A power distribution network stability control method based on multi-agent

Abstract

The application discloses a power distribution network stability control method based on multiple agents, comprising the following steps: deploying agents, collecting data and constructing graph data; establishing an improved graph echo state network and forming a structured input sequence; inputting the input sequence for time evolution and forming a dynamic echo state vector; exchanging the dynamic echo state vector with neighboring agents to splice and form comprehensive time sequence state information; each agent analyzes the comprehensive time sequence state information to generate a local stability control instruction; when a disturbance occurs, the comprehensive time sequence state information is updated and a disturbance updated local stability control instruction is generated; and the dynamic echo state vector is updated, the comprehensive time sequence state information is constructed and the local stability control instruction is generated periodically and continuously in a normal operation cycle. The application realizes power distribution network stability control by using an improved graph echo network, and has the advantages of strong real-time performance and fast disturbance response.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network operation control, and particularly relates to a power distribution network stability control method based on multiple agents. BACKGROUND

[0002] With high proportion of distributed new energy connected to the power distribution network, node voltage fluctuation is frequent, and power flow direction changes rapidly. The traditional power distribution network stability control mode with a central controller as the core gradually exposes limitations in real-time performance and scalability. In the prior art, centralized regulation and control or decentralized control strategies based on local measurements are commonly used. However, such methods are difficult to uniformly depict the topological relationship between nodes, cannot consistently model the time sequence operation state of each node, have insufficient response capability to rapid disturbances, and are prone to control bottlenecks under communication delays or local faults.

[0003] Existing research attempts to introduce multi-agent collaborative control ideas and machine learning methods based on graph structures, but generally lacks a unified modeling framework that can simultaneously reflect the topological constraints and time sequence dynamic characteristics of the power grid. The information structure exchanged between agents is loose, and cannot support stable collaborative decision-making across nodes. At the same time, the existing graph structure model is imperfect in terms of state update mechanism under disturbance scenarios, making it difficult to effectively decouple periodic operation control and disturbance response control. Therefore, there is an urgent need for a power distribution network stability control method that is distributed, highly real-time, and has strong disturbance adaptability under graph topology constraints.

[0004] Therefore, how to provide a power distribution network stability control method based on multiple agents is a problem that those skilled in the art need to solve. SUMMARY

[0005] One object of the present application is to provide a power distribution network stability control method based on multiple agents. The present application uses an improved graph echo network to achieve power distribution network stability control, with the advantages of strong real-time performance and fast disturbance response.

[0006] According to the power distribution network stability control method based on multiple agents of the present application, the following steps are included:

[0007] Deploy agents at each operating node of the power distribution network, collect the voltage amplitude, phase angle, active power, reactive power of the node, and the connection relationship between nodes, and construct graph data;

[0008] Establish an improved graph echo state network for each agent, set input weights, reservoir pool weights, and state update parameters, and form a structured input sequence;

[0009] Input the input sequence into the improved graph echo state network of the corresponding agent, perform time sequence evolution, obtain the current reservoir pool state, and form a dynamic echo state vector;

[0010] Each intelligent agent sends a dynamic echo state vector to a neighborhood agent, receives a dynamic echo state vector sent by a neighborhood agent, and splices to form comprehensive time sequence state information;

[0011] Each intelligent agent analyzes the voltage, active power, reactive power and correlation relationship change characteristics of the node and neighborhood node according to the comprehensive time sequence state information, and generates a local stability control instruction;

[0012] When a disturbance occurs in the distribution network, the disturbance time sequence operation data is obtained, a disturbance updated dynamic echo state vector is generated, a disturbance updated comprehensive time sequence state information is constructed, and a disturbance updated local stability control instruction is generated;

[0013] Each intelligent agent periodically obtains periodic time sequence operation data in a normal operation period, and continuously executes dynamic echo state vector updating, comprehensive time sequence state information construction and local stability control instruction generation.

[0014] Optionally, the generation of the graph data includes:

[0015] Each operating node of the distribution network is assigned a unique node index, and all node indexes are arranged in a fixed order to form a fixed node index sequence;

[0016] According to the fixed node index sequence, the connection relationship between any two nodes is mapped to an adjacency relationship matrix according to the corresponding relationship of the node index;

[0017] According to the fixed node index sequence, all node pairs connected by lines are recorded as an edge set in the order of the node index;

[0018] The fixed node index sequence, the adjacency relationship matrix and the edge set are uniformly organized, and the normalized voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of each node are associated with the corresponding node index to form the graph data.

