A power storage regulation method, system and device for a smart grid

Abstract

The application provides a power storage regulation and control method, system and device of a smart power grid, relates to the technical field of power grid regulation and control, constructs a power grid admittance matrix of a smart power grid topology based on a pre-management area, introduces a fuzzy controller based on an energy storage participation factor, performs network connection and receives a power storage regulation and control task, determines a set of energy storage factors based on power grid data and the power grid admittance matrix, performs equivalent reconstruction on the smart power grid topology, executes power storage regulation and control decisions under consensus constraints, determines a power storage regulation scheme, and executes the issuance of a decomposed scheme to drive regulation and control management. The application is used for solving the technical problem that the existing technology cannot achieve comprehensive and highly adaptive precise regulation and control driving, which limits the adaptive collaborative regulation and control of the power storage of the smart power grid in complex scenarios, and achieves flexible and highly adaptive comprehensive precise regulation and control of the smart power grid in multiple scenarios.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid regulation, and in particular to a power storage regulation method, system and device for a smart power grid. BACKGROUND

[0002] With large-scale access of new energy power generation, the smart power grid faces challenges such as frequent power fluctuations, complex load regulation, and dynamic imbalance between power supply and demand. Energy storage technology is used as a regulating means to support voltage, balance frequency, and optimize power flow.

[0003] Existing power storage regulation methods are mostly based on static scheduling models or single control variables for regulation, mainly relying on centralized scheduling instructions or control rules based on single-factor triggering, which is difficult to adapt to complex environments with frequent changes in power grid topology and dynamic changes in node state, and is difficult to achieve rapid response and regulation to local disturbances. In addition, the energy storage decision scheduling in the existing method is one-sided and cannot reasonably balance the demands of multiple parties, resulting in unstable regulation effect or unbalanced resource allocation.

[0004] In summary, the existing energy storage regulation technology lacks a complete system that is flexible and highly adaptable, making it difficult to achieve comprehensive and highly adaptive precise regulation driving, which limits the adaptive and collaborative regulation of power storage in the smart power grid in complex scenarios. SUMMARY

[0005] The present application provides a power storage regulation method, system and device for a smart power grid, which is used to solve the technical problem that it is difficult to achieve comprehensive and highly adaptive precise regulation driving in the prior art, which limits the adaptive and collaborative regulation of power storage in the smart power grid in complex scenarios.

[0006] In view of the above problems, the present application provides a power storage regulation method, system and device for a smart power grid.

[0007] In a first aspect, the application provides a power storage regulation method for a smart grid, the method comprising: constructing a grid admittance matrix based on a smart grid topology of a pre-management area, wherein the grid admittance matrix is dynamically updated; introducing a fuzzy controller based on a storage participation factor, connecting to the grid and receiving a power storage regulation task, determining a storage factor set based on grid data and the grid admittance matrix, wherein the fuzzy controller takes an SOC state, a price signal, and a grid frequency deviation as inputs, takes grid node voltage potential and line current vector potential as intermediate quantities, and takes an optimal control item as an output; according to the storage factor set, performing equivalent reconstruction on the smart grid topology, executing a power storage regulation decision under consensus constraint, and determining a power storage regulation scheme, wherein the consensus constraint is performed by a three-party consensus among a storage unit, a power distribution terminal, and a dispatch center; decomposing the power storage regulation scheme based on an access component belonging to the smart grid, and executing a decomposed scheme by the smart grid to drive regulation and management.

[0008] In a second aspect, the application provides a power storage regulation system for a smart grid, the system comprising: a matrix construction unit configured to construct a grid admittance matrix based on a smart grid topology of a pre-management area, wherein the grid admittance matrix is dynamically updated; a factor determination unit configured to introduce a fuzzy controller based on a storage participation factor, connect to the grid and receive a power storage regulation task, and determine a storage factor set based on grid data and the grid admittance matrix, wherein the fuzzy controller takes an SOC state, a price signal, and a grid frequency deviation as inputs, takes grid node voltage potential and line current vector potential as intermediate quantities, and takes an optimal control item as an output; a reconstruction decision unit configured to perform equivalent reconstruction on the smart grid topology according to the storage factor set, execute a power storage regulation decision under consensus constraint, and determine a power storage regulation scheme, wherein the consensus constraint is performed by a three-party consensus among a storage unit, a power distribution terminal, and a dispatch center; and a storage regulation unit configured to decompose the power storage regulation scheme based on an access component belonging to the smart grid, and execute a decomposed scheme by the smart grid to drive regulation and management.

[0009] In a third aspect, the application further provides an electronic device comprising: a memory configured to store executable instructions; and a processor configured to execute the executable instructions stored in the memory to implement a power storage regulation method for a smart grid.

[0010] One or more technical solutions provided in the application have at least the following technical effects or advantages:

[0011] This application provides a method for power storage regulation in a smart grid. It constructs a grid admittance matrix based on the smart grid topology of a pre-managed area, wherein the grid admittance matrix is ​​dynamically updated. A fuzzy controller based on energy storage participation factors is introduced. This controller connects to the grid and receives power storage regulation tasks. Based on grid data and the grid admittance matrix, it determines a set of energy storage factors. The fuzzy controller takes SOC status, electricity price signal, and grid frequency deviation as inputs, grid node voltage potential and line current vector potential as intermediate quantities, and the optimal control term as output. Based on the energy storage factor set, the smart grid topology is equivalently reconstructed, and power storage regulation decisions are executed under consensus constraints to determine a power storage regulation scheme. Consensus constraints are established through a three-way consensus among energy storage units, distribution terminals, and the dispatch center. The power storage regulation scheme is decomposed based on the smart grid's access component affiliation, and the smart grid executes the decomposed scheme's distribution-driven regulation management. This technology addresses the technical challenge of achieving comprehensive and highly adaptable precise control in existing technologies, which limits the adaptive and coordinated control of power storage in complex scenarios in smart grids. It enables flexible and highly adaptable comprehensive and precise control of smart grids across diverse scenarios. Attached Figure Description

[0012] Figure 1 This application provides a schematic flowchart of a smart grid power storage and control method.

[0013] Figure 2 This application provides a schematic diagram of the structure of a smart grid power storage and control system.

[0014] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0015] Explanation of reference numerals in the attached figures: Matrix construction unit 11, Factor determination unit 12, Reconstruction decision unit 13, Energy storage and regulation unit 14, Processor 21, Memory 22, Input device 23, Output device 24. Detailed Implementation

[0016] This application provides a method, system, and device for power storage regulation in smart grids, which addresses the technical problem in existing technologies where it is difficult to achieve comprehensive and highly adaptable precise regulation, resulting in limited adaptive and coordinated regulation of power storage in smart grids in complex scenarios.

[0017] Example 1: As Figure 1 As shown, this application provides a power storage and control method for a smart grid, the method comprising:

[0018] S1: constructing a power grid admittance matrix based on a pre-management area-based smart grid topology, wherein the power grid admittance matrix is dynamically updated.

[0019] In the embodiment, first, a topology structure of a smart grid is constructed based on a pre-management area, which is a logical division unit in the operation of the smart grid, and usually corresponds to a sub-area that is geographically continuous or has a correlation in scheduling function, for example, a city power distribution grid area, an industrial park or a regional load cluster.

[0020] Specifically, the topology structure of the smart grid is composed of a plurality of grid nodes and connecting branches between the nodes, the nodes can include substations, energy storage units, load access points or distributed power access points, and the connecting branches represent power transmission lines or transformer interfaces between nodes. The above topology structure serves as the basis for constructing the power grid admittance matrix, and the node numbering and branch connection relationship serve as the prerequisite for subsequent matrix element establishment.

