the power flow directly from the generator to the load is not considered. Therefore, it can be regarded as two systems in series. The equivalent impedance of the energy storage system connected to the grid is calculated as follows:
[0073] In the formula: Nodes represented by equivalent impedance and nodes Electrical distance between Nodes represented by equivalent impedance and nodes Electrical distance between For energy storage access nodes After that, node ,node ,node The equivalent impedance between.
[0074] The power limitation of transmission lines can be decomposed into the energy transmitted through the line divided by time. The sum of the charging and discharging times of energy storage will equal the time it takes for electricity to be transferred from the generator to the energy storage, and then from the energy storage to the load. The branch transmission capacity of the generator-load pair after energy storage is integrated into the system can be expressed as:
[0075] In the formula: This indicates the branch transmission capacity of the generator-load pair after distributed energy storage is connected to the grid. For the transmission capacity of the branch of the energy storage access system; , These are the charging and discharging times for energy storage, respectively. Represents a node and nodes between the capacities, Represents a node To the node The capacity between.
[0076] Based on the traditional calculation of net physical capacity, an information impact factor is introduced. Define the net information-physical capacity as:
[0077] In the formula, , These represent the number of generators and the number of load nodes, respectively. , These are the generator and the load node, respectively. , These are the sets of generators and load nodes, respectively. This indicates the branch transmission capacity of the generator-load pair after distributed energy storage is connected to the grid. This represents the equivalent resistance of distributed energy storage after it is connected to the power grid. For load nodes The information impact factor characterizes the reliability of the information system that the node depends on in its control capabilities.
[0078] Among them, information impact factor The calculation method is as follows:
[0079] in, For nodes The set of key information components along the information transmission path on which it depends, including but not limited to intelligent electronic devices (IEDs), communication links, optical line terminals (OLTs), optical network units (ONUs), passive optical splitters (POSs), servers, and smart terminals in the distribution area. For the first The availability of each information element is calculated using the following formula:
[0080] in, For components Mean time between failures (MTBF) For components The average repair time.
[0081] By adjusting and calculating the power transmission capacity and equivalent impedance after energy storage is connected to the grid, and by statistically obtaining the components... The mean time between failures (MTBF) and mean time to repair (MTBR) are used to calculate the system information-physical net capacity after energy storage is connected to each node. By comparing the system information-physical net capacity value at each node, the node with a larger information-physical net capacity is more beneficial to the system, and therefore it can be regarded as the optimal location for energy storage connection.
[0082] In one embodiment, the flowchart for calculating the system's net capacity is as follows: Figure 4 As shown, the specific steps are as follows: Step 1: Input the power distribution network system parameters, and calculate the power transfer distribution factor (PTDF) and distribution factor (PDF) sequentially according to the established mathematical model. ), maximum current carrying capacity of each branch, system admittance matrix ( ), impedance matrix ( The equivalent impedance matrix; and the mean time between failures (MTBF) of the components is obtained through historical statistical data. and average repair time .
[0083] Admittance matrix The calculation formula is:
[0084]
[0085] In the formula, For nodes Admittance; , They are nodes Resistors and inductors; For nodes Self-admittance, For nodes Number of connecting branches.
[0086] Impedance matrix The calculation is as follows:
[0087] The formula for calculating the equivalent impedance matrix is:
[0088] In the formula: For nodes The equivalent impedance; For the impedance matrix OK List.
[0089] Step 2: Based on Step 1, connect each node of the system to the energy storage system in sequence, and calculate the maximum current carrying capacity and equivalent impedance matrix of each branch after the energy storage is connected, as well as the information influence factor.
[0090] Step 3: Calculate and sort the improved cyber-physical net capacity equations of the system to determine the optimal access location for energy storage. Figure 5 The results of calculating net capability on the IEEE (Institute of Electrical and Electronics Engineers) 33-node system are shown in this embodiment of the invention.
[0091] Furthermore, based on the optimal access point of the distributed energy storage system, a multi-scenario collaborative optimization modeling module is established to create a multi-scenario collaborative optimization model for distributed energy storage. The model's objective functions are minimizing the system's daily comprehensive cost and minimizing the sum of squared net load changes at adjacent time points.