[0019] Optionally, the generation of the structured input sequence includes:

[0020] The voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence are normalized;

[0021] According to the fixed node index sequence in the graph data, the normalized voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of each node are arranged in the fixed node index sequence to form a time sequence operation quantity input structure of the corresponding node;

[0022] According to the fixed node index sequence, the input weight is fixedly mapped to the time sequence operation quantity of the node according to the node index sequence;

[0023] The reserve pool weight is constructed according to the adjacency relation matrix, so that any position in the reserve pool is only connected with the weight of the neighborhood nodes of the corresponding node;

[0024] The state update parameter is set according to the preset internal attenuation coefficient, input mapping coefficient, normalization coefficient and maximum change limit coefficient, so that the state update process performs time evolution under the topological constraint in the reserve pool;

[0025] The voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence arranged in time sequence are sequentially input to the improved graph echo state network of the corresponding agent according to the node index sequence, to form a structured input sequence of the improved graph echo state network.

[0026] Optionally, the generation of the dynamic echo state vector comprises:

[0027] The structured input sequence is input to the improved graph echo state network of the corresponding agent according to the fixed node index sequence;

[0028] According to the input weight, the structured input sequence at the current time is mapped to the corresponding position in the reserve pool according to the fixed node index sequence, so that the corresponding position obtains the input driving amount at the current time;

[0029] According to the reserve pool weight, after the reserve pool state at the last time is attenuated according to the internal attenuation coefficient, the attenuated reserve pool state and the reserve pool state at the last time of all neighborhood nodes having a connection relationship with it are superimposed according to the connection weight represented by the adjacency relation matrix, to form the intermediate state at the current time;

[0030] According to the state update parameter, the input driving amount and the intermediate state are added according to a fixed proportion to obtain the final state of the reserve pool at the current time, and the final state is arranged according to the fixed node index sequence to form the dynamic echo state vector.

[0031] Optionally, the generation of the comprehensive time sequence state information comprises:

[0032] After each agent completes the generation of the dynamic echo state vector of the corresponding node, the agent structures the dynamic echo state vector of the node obtained at the current time, and each agent sends the structured dynamic echo state vector of the node to all neighborhood agents having a connection relationship with it;

[0033] Each agent receives the neighborhood dynamic echo state vector generated at the current time from all neighborhood agents having a connection relationship with it, and preliminarily collects the received neighborhood dynamic echo state vector, and arranges the preliminarily collected neighborhood dynamic echo state vector according to the pre-set index sequence according to the fixed node index sequence;

[0034] The intelligent agent performs structure inspection on the arranged neighbor dynamic echo state vector, so that the length, order and node order identifier of the neighbor dynamic echo state vector are consistent with those of the node dynamic echo state vector;

[0035] The intelligent agent splices the node dynamic echo state vector and all the arranged and inspected neighbor dynamic echo state vectors according to a fixed combination format;

[0036] The intelligent agent performs consistency arrangement on the spliced data structure, sets the arranged spliced structure as the integrated time sequence state information at the current time, and stores the integrated time sequence state information in the internal time sequence information buffer.

[0037] Optionally, the generation of the local stability control instruction includes:

[0038] The intelligent agent obtains the integrated time sequence state information at the current time, and performs structure analysis on the node dynamic echo state vector and the neighbor dynamic echo state vector in the integrated time sequence state information, to obtain the structure-analyzed integrated time sequence state information.

[0039] The intelligent agent extracts the voltage amplitude variation feature, active power variation feature, reactive power variation feature and node association relationship variation feature of the node at the current time according to the structure-analyzed integrated time sequence state information, and combines the variation features in a fixed order to form a time sequence operation variation set of the node.

[0040] The intelligent agent extracts the voltage amplitude variation feature, active power variation feature, reactive power variation feature and node association relationship variation feature of all the neighbor nodes at the current time according to the structure-analyzed integrated time sequence state information, and arranges the variation features in a fixed node index order to form a neighbor time sequence operation variation set.

[0041] The intelligent agent fuses the time sequence operation variation set of the node and the neighbor time sequence operation variation set according to a fixed combination format, so that the fused time sequence operation variation structure can reflect the local topological relationship and the synchronous time sequence operation state of the node and the neighbor nodes.

[0042] The intelligent agent generates a node-level adjustment amount according to the fused time sequence operation variation structure.

[0043] The intelligent agent sets the node-level adjustment amount as a local stability control instruction at the current time, and executes the local stability control instruction through the corresponding execution device of the node.