[0021] Next, according to the smart grid topology, the self-admittance elements of each topology node are calculated one by one, the self-admittance element refers to the sum of the admittance values of all branches directly connected to a certain grid node, wherein the admittance element is a complex number representing the impedance degree of a power component to alternating current in a power system. The self-admittance data reflects the electrical coupling strength of the node location, and is used to construct the diagonal elements of the power grid admittance matrix.

[0022] Synchronously, for any two topology nodes with a connection relationship, the mutual admittance value is determined, wherein the mutual admittance value is the negative of the admittance value of the connecting branch, which reflects the coupling directionality and relative influence of current between the two nodes. Optionally, if there is no direct branch connection between the two nodes, the corresponding mutual admittance value is zero.

[0023] In summary, the self-admittance value is used to fill the diagonal elements of the power grid admittance matrix, and the mutual admittance value is used to fill the non-diagonal elements, to construct a complete power grid admittance matrix. The power grid admittance matrix is a complex symmetric matrix.

[0024] In this application, the power grid admittance matrix is used to provide energy storage control criteria and voltage and current state evaluation basis based on electrical coupling relationship.

[0025] Further, considering that the topology structure of each node in the smart grid has dynamic variability, for example, the addition, deletion of nodes, the switching of branches and the change of energy storage unit access position may all cause changes in admittance relationship, therefore the power grid admittance matrix needs to have dynamic updating capability.

[0026] Specifically, when a topology change event is monitored, such as the addition of a connection line, the self-admittance and mutual-admittance values of the related nodes are updated in real time, and the corresponding matrix elements in the power grid admittance matrix are updated at the corresponding positions, to ensure the accuracy and real-time performance of subsequent energy storage factor evaluation and control scheme development.

[0027] In summary, the power grid admittance matrix constructed and dynamically maintained can truly reflect the current operation structure and electrical state of the power grid, serving as a basic support structure for the control logic.

[0028] Further, the power grid admittance matrix based on the smart grid topology of the pre-management area is constructed, and the step S1 of the present application includes:

[0029] For the smart grid topology of the pre-management area, the self-admittance element is determined by summing the admittances of the nodes and the connected power grid branches. The mutual-admittance element is determined by the negative values of the admittances of the connecting power grid branches between any two topology nodes. The power grid admittance matrix is constructed with the self-admittance elements as the diagonal elements and the mutual-admittance elements as the non-diagonal elements, wherein the power grid admittance matrix is updated with the smart grid topology.

[0030] In this embodiment, for the smart grid topology of the pre-management area, the topology nodes are first identified one by one and the connection relationship with the adjacent power grid branches is constructed. The pre-management area refers to a sub-area with independent control authority and specific electrical characteristics, which is pre-defined according to the smart grid control planning, for local network analysis and control strategy deployment. The topology nodes include bus nodes, transformer nodes, energy storage device connection nodes, load access nodes or distributed power access points in the area, each node having a unique topology identifier.

[0031] Specifically, based on the identification of the nodes, each topology node is analyzed in turn, the admittance values of all the directly connected power grid branches are counted, and the self-admittance element of the node is determined by summing the admittances of the node and all the connected branches.

[0032] For example, for a node A connected with three branches with corresponding admittances Y1, Y2 and Y3, the self-admittance of the node A is Y1+Y2+Y3, which represents the comprehensive electrical coupling strength of the node A to its local power grid.

[0033] Synchronously, for any two topology nodes with a connection relationship, the negative value of the admittance of the branch between them is calculated, and this value is taken as the mutual-admittance element. Mutual-admittance refers to the admittance coupling relationship between two nodes due to the connection of the branch, representing the alternating current transmission capability between them and embodying the electrical coupling contribution of the position other than the diagonal line in the power grid admittance matrix.

[0034] For example, if there is a branch connection between node A and node B with an admittance of AB, in the admittance matrix, the element corresponding to node A in the row and node B in the column is -Y, where Y is the admittance value; similarly, the element corresponding to node B in the row and node A in the column is also -Y, thereby satisfying the symmetry of the admittance matrix. When there is no branch connection between two nodes, the mutual admittance is recorded as 0, indicating no electrical coupling.

[0035] Subsequently, after obtaining all the self-admittance elements and mutual admittance elements, the construction of the entire power grid admittance matrix is completed with the self-admittance as the diagonal elements and the mutual admittance as the non-diagonal elements of the admittance matrix.

[0036] In one specific embodiment, the power grid admittance matrix is an N x N complex symmetric matrix, where N is the total number of topological nodes, and its structure reflects the static coupling relationship and branch configuration of the entire pre-management area power grid, which is the key basis for power flow analysis, state estimation, stability control, and subsequent regulation and optimization.

[0037] In the embodiments of the present application, the admittance matrix is not only used for electrical state evaluation, but also serves as an important basis for fuzzy controller input and energy storage factor evaluation.

[0038] Further preferably, in order to adapt to the dynamic changes of the operating state of the smart grid, such as node addition, branch fault, topology reconfiguration, or dynamic access of energy storage units, the power grid admittance matrix has dynamic updating capability.

[0039] Specifically, when a change in the topological structure is detected, an automatic updating mechanism is triggered to recalculate the self-admittance and mutual admittance elements of the affected nodes and perform local updating in the admittance matrix, ensuring that the matrix always reflects the current true structure of the smart grid.

[0040] Preferably, the dynamic updating mechanism ensures that the regulation based on the admittance matrix can adapt to the changes in the power grid state in real time, improving the real-time performance and accuracy of energy storage participation control and enhancing the stability and robustness of the overall dispatching system.

[0041] S2: Introduce a fuzzy controller based on the energy storage participation factor, connect to the grid and receive power storage regulation tasks, determine the energy storage factor set based on the power grid data and the power grid admittance matrix, wherein the fuzzy controller takes the SOC state, the electricity price signal, and the power grid frequency deviation as inputs, takes the power grid node voltage potential and the line current vector potential as intermediate quantities, and takes the optimal control item as output.

[0042] In the embodiments, a fuzzy controller based on the energy storage participation factor is introduced to improve the flexibility and adaptability in the process of power storage regulation.

[0043] The fuzzy controller proposed in the application is a control unit that makes control decisions using fuzzy logic reasoning mechanism, and can generate control output meeting target optimization conditions according to experience rules and data correlation under the condition of incomplete parameter information or fuzzy system state.

[0044] In the present scheme, the fuzzy controller takes the energy storage participation factor as the core for construction, which represents the adaptability, response ability and load bearing potential of the energy storage unit to the regulation task under the current power grid structure and state, and is the core decision basis for intelligent scheduling of energy storage resources.

[0045] Specifically, the fuzzy controller realizes its embedded deployment in the smart grid environment by performing network connection. Optionally, the network connection mode can be communication access at the physical layer or control coordination access at the logical layer, so that the controller can obtain power grid data and admittance matrix information in real time and interact with the dispatch center.

[0046] After completing the access, the fuzzy controller receives the power storage regulation task, which can be sourced from the active inspection module of the smart grid or the instruction issued by the dispatch platform, and the content includes parameters such as the regulation target to be responded, the time window, the power regulation range, and the regional restriction.

[0047] In the specific implementation, during the task execution process, the fuzzy controller takes the power grid data and the power grid admittance matrix as the basis for analysis. The power grid data includes but is not limited to the node real-time state of charge (SOC), the dynamic price signal, and the offset of the power grid operating frequency, which are obtained by integrating the original power grid data and directly reflect the available capacity, responsiveness, and frequency support ability of the energy storage system to the power system.