[0092] Specifically, the multi-scenario collaborative optimization scheduling model of source-grid-load-storage in this invention is a multi-objective optimization model. In this embodiment, the weight coefficient method is used to transform it into a single-objective optimization model, the expression of which is:
[0093] In the formula: These are the weighting coefficients, and .
[0094] This section employs fuzzy hierarchical analysis and entropy weighting method for comprehensive evaluation, determining the optimal weight coefficients for each sub-objective in the multi-objective function. This approach can both take into account expert experience and reduce the influence of subjective factors on the results, thus allowing for a more reasonable determination of the weight of each evaluation indicator.
[0095] Based on the above, the weights obtained by fuzzy analytic hierarchy process are: The weights obtained by the entropy weight method are Combined weights The calculation formula is as follows:
[0096] In the formula: This represents the number of evaluation indicators.
[0097] ①System daily comprehensive cost: This invention comprehensively considers constraints such as on-load tap changer (OLTC), shunt capacitor bank (SCB), static var compensator (SVC), SOP optimization, and flexible load regulation, and proposes a multi-objective ADN DPV and DES joint optimization configuration model that minimizes annual comprehensive cost and average voltage deviation.
[0098] To better promote the integration of distributed photovoltaic (PV) power into energy storage and improve system operational stability, the daily comprehensive cost of the system considers the investment and operation and maintenance costs of DES and SOP, network loss costs, curtailment costs, and grid purchase costs. The objective function is:
[0099] in, The objective function for the system's daily comprehensive cost; , These are the DES investment costs and operation and maintenance costs, respectively. These are the SOP investment costs and operation and maintenance costs. Main grid electricity purchase cost, Indicates the cost of active power network loss. This indicates the cost of penalties for abandoning light.
[0100] To fully account for the impact of DPV output uncertainty on system operation, the costs under various typical scenarios are calculated using a weighted average. The cost calculation formulas for each part are as follows:
[0101] In the formula: For the scene The probability of occurrence; These are respectively a set of typical scenarios, a set of SOP ports, and a set of lines; For the discount rate, These refer to the investment periods for DES and SOP, respectively. These represent the number of DES and DPV connections, respectively. These are the unit capacity cost of DES, unit power cost of DES, unit operation and maintenance cost of DES, unit capacity cost of SOP, time-of-use electricity price, and unit curtailment cost; The first The installed capacity and power of the DES unit; They are respectively DES charging and discharging power at all times; For port capacity; This is the cost coefficient for SOP operation and maintenance; branch road The resistance; for Branch roads during time periods The current; The first Taiwan's predicted DPV output and actual dispatched output; for Active power flowing from the main grid to this grid during a given time period; The scheduling time is in units.
[0102] Therefore, in this invention, the daily comprehensive cost of the system is calculated by weighting the cost under typical distributed photovoltaic scenarios. Since distributed photovoltaic power generation is greatly affected by environmental factors, its output has a certain degree of uncertainty. Therefore, this invention uses Monte Carlo sampling to sample the probability distribution of photovoltaic output.
[0103] Assuming that the solar irradiance approximately follows a Beta distribution during the scheduling period, the photovoltaic output can be obtained through corresponding transformations.
[0104] Specifically, in this embodiment of the invention, the scene generation and reduction process is as follows: Figure 6 As shown, a distributed photovoltaic (PV) power output mathematical model is first established. Based on this model, multiple PV scenario sets are randomly generated, and a fast priori elimination technique based on probabilistic distance is used to reduce the number of scenarios. The specific steps are as follows: (1) Assuming that the generated photovoltaic power output scenarios approximately follow a Beta distribution during the scheduling period, Monte Carlo sampling is performed on the photovoltaic power output to generate N scenarios. Figure 7 For the 400 distributed photovoltaic power output scenarios generated through Monte Carlo sampling, the following formula is used to calculate the output of each scenario pair. 2-norm That is, Euclidean distance:
[0105] In the formula: It is a 2-norm; For the first The first scenario One value, For the first The first scenario There are several values, where L is the length of the scene sequence; (2) Assume the probability of each scenario occurring Same; Get each pair of scenes in all scenes L2 norm and scenarios probability of occurrence The set of minimum products, denoted by set D, is represented as:
[0106] (3) For the elements in set D, sort them in ascending order according to their probability distance; (4) According to step 3, reduce V scenarios that satisfy the following formula. This is to eliminate scenarios that are not representative and have a low probability of occurrence.