[0044] Optionally, the generation of the disturbance updated local stability control instruction includes:

[0045] The intelligent agent obtains disturbance timing operation data of the node when detecting new energy output fluctuation, load change or line disturbance, arranges the disturbance timing operation data into a disturbance update input sequence according to a fixed node index sequence;

[0046] The disturbance update input sequence is input into an improved graph echo state network to form a disturbance update reserve pool state;

[0047] The disturbance update reserve pool state is arranged according to the fixed node index sequence to form a disturbance update dynamic echo state vector of the node, which is sent to all neighborhood intelligent agents connected thereto, and all neighborhood disturbance update dynamic echo state vectors generated by the neighborhood nodes at the same disturbance moment are received;

[0048] All neighborhood disturbance update dynamic echo state vectors are arranged according to the fixed node index sequence;

[0049] The disturbance update dynamic echo state vector of the node and the arranged neighborhood disturbance update dynamic echo state vector are spliced according to a fixed combination format to form a disturbance update comprehensive timing state information;

[0050] The voltage amplitude change characteristics, active power change characteristics, reactive power change characteristics and node association relationship change characteristics of the node and the neighborhood nodes at the disturbance moment are extracted according to the disturbance update comprehensive timing state information, and the change characteristics are arranged into a disturbance update operation quantity set;

[0051] The node-level adjustment quantity is generated according to the disturbance update operation quantity set, and the node-level adjustment quantity is taken as a disturbance update local stability control instruction, and the disturbance update local stability control instruction is executed by a corresponding execution device.

[0052] Optionally, the generation of the local stability control instruction comprises:

[0053] Each intelligent agent periodically obtains periodic timing operation data of the node when the power distribution network is in a normal operation state, and arranges the periodic timing operation data into a periodic update input sequence according to a fixed node index sequence;

[0054] Each intelligent agent inputs the periodic update input sequence into an improved graph echo state network to form a periodic update reserve pool state, and arranges the periodic update reserve pool state according to a fixed node index sequence to form a periodic update dynamic echo state vector of the node;

[0055] Each intelligent agent sends the periodic update dynamic echo state vector of the node to all neighborhood intelligent agents connected thereto, and receives periodic update dynamic echo state vectors of all neighborhood nodes;

[0056] The intelligent agent arranges all the neighborhood period update dynamic echo state vectors in a fixed node index order, and splices the arranged neighborhood period update dynamic echo state vectors and the node period update dynamic echo state vector according to a fixed combination format;

[0057] The intelligent agent extracts the voltage amplitude change characteristics, active power change characteristics, reactive power change characteristics and node association relationship change characteristics of the node and the neighborhood node according to the period update comprehensive time sequence state information, and arranges the extracted change characteristics into a period update operation quantity set;

[0058] The intelligent agent generates a period update node-level adjustment quantity according to the period update operation quantity set, takes the period update node-level adjustment quantity as a period update local stability control instruction, and executes the period update local stability control instruction through a corresponding execution device;

[0059] After the intelligent agent completes the execution of the period update local stability control instruction, the intelligent agent distinguishes the period update input sequence, the period update reserve pool state, the period update dynamic echo state vector, the period update comprehensive time sequence state information and the period update node-level adjustment quantity from the corresponding contents generated in the disturbance update process, so that the period update control chain and the disturbance update control chain are independent in structure and can maintain overall stable operation in a double-track manner during the operation of the power distribution network.

[0060] The beneficial effects of the present application are:

[0061] The present application arranges intelligent agents at each node of the power distribution network, and is based on the graph data constructed by the fixed node index, the adjacency relationship matrix and the edge set, so that the time sequence operation quantities such as the voltage, the phase angle, the active power and the reactive power of the node form a strict one-to-one correspondence with the topology structure of the power distribution network, realizing the structured power grid modeling effect that cannot be achieved by traditional technologies. On this basis, the present application introduces an improved graph echo state network, directly embeds the topological constraint into the reserve pool weight, so that the state evolution in the reserve pool strictly follows the physical connection relationship of the power grid, forms a graph structure time sequence evolution chain of the node time sequence state, can continuously capture the dynamic coupling relationship between the nodes, and significantly enhances the time sequence expression ability of the complex electrical quantity change. Through the exchange of the dynamic echo state vectors between the intelligent agents, the present application constructs comprehensive time sequence state information containing the cooperative information of the node and the neighborhood node, so that each intelligent agent has the perception ability of the voltage and the trend of the power flow change in the regional range at the local, thereby avoiding the defects of the traditional centralized control structure, such as the dependence on the center node, the large communication pressure and the single point fault.