[0048] Among them, the SOC state represents the current charge level of the energy storage unit, which is the core index for measuring its charge and discharge capacity; the price signal reflects the economic incentive conditions of the current electricity market, which is an important parameter for cost optimization; and the power grid frequency deviation is used to reflect the stability fluctuation of the power system, which is an important index for adjusting the response timing and priority.

[0049] Further, in one preferred embodiment of the present application, after receiving the above inputs, the fuzzy controller first performs fuzzy processing on them and converts them into fuzzy language variables, for example, the SOC is divided into low, medium and high states, and the frequency deviation is divided into negative, zero and positive levels.

[0050] Subsequently, the grid node voltage potential and the line current vector potential are introduced as intermediate analysis quantities, which respectively represent the amplitude and phase changes of the node voltage and the directionality and strength of the current in the line, and are the bridge quantities for building the coupling relationship between the energy storage factor and the grid. The voltage potential and the current vector potential are calculated by the grid admittance matrix and the node injected current, which reflects the electrical influence and adjustment response range of the possible access point of the energy storage unit.

[0051] In a feasible embodiment, the grid node voltage potential and the line current vector potential are obtained as follows: first, load the grid admittance matrix of the current smart grid, which is a complex matrix obtained according to the connection relationship between each node in the grid topology and the branch parameters such as resistance, reactance, etc.; then, obtain the injected current value of each node, which is uploaded in real time by the energy storage device, load or distributed power supply, and its value reflects the complex current injected or drawn by each node to the grid. Then, by using linear algebra methods such as LU decomposition or Gauss-Seidel iteration method, the grid admittance matrix and the injected current vector are solved to obtain the complex voltage value of each node, i.e. the voltage potential. Further, according to the voltage potential difference between any two nodes and the admittance value of the corresponding branch, the amplitude and phase of the branch current are calculated according to Ohm's law to obtain the current vector potential.

[0052] Finally, the fuzzy controller analyzes the input and intermediate quantities based on the fuzzy rule base and the reasoning mechanism, and outputs the optimal control item. The control item includes the adjustment power size, adjustment time range and participation priority of the energy storage unit, and can be further used to generate the energy storage factor set. The energy storage factor set is a comprehensive quantitative representation of the availability, responsiveness and coordination of the energy storage unit in the current regulation task, which provides a key input basis for subsequent grid reconstruction and regulation decision.

[0053] In this application, by introducing the fuzzy controller, fine deployment and dynamic response of the energy storage resource can be realized under uncertain parameters, and the intelligent level of the regulation system is enhanced.

[0054] Further, the fuzzy controller based on the energy storage participation factor is introduced, and step S2 of the application comprises:

[0055] A first threshold is set according to the task constraint, wherein the first threshold performs filtering of the grid data and the grid admittance matrix; a second threshold is set according to the analysis of the grid node voltage potential and the line current vector potential; a third threshold is set according to the positioning of the optimal control item based on the regulation database, wherein the third threshold performs filtering of the energy storage participation item with dynamic weight as the constraint; the first threshold, the second threshold and the third threshold are cascaded, and the fuzzy controller is generated by supervised training to convergence.

[0056] In the embodiment, to realize the accurate response of the fuzzy controller to the energy storage regulation task in a complex power grid scenario, a three-level control threshold mechanism is designed, and the controller is trained to complete parameter convergence and reasoning optimization within the rule system, so as to build a fuzzy controller meeting the task requirements and system state.

[0057] Optionally, the control threshold comprises a first threshold, a second threshold and a third threshold in sequence, each of which acts on an information filtering process in different dimensions and presents a cascading relationship in the control structure, forming a control logic framework from coarse to fine.

[0058] Firstly, the first threshold is set with the power storage regulation task as an external task constraint. The regulation task usually contains target node range, power demand boundary, response time window and other restrictive conditions, and the first threshold performs preliminary filtering on the input power grid data and power grid admittance matrix according to the task constraint.

[0059] Specifically, the underlying logic includes: identifying the node area related to the task, extracting the voltage, current, frequency deviation of the nodes in the task area and the corresponding admittance submatrix, so as to narrow the control analysis range and improve the calculation efficiency and reasoning pertinence.

[0060] For example, if the regulation task requires priority scheduling of energy storage units in region A, the first threshold limits the power grid data to the real-time operating information of the nodes in region A, and extracts the local admittance matrix related to region A.

[0061] Secondly, based on the first threshold screening result, the extracted power grid node data is further analyzed, and the second threshold is set with the node voltage potential and line current vector potential as the core variables. The voltage potential is used to reflect the voltage amplitude and phase state of the node in the power grid, and the current vector potential represents the flow direction characteristics and amplitude intensity of the current in the branch, which together reveal the influence of energy storage access on the local electrical structure.

[0062] Among them, the second threshold will set criteria according to these electrical states to exclude candidate nodes that are in extreme voltage or current state or do not have stable access conditions in the current period, ensuring decision analysis within the safety boundary.

[0063] Then, a third threshold is set, which is oriented to the regulation task and performs screening optimization on the optimal control item in the regulation database. The regulation database stores the participation response data of different energy storage units under different network states in historical regulation scenarios, including specific power regulation elements such as response rate, stability coefficient, access risk level, etc. The third threshold identifies control items with high similarity by comparing the current input conditions with the known patterns in the database, and excludes low correlation energy storage participation items. At the same time, a dynamic weight constraint mechanism is introduced, and according to the key target of the current task, such as frequency support priority, electricity price income priority, stability priority, etc., a variable weight is applied to the participation item to adjust its priority order and improve the target adaptability of the screening result.

[0064] In summary, the above three thresholds are executed in the order of input data screening→ electrical state verification→ control item matching, forming the pre-screening structure of the fuzzy controller. To realize the optimal fusion of the logic of each threshold in the fuzzy controller, convergence training is performed through a supervised training mechanism.

[0065] Specifically, the existing historical data set of the power grid is used as a training sample, and the task input, control target and final response effect are used as training references. By continuously adjusting the fuzzy rule boundary and threshold parameters, the control output is optimized in the training iteration, so that the controller can finally give a control response result that meets the actual operation target when facing new inputs. The control structure formed after training convergence is the fuzzy controller in the present application, which has multi-threshold fusion reasoning ability and task adaptability control ability, and can be widely used in energy storage participation type regulation tasks in complex smart grid environments.

[0066] Further, after receiving the power storage regulation task, step S2 of the present application includes:

[0067] obtaining a power storage regulation task, wherein the power storage regulation task is an active task or a passive task, the active task is generated by self-checking of the smart grid, and the passive task is uploaded by the server; for the smart grid topology, mining energy storage participation items of parameter power storage regulation to form a regulation database, wherein the energy storage participation item is identified by a topology node code, and the regulation database is built-in in the fuzzy controller; by interpreting the power storage regulation task, the regulation database is framed to determine the task regulation database.

[0068] In the present embodiment, in order to realize the orderly calling of energy storage resources in a complex power grid scenario, first, the power storage regulation task is obtained as an external trigger signal for the fuzzy controller to execute judgment and reasoning.

[0069] In the specific implementation scenario of the present application, the power storage regulation and control task can be divided into active task and passive task according to its source, wherein the active task refers to the regulation and control demand automatically triggered by the smart grid through self-checking and self-diagnosis mechanism in the operation process, for example, the automatically generated charge and discharge adjustment instruction when detecting local load mutation, frequent voltage fluctuation, frequency deviation from normal threshold or high idle rate of energy storage unit; and the passive task refers to the uploaded task from the superior dispatching center, cloud server or user terminal, for example, the power regulation instruction, demand response task or economic optimization operation instruction issued through the dispatching platform, and the two together constitute the smart grid regulation and control task entry.