[0107] In the formula: To reduce the number of scenes The probability of occurrence; For the scene The probability of occurrence.
[0108] (5) Determine the scenario with the smallest sum of distances to the reduced scenario probabilities, then update the probability of that scenario, which is equal to the original probability of that scenario plus the probability of the reduced scenario, and finally update the number of elements in set D.
[0109] (6) Repeat steps (2)-(5). If the required number of scenarios is met, stop running the update; otherwise, continue to reduce the number of scenarios. Figure 8 The photovoltaic output scenarios and their probabilities after being reduced by the rapid previous generation elimination technology are shown.
[0110] ② Peak shaving and valley filling: Representing peak shaving and valley filling, this invention uses the minimum sum of squares of net load changes at adjacent time points as the expression:
[0111]
[0112] In the formula: A collection of scenes; For the scene The probability of occurrence; T is the scheduling period; The net load at time t; Let t be the load. , These are the energy storage charging and discharging power, respectively. This indicates the actual dispatch output of distributed photovoltaic power.
[0113] Furthermore, the objective function of the multi-scenario collaborative optimization model is solved.
[0114] The constraints of the multi-scenario collaborative optimization model in this invention include: (1) Second-order cone relaxation of power flow constraints: This invention employs a branch power flow (Distflow) model, which includes active / reactive power balance, node voltage, and branch current expressions, as detailed below:
[0115] In the formula: For time period index, Index of typical scenarios for distributed photovoltaic power generation Indicates in time From the node in the scenario Flow to Node active power, Indicates in time From the node in the scenario Flow to Node reactive power, In order to be in time Nodes in the scene The discharge power of the stored energy, In order to be in time The scene is a section as a point The charging power of the energy storage, , In order to be in time Nodes in the scene Active load and reactive load For nodes Conductivity to ground node Saturation to ground branch road The resistance, branch road Reactance, In order to be in time In this scenario, the active power of the upstream power grid is... In order to be in time In this scenario, the reactive power of the upstream power grid, In order to be in Time node as point Distributed photovoltaic active power, In order to be in Time node as point Distributed photovoltaic reactive power, , In order to be in time The scene is a section as a point The reactive power of the generator In order to be in time The scene is a section as a point SOP injected active power, exist time The scene is a section as a point SOP injects reactive power. for Time Branch The square of the current, ; for time The scene is a section as a point The square of the voltage at that point, ; and SCB and SVC are respectively in time Injecting nodes in the scene reactive power, , For nodes The minimum and maximum voltages, branch road Maximum current, These are the sets of nodes connected to generators, loads, DPV, DES, SOP, SVC, and SCB, respectively. The above formula includes system power flow constraints, voltage constraints, and branch current constraints.
[0116] 2) Photovoltaic output constraints:
[0117] In the formula: The power factor angle for DPV is a constant. For DPV at the node The number of installations is an integer variable. In order to be in Time node as point The maximum active power of distributed photovoltaic power. In order to be in Time node as point Distributed photovoltaic active power.
[0118] 3) OLTC adjustment:
[0119]
[0120] In the formula: For installation The voltage ratio of the branch transformer; To adjust the gear; Adjust the step size in units; This is the maximum value for gear adjustment; These are the upper and lower limits of the transformer tap position, respectively. This is the maximum number of adjustments per day. for The position and status of the transformer tap changer at all times.
[0121] 4) Reactive power compensation device:
[0122]
[0123] In the formula: For nodes The number of SCB groups used; This refers to the capacity of a single capacitor bank. The maximum number of groups that can be included; and Injected into the reactive power compensation device The upper and lower limits of reactive power; and These are the node sets for SCB and SVC to access the distribution network, respectively. For SCB in time Injecting nodes in the scene The reactive power.
[0124] 5) SOP Optimization Operation:
[0125]
[0126]
[0127] In the formula: For the feeder nodes at both ends of the SOP; and They are respectively in SOP at node in scenario s at time s Active power loss of the converter; and The SOP is the loss factor of the converters at both ends. In order to be in time The scene is a section as a point SOP injected active power, exist time The scene is a section as a point SOP injects reactive power. For point Apparent power injected by SOP, For point Apparent power injected by SOP.