[0062] In addition, the application sets a dual-track operation mechanism of the periodical update control chain and the disturbance update control chain, so that the intelligent agent can continuously update the power grid operation state in the normal operation period, and can quickly generate a disturbance update dynamic echo state vector and disturbance update comprehensive time sequence state information locally when new energy output fluctuation, load sudden change or line disturbance occurs, to realize disturbance response in seconds or even milliseconds. This mechanism solves the problems of high coupling degree of periodical control and disturbance control, slow disturbance response and insufficient dynamic robustness in the existing method. Through the structural analysis, change feature extraction and fusion of the comprehensive time sequence state information, the application can generate local stability control instructions with physical meaning for each node, which can directly act on voltage, active power, reactive power and power flow regulation, so that the control strategy has both distributed autonomy and consistency and collaboration in the whole network. In summary, the application significantly improves the dynamic stability, disturbance recovery speed, distributed collaboration ability and real-time online regulation ability of the distribution network under the condition of high proportion of new energy access, and has important engineering application value and technical progress significance for solving the stability control problem of complex distribution network operation scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0063] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings:

[0064] Figure 1 A general flowchart of a distribution network stability control method based on multiple intelligent agents is proposed for the application;

[0065] Figure 2 A structure diagram of an improved graph echo state network of a distribution network stability control method based on multiple intelligent agents is proposed for the application. DETAILED DESCRIPTION

[0066] The application will now be described in further detail with reference to the drawings. These drawings are simplified schematic diagrams, which only schematically show the basic structure of the application, and therefore only show the components related to the application.

[0067] Reference Figure 1 and Figure 2 A distribution network stability control method based on multiple intelligent agents, comprising the following steps:

[0068] Intelligent agents are deployed at each operating node of the distribution network, the voltage amplitude, phase angle, active power, reactive power of the node and the connection relationship between the nodes are collected, and a graph data is constructed according to the connection relationship, the graph data includes fixed node index, adjacency relationship matrix and edge set, so that the node operation quantity and the topological structure form a structured correspondence relationship;

[0069] establish an improved graph echo state network for each agent according to the graph data, set input weights, reservoir weights and state update parameters corresponding to the graph data, the reservoir weights are constructed according to the adjacency relationship matrix, so that any position in the reservoir is only connected with the weight of the neighborhood nodes of the corresponding node, and the voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of the node are arranged in time sequence and input to the improved graph echo state network of the corresponding agent to form a structured input sequence of the improved graph echo state network;

[0070] The input sequence is input to the improved graph echo state network of the corresponding agent, the topological connection relationship represented by the reservoir and the update rule are used to perform time sequence evolution on the reservoir state, the current reservoir state is obtained through the attenuation of the last time state, the superposition of the adjacent node states according to the connection weight and the input action, and a dynamic echo state vector is formed by arranging the node sequence;

[0071] Each agent sends the dynamic echo state vector to the neighborhood agent connected therewith, receives the dynamic echo state vector sent by the neighborhood agent, arranges the node index sequence, and splices the dynamic echo state vectors of the node and the neighborhood node in a fixed format to form comprehensive time sequence state information;

[0072] Each agent analyzes the voltage, active power, reactive power and correlation relationship change characteristics of the node and the neighborhood node according to the comprehensive time sequence state information, generates a local stable control instruction for adjusting the running state of the node, and adjusts the corresponding executing mechanism in real time;

[0073] When new energy output fluctuation, load change or line disturbance occurs in the distribution network, each agent obtains new time sequence running data, updates the improved graph echo state network, generates a disturbance update dynamic echo state vector, constructs a disturbance update comprehensive time sequence state information, and generates a disturbance update local stable control instruction to realize fast disturbance response;

[0074] Each agent periodically obtains new time sequence running data in the normal operation period, continuously performs dynamic echo state vector update, comprehensive time sequence state information construction and local stable control instruction generation, so that the agent maintains the distributed stable operation of the distribution network based on the dual-track mechanism of periodic update control and disturbance update control without the need for a central controller.

[0075] In the embodiment, the generation of the graph data comprises:

[0076] Each running node of the distribution network is allocated a unique node index, and all node indexes are arranged in a fixed sequence to form a fixed node index sequence;

[0077] According to the fixed node index sequence, a connection relationship between any two nodes is mapped to an adjacency relationship matrix according to the corresponding relationship of the node indexes, so that the row and column numbers are one-to-one corresponding to the fixed node index sequence;

[0078] According to the fixed node index sequence, all node pairs connected by lines are recorded in an edge set in the order of the node indexes, so that each edge in the edge set is identified by the node number in the fixed node index sequence;

[0079] The fixed node index sequence, the adjacency relationship matrix and the edge set are uniformly organized, and the normalized voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of each node are associated with the corresponding node index, so that the time sequence operation data of the nodes and the topology structure form a one-to-one correspondence at the data level, and form a graph data for describing the structure of the power distribution network.