[0070] Further, after receiving the regulation and control task, the parameterized energy storage participation item related to the task is mined according to the current smart grid topology. The energy storage participation item is a structured entry after modeling the response capability of the energy storage unit at each topology node under the specific electrical state, control boundary and task target, including node number, access power range, response time, historical stability evaluation index, state of charge (SOC) interval, regulation priority and other information.

[0071] Preferably, in order to realize efficient retrieval and rapid adaptation, the energy storage participation item is accompanied by a unique topology node code, which can be directly mapped with the node identifier in the grid admittance matrix structure, so as to accurately locate the target node and its state in the regulation and control execution.

[0072] In the present application, all the above participation items constitute a structured regulation and control database, which is integrated in the fuzzy controller in an embedded manner as the core knowledge support set for the execution of inference judgment and output control item.

[0073] Further, in order to improve the response efficiency and task adaptability of the controller, the regulation and control database is subjected to targeted screening after receiving the task, forming a data subset strongly associated with the current task, referred to as task regulation and control database.

[0074] One screening process provided by the present application includes identifying the target area or node range in the regulation and control task, identifying the adjustment target such as frequency support, load peak clipping, electricity price arbitrage, and identifying the time requirement and energy boundary requirement of the task. Based on these task elements, the regulation and control database is subjected to optional framing, that is, selecting those energy storage participation items in the database that meet the conditions of spatial position, electrical capacity and response time sequence, and eliminating the items that are not related to the current task target or the resources are not available, so as to quickly build an efficient response set serving the current task.

[0075] Further, the task regulation database will be executed as a fuzzy controller after the subsequent multi-threshold reasoning, energy storage factor screening and control item generation of the exclusive data range, so that the fuzzy control process is more targeted and real-time, and the intelligent level and system adaptation ability of the overall energy storage regulation are improved.

[0076] Further, the fuzzy controller is generated by supervised training to convergence, and step S2 of the present application includes:

[0077] The energy storage controller is determined by supervised training to convergence; the input-multiple-level threshold output is the first decoupling target, the reasoning logic of the input-output of the supervised training is the second decoupling target, the energy storage controller is decoupled, the fuzzy controller and the logic reasoner are determined; the connection between the fuzzy controller and the data block stored by the smart grid and the grid admittance matrix is established, and the logic reasoner is used as the edge branch of the fuzzy controller, wherein the logic reasoner is optional and is used for logical verification of the output of the fuzzy controller.

[0078] In the present embodiment, in order to realize accurate modeling and rule optimization of the fuzzy controller, first, a supervised training model is constructed based on the grid operation data set, historical regulation tasks and energy storage response results, and through the training process, the controller forms a stable input-output mapping relationship in multiple tasks and multiple state scenarios.

[0079] The process is specifically: in the existing regulation data set, the input feature vector is extracted, including the SOC state of the energy storage unit, the electricity price signal, the grid frequency deviation and other basic inputs, and the grid node voltage potential, line current vector potential and other intermediate variables; the historical response results (such as the adjustment power size, response time window, task completion efficiency, etc.) are used as supervision labels, input into the training framework, and through repeated iterative training, the fuzzy rules, parameter boundaries, etc. are optimized, the model is converged and stabilized, and finally the structure and reasoning ability of the energy storage controller are determined.

[0080] Subsequently, the energy storage controller after completing the preliminary training still contains a relatively complex internal logical coupling structure, which is not conducive to subsequent modularization calling and reasoning process optimization, so that its real-time response ability is limited. Therefore, the control decoupling strategy is introduced in the present embodiment, and the controller is structurally split with two decoupling targets.

[0081] Specifically, the first decoupling target is an input-multilevel threshold output path, that is, the input features (such as SOC, electricity price, frequency deviation) are structured and split through a multi-level control logic output path including a first threshold (data screening), a second threshold (electrical state analysis), a third threshold (historical response screening), and the like, so that each threshold control logic module is independently controllable and the parameters are adjustable; the second decoupling target is a supervised reasoning logic of input and output, that is, a fuzzy reasoning rule ontology is abstracted from the training model, the semantic interpretation rules and logical judgment paths related to the original data and intermediate variables are stripped, and a clear reasoning chain structure is formed.

[0082] Based on the above decoupling, the energy storage controller is split into two independent sub-modules: a fuzzy controller and a logic reasoner. The fuzzy controller is mainly responsible for the membership degree processing of fuzzy variables, the application of multi-threshold fuzzy rules, and the output of the best control item, constituting the main control subject; the logic reasoner serves as an auxiliary, and realizes constraint verification, boundary rationality checking, and rule consistency verification on the output of the fuzzy controller.

[0083] In the embodiment of the present application, through decoupling, the fuzzy controller eliminates the internal complex logic information, that is, analyzes in a gray box state. At the same time, the logic reasoner can be triggered to verify the logic of the output of the fuzzy controller, so as to balance efficiency and accuracy.

[0084] Further, in order to guarantee the deployability and data adaptation ability of the fuzzy controller in the smart grid environment, the embodiment further establishes a connection relationship between the fuzzy controller and the data storage block in the smart grid system. The data block includes a grid admittance matrix, a node state cache area, a storage response state set, and the like, and is a structured storage area of real-time operation data of the smart grid. The fuzzy controller can realize real-time synchronization of the grid state, the admittance change, and the boundary condition of the energy storage unit by calling the data block, thereby supporting the timeliness and accuracy of reasoning calculation.

[0085] In addition, the logic reasoner is configured to be selectively triggered, that is, when there is a regulation and control exception, a control conflict, or the fuzzy output exceeds the expected control boundary, the logic reasoner can be activated to perform a reasoning chain verification operation, to perform logic consistency checking and dynamic correction on the control instruction generated by the fuzzy controller, and when necessary, feedback to constrain the fuzzy controller for secondary reasoning, thereby constructing a closed-loop regulation and control mechanism of fuzzy output + logic verification, and improving the robustness and fault tolerance of the control result.

[0086] In summary, the structural separation design not only enhances the maintainability of the controller modules, but also improves the flexibility and safety of the controller in complex dynamic grid scenarios.

[0087] Further, based on the grid data and the grid admittance matrix, the energy storage factor set is determined, and the step S2 includes:

[0088] With the acquisition of the power storage regulation task, grid data is retrieved, wherein the grid data determines the structured SOC state, electricity price signal, and grid frequency deviation via a first threshold screening structure. The grid admittance matrix is synchronized, wherein the grid admittance matrix determines the task correlation matrix via the first threshold screening. For the grid data and the grid admittance matrix, a second threshold is triggered to determine the intermediate quantity matrix. With the task regulation database as the regulation range, a third threshold is triggered to perform dynamic weight assignment and control item screening based on the grid data, the grid admittance matrix, and the intermediate quantity matrix to determine the energy storage factor set, wherein each energy storage factor is identified with an adjustable interval.

[0089] In the present embodiment, for the acquired power storage regulation task, a data preparation process matched with the task is first performed to ensure that the fuzzy controller can analyze the energy storage participation and generate the control items based on complete and valid data input.

[0090] Specifically, after receiving the regulation task, the grid data at the current time is immediately retrieved, which includes but is not limited to the state parameters of the energy storage units at each node, the electricity market price information, the system frequency state, etc., as the source of the controller input variables.