[0128] 6) DES grid connection optimization:
[0129]
[0130]
[0131] In the formula: In order to be in time The scene is a section as a point The charging power of the energy storage, In order to be in time Nodes in the scene The discharge power of the stored energy, the charging and discharging process of DES cannot occur simultaneously, so a 0-1 variable is introduced. The state of DES is represented by a value of 1, indicating that the energy storage device is in a discharging state, and vice versa. For nodes Reactive power at DES; For nodes Apparent power of DES installed at the location; For nodes The capacity of the DES installed at the location; and These are the charging and discharging efficiencies of DES, respectively. For nodes The remaining battery power of the DES; This represents the maximum depth of discharge (DES).
[0132] 7) Flexible load constraints: This paper takes reduceable load as the research object, and the constraints include reduceable time period, reduction rate, and reduction duration, which can be expressed as:
[0133] In the formula: To reduce capacity to reduce load, For variables of 0-1, the value is 1 when participating in flexible adjustment; This is the power threshold coefficient that can be reduced; Separate time periods and The reduction ratio coefficient; This is the load reduction rate coefficient; To reduce the total duration of load participation in regulation; These are the lower and upper limits of the time that can be reduced, respectively.
[0134] 8) Constraints on smart terminals in the distribution area: The constraints on smart terminals in the distribution area include communication bandwidth constraints and edge computing latency constraints. Among them: ① Communication bandwidth constraints:
[0135] In the formula , For each of the smart terminals in the distribution area Uplink and downlink communication rates at any given time (unit: Mbps); , These represent the maximum allowed uplink and downlink bandwidth (unit: Mbps).
[0136] ② Edge computing latency constraints:
[0137] In the formula: The total latency (in milliseconds) from receiving instructions from the main station to completing local autonomous control decisions and sending them to the energy storage device. This represents the maximum allowable delay.
[0138] The model constructed above is a mixed integer nonconvex nonlinear model, which is difficult to solve directly. This invention uses second-order cone theory and the big M method to transform the model.
[0139] First, by introducing the theory of second-order cone programming, the SOP optimization constraint is transformed into the following form:
[0140] In the formula, , In order to be in time The scene is a section as a point and nodes SOP injected active power, , In order to be in time The scene is a section as a point and nodes SOP injects reactive power. Indicates the connection node and nodes Apparent power of SOP.
[0141] Secondly, the DES grid-connected operation constraint formula contains 0-1 binary variables, which are nonlinear constraints. Using the Big M method, it is transformed into the following form:
[0142] In the formula: , In order to be in time The section in the scene is a node The energy storage charging and discharging power at the location, , This indicates the charging and discharging efficiency of energy storage. This indicates the rated power capacity of the energy storage. Indicates the energy storage charging and discharging status. It is a constant.
[0143] In this embodiment of the invention, the multi-scenario collaborative optimization strategy of source-grid-load-storage can utilize a commercial solver to achieve the goals of minimizing the daily comprehensive cost of the system and peak shaving and valley filling, with the solution model being a mixed integer second-order cone programming problem, provided that the constraints are met.
[0144] In one embodiment, the proposed model is solved using the YALMIP toolbox called by MATLAB and the Gurobi algorithm package. Specific steps include: Step 1: System Mathematical Model Construction. Input the network topology and branch node information, and construct the system mathematical model considering factors such as distributed energy storage charging and discharging power, charging and discharging status, power balance during the charging and discharging cycle, and state of charge. Specifically, based on the location and capacity of distributed photovoltaic access and the historical statistical data of components, including the smart terminals in the transformer area, obtain the network topology and branch node information; calculate the system information - physical net capacity to determine the location of distributed energy storage access.