[0080] In the embodiment, the generation of the structured input sequence includes:

[0081] The voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence are normalized to map all time sequence operation quantities to a unified interval in the numerical range;

[0082] According to the fixed node index sequence in the graph data, the normalized voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of each node are arranged in the order of the fixed node index, forming a time sequence operation quantity input structure of the corresponding node;

[0083] The generation of the time sequence operation quantity input structure specifically includes: arranging the voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence collected in the continuous time of each node in time order, so that each type of time sequence operation quantity is arranged in consistent in the time dimension; according to the fixed node index order, each type of time sequence operation quantity after arrangement is one-to-one corresponding to the corresponding node, so that the time sequence operation quantity of each node occupies a stable position in structure; the voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of each node are combined according to the fixed combination format, forming a node-level time sequence operation quantity structure; the node-level time sequence operation quantity structures of all nodes are arranged in the order of the fixed node index in turn, forming the overall time sequence operation quantity input structure;

[0084] According to the fixed node index sequence, the input weight is fixedly mapped to the time sequence operation quantity of the node in the order of the node index, so that each position of the input weight is one-to-one corresponding to the time sequence operation quantity of the corresponding node;

[0085] The reserve pool weight is constructed according to the adjacency relationship matrix, so that any position in the reserve pool is connected with the neighborhood nodes of the corresponding node by the weight, and the connection relationship of all reserve pool weights is consistent with the structure of the adjacency relationship matrix;

[0086] The state update parameter is set according to the preset internal attenuation coefficient, input mapping coefficient, normalization coefficient and maximum change limiting coefficient, so that the state update process performs time evolution under the topological constraint in the reserve pool;

[0087] The time evolution specifically includes: attenuating the reserve pool state at the last time according to the internal attenuation coefficient, so that the state at the last time forms a basis state after attenuation at the current time; according to the connection relationship between nodes represented by the reserve pool weight, the basis state after attenuation and the reserve pool state at the last time of all neighborhood nodes connected therewith are weighted and superimposed, so that the result after weighted superposition reflects the propagation influence of the node under the graph topological constraint; on this basis, the input action and the weighted superposition result are combined in a fixed proportion, so that the combined state forms the reserve pool state at the current time; and the reserve pool state at the current time is taken as the basis input for generating the subsequent dynamic echo state vector, so that the reserve pool state forms a time evolution process with graph structure constraint in continuous time;

[0088] The voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence arranged in time sequence are input to the improved graph echo state network of the corresponding agent in node index order, so that the input sequence and the reserve pool structure establish a structured corresponding relationship, and form a structured input sequence of the improved graph echo state network.

[0089] In the embodiment, the generation of the dynamic echo state vector includes:

[0090] The structured input sequence is input to the improved graph echo state network of the corresponding agent in a fixed node index order, so that each time sequence operation quantity establishes a fixed corresponding relationship with the position of the corresponding node in the reserve pool;

[0091] According to the input weight, the structured input sequence at the current time is mapped to the corresponding position in the reserve pool in a fixed node index order, so that the corresponding position obtains the input driving quantity at the current time;

[0092] The generation of the input driving quantity specifically comprises: arranging the voltage amplitude, phase angle, active power and reactive power collected by the node at the current time in order according to the fixed node index sequence, so that each type of time sequence operation quantity of the same node occupies a fixed position in structure; according to the input weight set in the improved echo state network, proportionally acting each type of time sequence operation quantity after the arrangement on the corresponding input weight, so that the voltage amplitude forms a voltage input component, the phase angle forms a phase angle input component, the active power forms an active power input component, and the reactive power forms a reactive power input component; combining all the input components in a fixed combination format, so that the combined structure contains driving information of each type of time sequence operation quantity at the current time on the reserve pool state at the data level; and setting the combined structure as the input driving quantity;

[0093] According to the reserve pool weight, the reserve pool state at the last time is attenuated according to the internal attenuation coefficient, and then the attenuated reserve pool state is superimposed with the reserve pool state at the last time of all the adjacent nodes having a connection relationship with the attenuated reserve pool state according to the connection relationship represented by the adjacency relationship matrix according to the connection weight, to form an intermediate state at the current time;

[0094] According to the state update parameter, the input driving quantity and the intermediate state are added in a fixed proportion to obtain the final state of the reserve pool at the current time, and the final state is arranged in accordance with the fixed node index sequence to form a dynamic echo state vector for representing the time sequence operation characteristics of the node under the constraint of the topological structure.