[0091] After retrieval, the grid data needs to be screened via the first threshold. The first threshold is the preliminary screening logic driven by the task constraints, which removes the irrelevant information in the grid data according to the target node range, time window, voltage / frequency regulation target, etc. in the task, and only retains the data fields strongly related to the task area and target, and extracts the key input variables. The screening result forms three types of standardized input data: one is the SOC state, i.e. the current state of charge of the energy storage unit, the second is the electricity price signal, i.e. the electricity market price within the task period, and the third is the grid frequency deviation, i.e. the offset between the current system frequency and the reference frequency (such as 50 Hz).

[0092] Optionally, the data structure standard of the screening result can be determined based on the demand debugging, including but not limited to the above-mentioned content.

[0093] At the same time, the current grid admittance matrix is called synchronously and also screened by the first threshold to extract the sub-matrix related to the task target to form the task correlation matrix. The task correlation matrix retains the electrical coupling relationship between the topological nodes involved in the regulation task, and is the key data basis for judging the feasibility of the energy storage unit access, the electrical response capability, and the action range.

[0094] Subsequently, a second threshold is triggered to enter a second stage of the process. The second threshold is centered on electrical analysis, which jointly analyzes the input grid data and admittance information to construct an intermediate quantity array containing multiple node states and branch response characteristics. The intermediate quantities mainly include voltage potential (node voltage amplitude and phase relationship) and current vector potential (branch current directionality and intensity characteristics), which can be used to reflect node control sensitivity, access response strength, and system disturbance feedback path, providing an intermediate layer basis for the controller to determine whether the energy storage access behavior has electrical significance and execution space.

[0095] Further, the task control database is used as the current control range, and a third threshold is triggered to perform control item screening and weight assignment process. The role of the third threshold is to assign different energy storage items with task target-oriented weight coefficients in the screened energy storage participation items, combining task input variables (SOC, electricity price, frequency), electrical response variables (voltage potential, current vector potential), and historical control records.

[0096] Specifically, the dynamic weight mechanism adjusts the constraints according to the task target priority (such as cost minimization, fastest response, and optimal stability), allowing the controller to select the optimal energy storage combination in the multi-objective trade-off of adjusting power size, response time window, and scheduling priority.

[0097] Finally, under the three-level threshold linkage mechanism, a structured energy storage factor set is determined, where each energy storage factor is identified with its adjustable interval, i.e., maximum dischargeable power, maximum charge capacity, SOC upper and lower limit boundary, and node physical connection constraints, providing quantitative input guarantee for subsequent smart grid topology reconstruction and control strategy output.

[0098] In summary, in the environment of multiple input data sources and complex structure, the precise quantification and task adaptability evaluation of energy storage unit participation capability are completed, and the simultaneous improvement of control efficiency and response accuracy is achieved.

[0099] S3: According to the energy storage factor set, the smart grid topology is equivalently reconstructed, and the power storage control decision under consensus constraint is executed to determine the power storage adjustment scheme, wherein the three-party consensus of energy storage unit, distribution terminal, and dispatch center is used for consensus constraint.

[0100] In this embodiment, based on the previously determined energy storage factor set, the original topology structure of the smart grid is further equivalently reconstructed to adapt to the adjustment target and network state of the current power storage control task.

[0101] The energy storage factor set, which clearly defines the adjustable power range, response capability, electrical connection node, and current state of each energy storage unit, is a crucial foundation for constructing the control strategy. Smart grid topologies typically include all physical nodes and connecting branches, but are not optimized for specific control tasks. Therefore, to achieve task-specific control efficiency and system coupling optimization, it is necessary to abstract and transform the topology through factor constraints, forming a reconfigured topology structure with equivalent control capabilities but lower computational overhead and more accurate decision-making.

[0102] Specifically, the energy storage factor set is first analyzed to extract two types of implicit constraints: one type is the first constraint, namely the restrictive adjustment boundary, including the nodes for priority energy storage deployment, voltage / frequency limiting range, and upper and lower limits of SOC; the other type is the second constraint, namely the topology adaptability requirements, including the feasibility of physical node access, the controllability of current vector potential, and the ability to maintain voltage stability range.

[0103] Based on this, by logically eliminating or equivalently converting nodes and branches in the power grid topology that are not directly related to the energy storage factor set or do not meet the constraints, a reconstructed smart grid topology is constructed. This topology retains the core paths, electrical channels and active nodes related to the control task, which significantly improves the efficiency of control operations and the accuracy of local response.

[0104] Subsequently, after completing the equivalent reconfiguration, power storage control decisions under consensus constraints are executed. This involves each of the three roles—energy storage unit, distribution terminal, and dispatch center—independently evaluating and making local decisions regarding the control tasks in the reconfigured topology, which are then integrated into a global control scheme. Specifically:

[0105] The energy storage unit proposes a first adjustment scheme based on its adjustable range and the current SOC state, focusing on local electrical load balance and energy release priority.

[0106] On the distribution terminal side, a second regulation scheme is proposed based on the stability of node voltage and current and power flow distribution, focusing on the operational safety and accessibility of the local network.

[0107] Based on global load forecasting, electricity price fluctuations, and other macroeconomic objectives, the dispatch center generates a third adjustment scheme, which is mainly aimed at economic efficiency and overall dispatch coordination.

[0108] In a further implementation, in order to form a consistent and executable regulatory decision, a tripartite consensus mechanism is introduced to integrate the three independent schemes and to compensate and optimize the conflicting or inconsistent regulatory instructions.

[0109] Optionally, the compensation process adopts a multi-party balanced optimization method of non-consensus part, balances the control range, time window and node priority under the premise of not violating the core interests of the three parties, and generates the final power storage adjustment scheme. The scheme has the comprehensive properties of task target consistency, local response effectiveness and system scheduling coordination, which can be used as the basis for subsequent decomposition and execution control, and completes the adaptive and efficient collaborative management of energy storage resources in smart grid.

[0110] Further, the smart grid topology is equivalently reconstructed, and the step S3 of the application comprises:

[0111] The set of energy storage factors is identified, and first constraint conditions and second constraint conditions are determined, wherein the first constraint conditions are used to constrain the priority of energy storage and limit the energy storage action, and the second constraint conditions are used to constrain the smart grid topology and the topology node factor; the smart grid topology is equivalently reconstructed based on the first constraint conditions and the second constraint conditions, and a reconstructed smart grid is determined.

[0112] In the embodiment, in order to realize the structural adaptation of the energy storage control scheme and the smart grid operation state, the current grid topology structure needs to be simplified and transformed based on the previously generated set of energy storage factors, and an equivalent reconstructed topology structure serving the current control task is constructed.

[0113] Therefore, first, the set of energy storage factors is structurally identified, and the core elements constituting the control constraints are extracted and divided into two types of control boundary conditions: first constraint conditions and second constraint conditions.

[0114] In the application, the first constraint conditions mainly limit the operation boundary of the energy storage unit itself, and reflect the action ability and adjustment priority of the energy storage device. The conditions include but are not limited to the following contents: first, the priority of energy storage nodes, i.e. the energy storage units that need to participate in priority under the current task target, and the selection basis can include: SOC state is at a high value, electricity price is at a peak value, or there is a frequency deviation in the region where the node is located, etc.; second, the nodes or sections that limit the energy storage action, i.e. the energy storage units that are not suitable for adjustment operation due to the current SOC at a low value, abnormal voltage / current state, unstable grid load, etc. According to these constraints, the energy storage devices without adjustment qualification are screened out, and a core energy storage cluster participating in task execution is constructed.