[0145] Step 2: Establish a distributed energy storage optimization scheduling model for multiple operating scenarios involving source-grid-load coordination (source-grid-load-storage multi-scenario collaborative optimization model). Taking into account typical distributed photovoltaic scenarios, distributed power sources, and the capacity of energy storage access locations, an objective function for optimizing the energy storage system's scheduling is established based on the system's daily comprehensive cost and peak shaving / valley filling. Simultaneously, source-grid-load-storage collaborative optimization scheduling constraints are established, considering both equipment constraints and power flow constraints. Step 3: In the multi-scenario collaborative optimization model of source-grid-load-storage, SOP constraints and other constraints have strong non-convexity, and the distributed energy storage charging and discharging standard contains binary variables. Using second-order cone relaxation and the Big-M method, the original mathematical model is transformed into a mixed integer second-order cone programming model. Step 4: Solve the power generation-grid-load-storage coordinated control strategy for this distribution network using an optimization solver. Specifically, a mixed-integer optimizer is used to solve the mixed-integer programming problem, which in turn calls an interior-point optimizer to solve the second-order cone programming problem. This yields information on the distributed photovoltaic power output of the distribution network, the optimal energy storage configuration scheme, and the coordinated optimization scheduling strategy.
[0146] Step 5: The smart terminal in the distribution area receives the energy storage operation strategy issued by the main station and simultaneously calculates the information impact factor at the current moment. ,like (threshold) If the value can be set to 0.95, the master station command will be executed; otherwise, the terminal will automatically switch to the local autonomous control mode, regulate the energy storage in the transformer area according to the local measurement information (voltage, frequency, load change rate) and the preset local control strategy (such as constant power charging and discharging, voltage droop control), and send the operating status and local decision results back to the master station.
[0147] In summary, this invention provides a distributed energy storage cloud-edge collaborative optimization configuration system and method, which is integrated into the distribution network scheduling and operation function block of the distribution automation system. By calculating the power transfer distribution factor after energy storage is connected to the distribution network and introducing the information impact factor, the availability of information nodes is incorporated into the net capacity evaluation system. A method for calculating the optimal location of energy storage based on information-physical net capacity is proposed. Unlike traditional net capacity calculations that only consider physical topology and branch parameters, this system can quantify the impact of random failures of the information system on the actual dispatchable capacity of energy storage, avoiding the installation of energy storage at weak information nodes. Based on this, considering the operating characteristics of energy storage and the uncertainty of distributed photovoltaic output, a multi-scenario collaborative optimization scheduling model of source-grid-load-storage is established with the objectives of system daily comprehensive cost and peak shaving and valley filling. This model is then transformed into a mixed-integer second-order cone programming problem using second-order cone relaxation and the Big-M method. The optimal location calculation method based on net capacity fully considers physical attributes such as distribution network branch impedance and branch transmission power limits, as well as the distribution network topology, and can decouple the system topology from other factors. This invention fully utilizes the spatiotemporal shift characteristics of energy storage. Energy storage is charged during periods of low load or low electricity prices and discharged during periods of high load and high electricity prices. This strategy can significantly improve the absorption capacity of distributed photovoltaic power while reducing the load peak-valley difference and daily comprehensive cost. Furthermore, it enhances the robustness against information failure through the edge autonomy mechanism of the intelligent terminal in the distribution area.
[0148] Those skilled in the art will understand that embodiments of the present invention can be provided as systems, methods, electronic devices, or computer program products, etc. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0149] It should be noted that the word "comprising" does not exclude the presence of components or steps not listed in the claims. The words "a" or "an" preceding a component do not exclude the presence of a plurality of such components. This invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer.
[0150] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0151] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A distributed energy storage cloud-edge collaborative optimization configuration system, characterized in that, This system is integrated into the distribution network dispatch and operation extension function block, including: The data acquisition and preprocessing module is used to acquire physical layer data and information layer data of the power distribution network from external systems, and to preprocess the acquired data. The information-physical net capacity calculation and location module is used to calculate the information-physical net capacity of the distributed energy storage system based on preprocessed data, and determine the optimal access point of the distributed energy storage system according to the information-physical net capacity; wherein the calculation of the information-physical net capacity introduces an information influence factor, and the information influence factor incorporates the availability rate of the smart terminals in the distribution area; The multi-scenario collaborative optimization modeling module is used to establish a multi-scenario collaborative optimization model of source-grid-load-storage based on the optimal access point of the distributed energy storage system, with the objectives of system daily comprehensive cost and peak shaving and valley filling. The optimization solution and strategy output module is used to solve the objective function of the source-grid-load-storage multi-scenario collaborative optimization model, obtain the operation strategy of the distributed energy storage system, and send the operation strategy to the intelligent terminal of the distribution area. The intelligent terminal of the distribution area selects to execute the scheduling command or start the local autonomous control based on the evaluation results of the local information impact factor.