[0095] In the embodiment, the generation of the comprehensive time sequence state information comprises:

[0096] After each agent completes the generation of the dynamic echo state vector at the corresponding node, the dynamic echo state vector of the node obtained at the current time is structurally arranged to maintain the same arrangement as the fixed node index sequence, and the agent sends the arranged dynamic echo state vector of the node to all the adjacent agents having a connection relationship with the agent, so that the adjacent agents can receive the dynamic echo state vector of the node at the current time;

[0097] Each agent receives the adjacent dynamic echo state vector generated at the current time from all the adjacent agents having a connection relationship with the agent, and preliminarily collects the received adjacent dynamic echo state vector, arranges the preliminarily collected adjacent dynamic echo state vector in a predetermined index sequence according to the fixed node index sequence, so that the arranged adjacent dynamic echo state vector maintains a consistent index correspondence relationship in structure;

[0098] The intelligent agent performs structure checking on the arranged neighbor dynamic echo state vector, so that the length, order and node order identifier of the neighbor dynamic echo state vector are consistent with those of the node dynamic echo state vector, to ensure that the splicing can be combined according to a unified structure;

[0099] The intelligent agent splices the node dynamic echo state vector and all the arranged and checked neighbor dynamic echo state vectors according to a fixed combination format, so that the spliced result contains the dynamic echo state vectors of the node and all the neighbor nodes generated at the current time in a single data structure in sequence;

[0100] The intelligent agent performs consistency arrangement on the spliced data structure, so that the spliced structure is consistent with the spliced structure at the previous time step in length, node corresponding position, sorting rule and overall format, to ensure that the structure is stable when the integrated time sequence state information is continuously used at different times, and the arranged spliced structure is set as the integrated time sequence state information at the current time, so that the integrated time sequence state information can reflect the time sequence running characteristics and topological structure relationship of the node and the neighbor nodes at the current time, and is stored in the internal time sequence information buffer.

[0101] In the embodiment, the generation of the local stability control instruction includes:

[0102] The intelligent agent obtains the integrated time sequence state information at the current time, and performs structure analysis on the node dynamic echo state vector and the neighbor dynamic echo state vector in the integrated time sequence state information, so that each time sequence running quantity can be corresponded to the respective node position according to the fixed node index order after the analysis, to obtain the structure-analyzed integrated time sequence state information.

[0103] The generation of the structure-analyzed integrated time sequence state information specifically includes: identifying the position of the node dynamic echo state vector and all the neighbor dynamic echo state vectors in the integrated time sequence state information according to the fixed node index order, so that the position of each dynamic echo state vector in the overall structure is determined; extracting the node dynamic echo state vector from the integrated time sequence state information, and sequentially arranging the various time sequence running quantities contained in the node dynamic echo state vector, so that the structured running quantity of the node maintains a fixed format in structure; extracting all the neighbor dynamic echo state vectors one by one according to the fixed node index order, and performing structure arrangement on the time sequence running quantities in the neighbor dynamic echo state vectors, so that the arranged neighbor structured running quantities are consistent with the node structured running quantity; arranging the node structured running quantity at the first position of the combination structure, and arranging all the neighbor structured running quantities according to the fixed node index order; combining the arranged structured running quantities according to the fixed combination format, to form the structure-analyzed integrated time sequence state information.

[0104] The intelligent agent extracts the voltage amplitude variation characteristics, active power variation characteristics, reactive power variation characteristics and node association relationship variation characteristics of the current time of the node according to the structural analysis of the comprehensive time sequence state information, and combines the variation characteristics into a time sequence operation variation set of the node in a fixed order;

[0105] The intelligent agent extracts the voltage amplitude variation characteristics, active power variation characteristics, reactive power variation characteristics and node association relationship variation characteristics of the current time of all the neighboring nodes according to the structural analysis of the comprehensive time sequence state information, and arranges the variation characteristics into a neighboring time sequence operation variation set in a fixed node index order;

[0106] The intelligent agent fuses the time sequence operation variation set of the node and the neighboring time sequence operation variation set according to a fixed combination format, so that the fused time sequence operation variation structure can simultaneously reflect the local topological structure relationship and the synchronous time sequence operation state of the node and the neighboring nodes;