[0115] The second constraint condition is used to limit or optimize the structure topology of the smart grid, to ensure that the energy storage adjustment behavior can be carried out on the controllable and stable physical basis of the power grid. The condition includes but is not limited to: first, the connectivity requirement of the topology structure, such as whether the node where the energy storage access point is located is physically connected with the main load path or the main branch; second, the topology node factor limit, that is, whether the voltage potential, short-circuit capacity and current allowable value of the node itself are in the adjustable interval, whether the node has the risk of electrical island, etc. Through the above restrictions, the edge nodes, isolated nodes or critical nodes with high risk which are not suitable for electrical adjustment in structure can be excluded, to ensure that the reconstructed topology has the implementability of adjustment behavior.

[0116] Subsequently, after identifying and establishing the above two constraint conditions, the original smart grid topology is equivalently reconstructed based on the content thereof.

[0117] The following is a specific method for the feasibility of equivalent reconstruction:

[0118] With the energy storage unit access nodes satisfying the first constraint condition and the main branch satisfying the second constraint condition as the core, a new set of grid nodes and a set of branches are constructed; the part that does not satisfy the condition is topologically excluded, edge is converted or equivalently merged, so as to obtain a reconstructed smart grid with more compact structure, higher control pertinence and calculation efficiency.

[0119] Preferably, the reconstruction result only retains the electrical path and control node closely related to the current control task, which reduces the subsequent controller reasoning complexity and execution overhead, and lays a data foundation and structural support for constructing fine energy storage control strategy in the minimum control domain.

[0120] In the above manner, the scheme of the present application can dynamically adapt to different task requirements and operating environments, and realize the coordinated response and dynamic evolution of the topology structure to the control target.

[0121] Further, the power storage adjustment scheme is determined by executing the power storage adjustment decision under the consensus constraint. The step S3 of the present application comprises:

[0122] For the reconstructed smart grid, the energy storage unit makes a power storage adjustment decision to determine a first adjustment scheme; the power distribution terminal makes a power storage adjustment decision to determine a second adjustment scheme; the dispatching center makes a power storage adjustment decision to determine a third adjustment scheme; the first adjustment scheme, the second adjustment scheme and the third adjustment scheme are compensated under the consensus constraint to determine the power storage adjustment scheme, wherein the consensus constraint compensation is a multi-party balanced optimization control of the non-consensus part of the scheme.

[0123] In the embodiment, based on the aforementioned reconstructed smart grid completed, station analysis is performed from the perspectives of the energy storage unit, the power distribution terminal and the dispatching center respectively, and power storage adjustment decisions are independently executed to realize dynamic regulation and control strategy generation under the collaborative participation of multiple roles.

[0124] In the implementation process, due to the different responsibilities, perceptible ranges and regulation targets of the subjects in the smart grid, the three-party decision results are different, so it is necessary to coordinate through the consensus mechanism while maintaining the autonomy, and finally form a power storage adjustment scheme with global consistency.

[0125] First, the energy storage unit makes power storage adjustment decisions to determine the first adjustment scheme. It focuses on the evaluation of the state and local response capability of the energy storage resource, determines the willingness and achievable power output range of the energy storage unit according to the current SOC level, temperature control state, charge and discharge capability, grid access point voltage stability and other parameters. It reflects the active response capability of the energy storage unit to the task and proposes its optimal response strategy under the guidance of the regulation target.

[0126] For example, if the current SOC of a certain energy storage unit is high and the node is in a high frequency offset area, a discharge strategy is generated to support frequency regulation.

[0127] Second, the power distribution terminal makes power storage adjustment decisions to determine the second adjustment scheme. It focuses on the electrical stability and local power flow safety of the smart grid. The power distribution terminal analyzes the voltage, current load, harmonic, voltage flicker and other parameters of the nodes under its jurisdiction, determines whether the access behavior of the energy storage unit may cause node voltage abnormalities, current overload or branch overvoltage, and proposes relatively conservative or enhanced adjustment suggestions according to the local power flow direction and path impedance. The decision of the power distribution terminal has a network constraint priority orientation to ensure that the adjustment operation does not pose a threat to the stability of the distribution network.

[0128] Third, the dispatching center makes power storage adjustment decisions to determine the third adjustment scheme. This scheme is based on multi-dimensional data such as global load forecasting, electricity price trend, reserve capacity and regulation cost, and optimizes the adjustment timing, adjustment amount allocation and operation cost structure from a macro perspective. The dispatching center prioritizes global economy, service level response and reasonable scheduling of reserve resources, and sets the energy storage power scheduling window and priority based on this.

[0129] In summary, the above three adjustment schemes respectively reflect different considerations of local response (energy storage unit), node physical constraints (power distribution terminal) and system coordination (dispatching center), and the results need to be consensus-constrained and compensated before unified regulation is executed.

[0130] Specifically, the three adjustment schemes are compared to identify non-consensus parts, i.e., conflicting or inconsistent adjustment nodes, power settings, timing strategies, etc. In the embodiments of the present application, a multi-party balanced optimization adjustment method is adopted to generate a balanced optimal fusion scheme by comprehensively adjusting related parameters based on a preset weight system or a negotiation feedback mechanism under the conditions of meeting the minimum safety constraints and optimal economic constraints.

[0131] For example, for single adjustment with a preset amplitude modulation, the non-consensus parts are iteratively adjusted and cost evaluated, and the scheme with the lowest cost is selected as the final fusion scheme.

[0132] In summary, the output power storage adjustment scheme covers all the target nodes, adjustment power, execution timing and priority of the energy storage participating units in terms of content, and has actual executability and full-system coordination consistency under the three-party consensus framework.

[0133] Through the above mechanism, the present application realizes a collaborative power storage regulation process dominated by multiple control subjects, and improves the comprehensiveness, system stability and resource response efficiency of the scheme.

[0134] S4: Based on the ownership of the smart grid access components, the power storage adjustment scheme is decomposed, and the smart grid executes the decomposed scheme to issue and drive the regulation management.

[0135] In the present embodiment, in order to realize accurate execution of the finally generated power storage adjustment scheme in the smart grid, the overall adjustment scheme needs to be structurally decomposed based on the ownership of the smart grid access components.

[0136] The access component ownership refers to the physical domain, control domain or management unit to which each energy storage unit, node device and execution terminal belongs in the smart grid system, for example, a certain energy storage unit belongs to a certain substation control area, a certain distribution terminal belongs to a certain local automation system, a certain dispatching command needs to be issued to the substation node through the master control platform, etc.

[0137] Specifically, first, all control units, energy storage nodes and adjustment instruction items in the power storage adjustment scheme are parsed and classified and identified, and a mapping relationship table between the adjustment instructions and the access components is established.

[0138] Optionally, the relationship table is generated based on the power grid device archives and the communication architecture database, which can determine which smart control node each adjustment instruction should be sent to, via which communication channel, and which subsystem or control domain is responsible for execution, thereby avoiding instruction broadcasting or invalid scheduling and improving regulation efficiency.

[0139] After completing the home identification, according to the mapping relationship, the adjustment scheme is task decomposed to form multiple sub-adjustment schemes facing different control domains, and each sub-scheme only contains part of the adjustment content required to be executed by the corresponding control component.

[0140] For example, the adjustment power, execution period and SOC limit involved in a storage unit for region A will be packaged into the adjustment sub-scheme dedicated to the region master controller, and the control items of the storage unit of region B will be issued to another control sub-network, without interfering with each other, so as to avoid wasting scheduling resources due to low region coupling degree.