2. The system according to claim 1, characterized in that, The data acquisition and preprocessing module acquires data through the following methods: obtaining physical layer data and information layer data from the production management system, dispatch automation system, power grid GIS platform, marketing platform, and intelligent terminals in the distribution area. The physical layer data includes real-time data, historical data, electrical single-line diagrams, network topology, equipment information, power outage information, and distributed energy storage data. The information layer data includes the historical mean time between failures (MTBF) and mean time to repair (MTTR) of components, as well as the MTBF and MTTR of intelligent terminals in the distribution area.
3. The system according to claim 1, characterized in that, In the aforementioned cyber-physical net capacity calculation and location module, the formula for calculating cyber-physical net capacity is as follows: In the formula, Represents information - net physical capacity; , These represent the number of generators and the number of load nodes, respectively. , These are the generator and the load node, respectively. , These are the sets of generators and load nodes, respectively. This indicates the branch transmission capacity of the generator-load pair after the distributed energy storage system is connected to the power grid; This represents the equivalent resistance of a distributed energy storage system after it is connected to the power grid. For load nodes The information impact factor characterizes the reliability of the information system that the node depends on in influencing its regulatory capabilities.
4. The system according to claim 3, characterized in that, In the information-physical net capacity calculation and site selection module, the information impact factor The calculation method is as follows: in, For nodes The set of key information components on the information transmission path on which it depends For the first Availability of each information element.
5. The system according to claim 1, characterized in that, In the multi-scenario collaborative optimization modeling module, the objective function expression of the source-grid-load-storage multi-scenario collaborative optimization model is: in, The total daily cost of the system, This is an indicator for peak shaving and valley filling. These are the weighting coefficients.
6. The system according to claim 5, characterized in that, The objective function for the system's daily comprehensive cost is: in, , These are the investment costs and operation and maintenance costs of distributed energy storage, respectively. These are the investment costs and operation and maintenance costs of intelligent soft switches, respectively. Main grid electricity purchase cost, Indicates the cost of active power network loss. This indicates the cost of penalties for abandoning light.
7. The system according to claim 5, characterized in that, The objective function for peak shaving and valley filling is: in, It is a collection of multiple scenarios; For the scene The probability of occurrence; T is the scheduling period; The net load at time t; The net load at time t-1.
8. The system according to claim 1, characterized in that, The constraints of the multi-scenario collaborative optimization model of source-grid-load-storage include: second-order cone relaxation power flow constraint, photovoltaic power output constraint, OLTC regulation, reactive power compensation device, SOP optimized operation and DES grid connection optimization, as well as communication bandwidth constraint and edge computing latency constraint of intelligent terminals in the distribution area.
9. The system according to claim 8, characterized in that, In the optimization solution and strategy output module, the model is transformed into a mixed-integer second-order cone programming problem by using second-order cone relaxation and the Big-M method.
10. A distributed energy storage cloud-edge collaborative optimization configuration method, characterized in that, This method applies the system described in any one of claims 1–9 to achieve distributed energy storage cloud-edge collaborative optimization configuration, and the method includes the following steps: S1. Obtain physical layer data and information layer data of the distribution network from external systems, and preprocess the obtained data; S2. Based on the preprocessed data, calculate the net information-physical capacity of the distributed energy storage system, and determine the optimal access point of the distributed energy storage system according to the net information-physical capacity; wherein the calculation of the net information-physical capacity introduces an information impact factor, and the information impact factor incorporates the availability rate of the smart terminals in the distribution area; S3. Based on the optimal access point of the distributed energy storage system, establish a multi-scenario collaborative optimization model of source-grid-load-storage with the goal of system daily comprehensive cost and peak shaving and valley filling; S4. Solve the objective function of the source-grid-load-storage multi-scenario collaborative optimization model to obtain the operation strategy of the distributed energy storage system, and send the operation strategy to the intelligent terminal of the distribution area. The intelligent terminal of the distribution area selects to execute the scheduling command or start the local autonomous control based on the evaluation result of the local information impact factor.