[0107] The generation of the fused time sequence operation variation structure specifically includes: taking the time sequence operation variation set of the node as the first group of data of the fused structure, so that it occupies a fixed position in the overall structure; adding the time sequence operation variation sets of all the neighboring nodes one by one according to the fixed node index order, so that the time sequence operation variation set of each neighboring node is arranged in a predetermined order in the fused structure; combining the voltage amplitude variation, active power variation, reactive power variation and node association relationship variation in the time sequence operation variation set of the node according to a fixed format, and combining the corresponding variation characteristics in the time sequence operation variation sets of all the neighboring nodes according to the same format, so that the variation characteristics of the node and the neighboring nodes maintain the same arrangement in the structure; structurally splicing the combined variation characteristics according to the fixed combination format, so that the spliced fused structure simultaneously contains the time sequence operation variation characteristics of the node and all the neighboring nodes at the current time; performing a unified arrangement on the spliced fused structure, so that it is consistent with the fused structure of the previous time in node order, length, field arrangement and overall format, and setting the arranged structure as the fused time sequence operation variation structure;

[0108] The intelligent agent generates a node-level adjustment amount according to the fused time sequence operation variation structure, so that the node-level adjustment amount can correspond to the adjustment actions of adjusting the voltage, active power, reactive power or power flow distribution of the node;

[0109] The intelligent agent sets the node-level adjustment amount as a local stability control instruction at the current time, and executes the local stability control instruction through the corresponding execution device of the node, so that the node realizes stable adjustment based on the graph topology-time sequence combined operation characteristics.

[0110] In this embodiment, the generation of the disturbance update local stability control instruction comprises:

[0111] When the agent detects new energy output fluctuation, load change or line disturbance, the agent obtains the disturbance time sequence operation data of the node, arranges the disturbance time sequence operation data into a disturbance update input sequence according to the fixed node index order, and makes the disturbance update input sequence consistent in structure with the structured input sequence at the previous time;

[0112] The disturbance update input sequence is input into the improved graph echo state network, the reservoir performs a reservoir state update process independent of the cycle update according to the disturbance time sequence operation data, and a disturbance update reservoir state is formed;

[0113] The disturbance update reservoir state is arranged in the fixed node index order to form a disturbance update dynamic echo state vector of the node, which is sent to all neighboring agents connected thereto, and all neighboring disturbance update dynamic echo state vectors generated by the neighboring nodes at the same disturbance time are received;

[0114] All neighboring disturbance update dynamic echo state vectors are arranged in the fixed node index order, so that the arranged neighboring disturbance update dynamic echo state vectors can be consistent with the disturbance update dynamic echo state vector of the node in the same structure format;

[0115] The disturbance update dynamic echo state vector of the node and the arranged neighboring disturbance update dynamic echo state vector are spliced in the fixed combination format to form a disturbance update comprehensive time sequence state information;

[0116] According to the disturbance update comprehensive time sequence state information, the voltage amplitude change characteristics, active power change characteristics, reactive power change characteristics and node association relationship change characteristics of the node and the neighboring nodes at the disturbance time are extracted, and the change characteristics are arranged into a disturbance update operating quantity set;

[0117] According to the disturbance update operating quantity set, a node-level adjustment quantity for the disturbance scene is generated, so that the node-level adjustment quantity can be used to adjust the voltage, active power, reactive power or power flow distribution of the node. The node-level adjustment quantity is used as a disturbance update local stability control instruction, and the disturbance update local stability control instruction is executed by the corresponding execution device, so that the node completes rapid adjustment based on the graph topology and the time sequence operation change characteristics at the disturbance time.

[0118] In this embodiment, the generation of the local stability control instruction comprises:

[0119] When the power distribution network is in a normal operation state, each agent periodically obtains the cycle time sequence operation data of the node, and arranges the cycle time sequence operation data into a cycle update input sequence according to the fixed node index order;

[0120] The agent inputs the periodic update input sequence into the improved graph echo state network, and the reservoir performs a periodic update reservoir state generation process according to the periodic update input sequence, and forms a periodic update reservoir state. The periodic update reservoir state is arranged in a fixed node index order to form a periodic update dynamic echo state vector of the node;

[0121] The agent sends the periodic update dynamic echo state vector of the node to all neighboring agents connected thereto, and receives the periodic update dynamic echo state vectors of all neighboring nodes;

[0122] The agent arranges all the neighboring periodic update dynamic echo state vectors in a fixed node index order, and splices the arranged neighboring periodic update dynamic echo state vectors and the periodic update dynamic echo state vector of the node according to a fixed combination format, so that the spliced data form a periodic update comprehensive time sequence state information;

[0123] The agent extracts the voltage amplitude change feature, active power change feature, reactive power change feature and node association relationship change feature of the node and the neighboring nodes according to the periodic update comprehensive time sequence state information, and arranges the extracted change features into a periodic update operating quantity set;

[0124] The agent generates a periodic update node-level regulating quantity according to the periodic update operating quantity set, so that the periodic update node-level regulating quantity can be used to adjust the voltage, active power, reactive power or power flow distribution of the node. The periodic update node-level regulating quantity is used as a periodic update local stability control instruction, and the corresponding execution device is used to execute the periodic update local stability control instruction, so that the node can complete continuous stability adjustment according to the periodic update comprehensive time sequence state information in the normal operation period.