[0141] Subsequently, the control issuing mechanism of the smart grid is triggered, and each decomposed sub-scheme is issued to the corresponding control node or device terminal through a scheduling interface or a communication bus. The issuing process is unified by the smart grid, supports periodic refreshing, state feedback collection and instruction confirmation mechanism, and ensures that each adjustment task can be executed, the process can be tracked, and the results can be fed back.

[0142] Finally, a closed-loop execution chain composed of global generation, local decomposition, sub-domain execution and unified feedback is formed, which realizes the collaborative driving and stable landing of the power storage adjustment scheme in a multi-source heterogeneous system, and significantly enhances the effectiveness of the regulation scheme, the accuracy of the instruction transmission and the controllability of the system response.

[0143] The power storage regulation method of the smart grid provided by the application has the following technical effects:

[0144] 1. Dynamic grid model construction: based on the smart grid topology of the pre-management region, the grid admittance matrix is constructed by calculating the self-admittance and mutual-admittance of the nodes, and is updated in real time with the change of the topology. The dynamic changes of the grid structure and parameters are accurately reflected, providing an accurate mathematical model basis for power regulation and improving the adaptability of the regulation strategy. Fuzzy controller intelligent decision: a fuzzy controller is built with the energy storage participation factor, inputting data such as SOC state and electricity price signal, and outputting the best control item to determine the energy storage factor set through multi-level threshold screening and supervised training. Dynamic adjustment of energy storage strategy based on multi-dimensional information balances grid stability, economy and energy storage device state, and optimizes resource allocation efficiency. Decoupling verification and dynamic compensation: decouple the fuzzy controller and the logic reasoner, verify the output results, and optimize the non-consensus part of the multi-party adjustment scheme. Through the logic verification and dynamic compensation mechanism, the accuracy and stability of the regulation decision are guaranteed, and the power storage adjustment scheme is effectively landed.

[0145] 2. Topology reconfiguration and consensus constraint: based on the equivalent reconfiguration of the power grid topology of the energy storage factor set, the power storage control decision is made in combination with the three-party consensus of the energy storage unit, the power distribution terminal and the dispatch center. Through topology optimization and multi-party cooperation, the control scheme is ensured to meet the needs of different subjects, the decision feasibility is enhanced, the single-point conflict is avoided, and the overall control ability of the power grid is improved.

[0146] 3. Task-driven database adaptation: according to the active or passive control task, the energy storage participation item is mined, the control database is constructed, and the task-specific database is generated to guide the fuzzy controller decision. The task-level customization of the control strategy is realized, the redundant calculation is reduced, the optimal energy storage scheme is quickly matched, and the response efficiency and resource utilization are improved.

[0147] Embodiment two: based on the same inventive concept as the power storage control method of the intelligent power grid in the foregoing embodiment, as shown in Figure 2 The application provides an intelligent power grid power storage control system, which comprises:

[0148] A matrix construction unit 11 is configured to construct a power grid admittance matrix based on the topology of the intelligent power grid in the pre-management area, wherein the power grid admittance matrix is dynamically updated.

[0149] A factor determination unit 12 is configured to introduce a fuzzy controller based on the energy storage participation factor, receive the power storage control task by connecting to the grid, and determine the energy storage factor set based on the grid data and the power grid admittance matrix, wherein the fuzzy controller takes the SOC state, the electricity price signal and the grid frequency deviation as inputs, takes the grid node voltage potential and the line current vector potential as intermediate variables, and takes the optimal control item as output.

[0150] A reconfiguration decision unit 13 is configured to perform equivalent reconfiguration of the intelligent power grid topology based on the energy storage factor set, execute the power storage control decision under the consensus constraint, and determine the power storage adjustment scheme, wherein the consensus constraint is performed based on the three-party consensus of the energy storage unit, the power distribution terminal and the dispatch center.

[0151] An energy storage control unit 14 is configured to decompose the power storage adjustment scheme based on the access components belonging to the intelligent power grid, and the intelligent power grid executes the decomposed scheme to drive the control management.

[0152] Further, the matrix construction unit 11 performs the following steps: for the topology of the intelligent power grid in the pre-management area, the sum of the self-admittance elements of the nodes and the admittance of the connected power grid branches is determined; for any two topology nodes, the negative value of the admittance of the connected power grid branches between the nodes is determined as the mutual admittance element; the self-admittance element is taken as the diagonal element, and the mutual admittance element is taken as the non-diagonal element to construct the power grid admittance matrix, wherein the power grid admittance matrix is updated with the intelligent power grid topology.

[0153] Further, the factor determination unit 12 performs the following steps: setting a first threshold with task constraints, wherein the first threshold performs filtering of grid data and grid admittance matrix; setting a second threshold with analysis of grid node voltage potential and line current vector potential; setting a third threshold based on optimal control item positioning of the regulation database, wherein the third threshold performs filtering of energy storage participation items with dynamic weight constraints; cascading the first threshold, the second threshold and the third threshold, and generating the fuzzy controller through supervised training to convergence.

[0154] Further, the factor determination unit 12 performs the following steps: obtaining a power storage regulation task, wherein the power storage regulation task is an active task or a passive task, the active task is generated by smart grid self-checking, and the passive task is uploaded by the server; mining energy storage participation items of parameter power storage regulation for smart grid topology to form a regulation database, wherein the energy storage participation item is identified by a topology node code, and the regulation database is built-in the fuzzy controller; determining a task regulation database by interpreting the power storage regulation task and selectively framing the regulation database.

[0155] Further, the factor determination unit 12 performs the following steps: determining an energy storage controller through supervised training to convergence; decoupling the energy storage controller to determine the fuzzy controller and the logic reasoner, with input-multiple-level threshold output as a first decoupling target, and inference logic of input-output of supervised training as a second decoupling target; establishing a connection between the fuzzy controller and the data block stored by the smart grid and the grid admittance matrix, and taking the logic reasoner as a branch of the fuzzy controller, wherein the logic reasoner is selectively triggered to perform logical verification on the output of the fuzzy controller.

[0156] Further, the factor determination unit 12 performs the following steps: with the acquisition of the power storage regulation task, retrieving grid data, wherein the grid data is filtered through the first threshold to determine the structured SOC state, electricity price signal and grid frequency deviation; synchronizing the grid admittance matrix, wherein the grid admittance matrix is determined by the first threshold filtering to determine the task correlation matrix; triggering the second threshold for the grid data and the grid admittance matrix to determine the intermediate quantity array; triggering the third threshold with the task regulation database as the regulation range, and performing dynamic weight assignment and control item filtering with the grid data, the grid admittance matrix and the intermediate quantity array to determine the energy storage factor set, wherein each energy storage factor is identified by an adjustable interval.

[0157] Further, the reconfiguration decision unit 13 performs the following steps: identifying the set of energy storage factors, determining a first constraint condition and a second constraint condition, wherein the first constraint condition is used to constrain the priority of energy storage and limit energy storage actions, and the second constraint condition is used to constrain the smart grid topology and the topology node factor; and performing equivalent reconfiguration on the smart grid topology based on the first constraint condition and the second constraint condition to determine a reconfigured smart grid.

[0158] Further, the reconfiguration decision unit 13 performs the following steps: performing power storage adjustment decision for the reconfigured smart grid, determining a first adjustment scheme; performing power storage adjustment decision for the distribution terminal, determining a second adjustment scheme; performing power storage adjustment decision for the dispatching center, determining a third adjustment scheme; and performing consensus constraint compensation on the first adjustment scheme, the second adjustment scheme, and the third adjustment scheme to determine the power storage adjustment scheme, wherein the consensus constraint compensation is a multi-party balanced optimization control for non-consensus parts of the scheme.

[0159] Based on the foregoing embodiments, the embodiments of the present application also provide an electronic device. Figure 3 FIG. 1 is a structural schematic diagram of an electronic device provided by an embodiment of the present application, which shows a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present application. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functions and use range of the embodiments of the present application. The electronic device is in the form of a general computing device, and its components can include but are not limited to a processor 21, a memory 22, an input device 23, and an output device 24. The processor 21 can be one or more; the memory 22 can include a computer readable medium and at least one program product, which has a set of (at least one) program modules configured to perform the functions of the embodiments of the present application.

[0160] The memory 22 shown in the embodiments of the present application can adopt any combination of one or more computer readable media; the computer readable storage medium can be but is not limited to an infrared ray, a semiconductor system, a device, or a component, or any combination of the above, for storing software programs, computer executable programs, and modules, such as the program instructions / modules of the power storage control method of the smart grid in the embodiments of the present application. The processor 21 performs various functional applications and data processing of the computer device by running the software programs, instructions, and modules stored in the memory 22, that is, implements the power storage control method of the smart grid.

[0161] The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Modifications of these embodiments will occur to persons of skill in the art, and that the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A power storage regulation method for a smart grid, characterized by, The method comprises: constructing a power grid admittance matrix based on a pre-management area intelligent power grid topology, wherein the power grid admittance matrix is dynamically updated; introducing a fuzzy controller based on a storage participation factor, connecting to the grid and receiving a power storage regulation task, determining a storage factor set based on grid data and the power grid admittance matrix, wherein the fuzzy controller takes the SOC state, the electricity price signal, and the grid frequency deviation as inputs, takes the grid node voltage potential and the line current vector potential as intermediate quantities, and takes the optimal control item as output; reconstructing the intelligent power grid topology according to the storage factor set, executing the power storage regulation decision under the consensus constraint, and determining the power storage regulation scheme, wherein the consensus constraint is performed by the three-party consensus of the storage unit, the power distribution terminal, and the dispatching center; attributing the access components based on the intelligent power grid, decomposing the power storage regulation scheme, and driving the regulation and management of the intelligent power grid executing the decomposition scheme.

2. The power storage regulating method for a smart grid according to claim 1, wherein, Constructing a power grid admittance matrix based on a pre-management area intelligent power grid topology, comprising: For the intelligent power grid topology of the pre-management area, determine the self-admittance element by summing the admittance of the node and each connected power grid branch for each topology node; For any two topology nodes, determine the mutual admittance element by the negative value of the admittance of the connected power grid branch between the nodes; Construct the power grid admittance matrix with the self-admittance element as the diagonal element and the mutual admittance element as the non-diagonal element, wherein the power grid admittance matrix is updated with the intelligent power grid topology.

3. The power storage regulating method for a smart grid of claim 1, wherein, Introducing a fuzzy controller based on a storage participation factor, comprising: Setting a first threshold with task constraints, wherein the first threshold performs filtering of grid data and the power grid admittance matrix; Setting a second threshold based on analysis of grid node voltage potential and line current vector potential; Setting a third threshold based on optimal control item positioning of the regulation database, wherein the third threshold performs filtering of the storage participation item with dynamic weight as a constraint; Cascade the first threshold, the second threshold, and the third threshold, and generate the fuzzy controller by supervised training to convergence.

4. The power storage regulating method of a smart grid according to claim 3, wherein, After receiving the power storage regulation task, comprising: Obtaining a power storage regulation task, wherein the power storage regulation task is an active task or a passive task, the active task is generated by intelligent power grid self-checking, and the passive task is uploaded by the server; For the intelligent power grid topology, mine the storage participation item of the parameter power storage regulation to form a regulation database, wherein the storage participation item is identified by topology node coding, and the regulation database is built-in the fuzzy controller; By interpreting the power storage regulation task, the regulation database is framed for selection to determine the task regulation database.

5. The power storage regulating method of a smart grid according to claim 4, wherein, Generating the fuzzy controller by supervised training to convergence, comprising: Determine the storage controller by supervised training to convergence; Decouple the storage controller by taking input-multiple-level threshold output as the first decoupling target and the inference logic of the input-output of supervised training as the second decoupling target to determine the fuzzy controller and the logical reasoner; The fuzzy controller is connected with the smart grid and a data block of the grid admittance matrix, and the logic inference device is used as a branch of the fuzzy controller, wherein the logic inference device is selectively triggered to perform logical verification on the output of the fuzzy controller.

6. The power storage regulating method of a smart grid according to claim 5, wherein, Based on the grid data and the grid admittance matrix, a set of energy storage factors is determined, including: Upon obtaining the power storage control task, the grid data is retrieved, wherein the grid data is used to determine the structured SOC state, electricity price signal, and grid frequency deviation through first threshold screening; The grid admittance matrix is synchronized, wherein the grid admittance matrix is used to determine the task correlation matrix through first threshold screening; For the grid data and the grid admittance matrix, a second threshold is triggered to determine an intermediate quantity matrix; Within the control range of the task control database, a third threshold is triggered to perform dynamic weight assignment and control item screening based on the grid data, the grid admittance matrix, and the intermediate quantity matrix, thereby determining the set of energy storage factors, wherein each energy storage factor has an adjustable interval.

7. The power storage regulating method for a smart grid of claim 1, wherein, The smart grid topology is equivalently reconstructed, including: The set of energy storage factors is identified to determine first and second constraint conditions, wherein the first constraint condition is used to constrain the priority of deploying energy storage and limit energy storage actions, and the second constraint condition is used to constrain the smart grid topology and topology node factors; The smart grid topology is equivalently reconstructed based on the first and second constraint conditions to determine a reconstructed smart grid.

8. The power storage regulating method of a smart grid according to claim 7, wherein, The power storage control decision under consensus constraints is executed to determine a power storage adjustment scheme, including: For the reconstructed smart grid, an energy storage unit is positioned to make a power storage adjustment decision to determine a first adjustment scheme; A power distribution terminal is positioned to make a power storage adjustment decision to determine a second adjustment scheme; A dispatch center is positioned to make a power storage adjustment decision to determine a third adjustment scheme; The first, second, and third adjustment schemes are subjected to consensus constraint compensation to determine the power storage adjustment scheme, wherein the consensus constraint compensation is a multi-party balanced optimization control for non-consensus parts of the schemes.

9. A power storage regulating system for a smart grid, characterized by, A system for executing the power storage control method of any one of claims 1-8, the system comprising: A matrix construction unit for constructing a grid admittance matrix based on the topology of the smart grid in the pre-management area, wherein the grid admittance matrix is dynamically updated; A factor determination unit for introducing a fuzzy controller based on the energy storage participation factor, receiving the power storage control task through network connection, and determining a set of energy storage factors based on the grid data and the grid admittance matrix, wherein the fuzzy controller takes the SOC state, electricity price signal, and grid frequency deviation as input, takes the grid node voltage potential and line current vector potential as intermediate quantities, and takes the optimal control item as output; A reconstruction decision unit for equivalently reconstructing the smart grid topology based on the set of energy storage factors, executing the power storage control decision under consensus constraints, and determining the power storage adjustment scheme, wherein the consensus constraint is performed based on the three-party consensus of the energy storage unit, power distribution terminal, and dispatch center. The energy storage regulation unit is used to decompose the power storage regulation scheme based on the access component belonging to the smart grid, and the smart grid executes the decomposed scheme to drive the regulation and management.

10. An electronic device, comprising: The electronic device comprises: a memory for storing executable instructions; a processor for executing the executable instructions stored in the memory to implement the power storage regulation method of the smart grid according to any one of claims 1 to 8.

Images