[0125] After the agent completes the execution of the periodic update local stability control instruction, the agent distinguishes the periodic update input sequence, the periodic update reservoir state, the periodic update dynamic echo state vector, the periodic update comprehensive time sequence state information and the periodic update node-level regulating quantity from the corresponding contents generated in the disturbance update process, so that the periodic update control chain and the disturbance update control chain are independent in structure and can maintain overall stable operation in a double-track manner during the operation of the power distribution network.

[0126] Embodiment 1:

[0127] To verify the feasibility of the application in implementation, the application is applied to a power supply area around a city in a southeast coastal area. The distribution network of the area covers coastal residential areas, sea wind corridor zones and several distributed new energy access points, including roof photovoltaic, industrial and commercial side photovoltaic arrays, small-scale wind turbine generators and part of energy storage systems. Due to the open terrain and rapid climate change, the distribution network of the area has always been plagued by problems such as large voltage fluctuations, frequent reverse of power flow direction and rapid changes in node burden, especially in the case of strong summer sunlight and sudden sea wind, the output fluctuation of photovoltaic and wind power is extremely obvious, making the regulation and stability of the distribution network a long-term problem for local operation and maintenance personnel.

[0128] In the distribution network, the intelligent agent collaborative control system proposed by the application is introduced, and an intelligent agent device capable of independently collecting the voltage amplitude, phase angle, active power and reactive power of each operating node is deployed. The intelligent agent organizes the above data and the connection relationship between the nodes into graph data, so that the topology structure of the entire distribution network is clearly expressed in the data layer in the form of fixed node index, adjacency relationship matrix and edge set. Compared with the traditional regulation and control method relying on single measurement data, the application enables each node to obtain a structured expression consistent with the entire network in the digital structure, avoiding the judgment bias caused by inconsistent measurement sources in the past.

[0129] After the deployment of the intelligent agent, an improved graph echo state network is constructed according to the graph data, so that the time series operation quantity of each node can be propagated in the reserve pool in a topological structure constrained manner. In actual application, when the photovoltaic output starts to rise and some nodes appear voltage lifting trend in the morning when the sunlight rapidly increases, the intelligent agent inputs the continuously collected time series operation quantity into the improved graph echo state network to generate a dynamic echo state vector through the time evolution process of the reserve pool. Since the weights of the reserve pool are completely based on the adjacency relationship, the dynamic echo state vector not only reflects the time series characteristics of the node itself, but also implicitly contains the structured information of the influence of the neighborhood nodes on it, thereby effectively overcoming the defect that the traditional local control cannot obtain regional dynamic characteristics.

[0130] In order to realize the collaborative regulation between nodes, each intelligent agent sends the dynamic echo state vector generated by the node to all adjacent nodes connected to it, and receives the dynamic echo state vectors of the adjacent nodes. The received information is arranged in order according to the fixed node index and formatted spliced to form comprehensive time series state information. Since this structure strictly follows a unified format, the intelligent agent can directly identify the operation trend of the node and the adjacent nodes at the same time without complex data alignment. In the time period of rapid change of photovoltaic, the intelligent agent can judge the change direction of voltage trend, the transfer trend of active power flow and the synchronous change relationship of reactive power through the comprehensive time series state information, so as to generate a local stable control instruction more suitable for the current state.

[0131] In this region, strong sea breezes often occur in the afternoon, causing significant fluctuations in wind power output and easily leading to simultaneous disturbances at multiple nodes. To address this issue, the disturbance update mechanism of this invention enables the agent to immediately collect new time-series operational data upon detecting a disturbance, rapidly generating a new disturbance update dynamic echo state vector. This vector is then concatenated with the corresponding vectors from neighboring nodes to form comprehensive disturbance update time-series state information. Based on this information, the agent generates disturbance update local stability control commands, ensuring stable operation of nodes during the duration of the disturbance. Under this mechanism, voltage fluctuations are rapidly suppressed, power flow impacts are significantly mitigated, the overall distribution network becomes more stable, and cascading fluctuations caused by disturbance propagation no longer occur.

[0132] During stable operation of the distribution network, the intelligent agent periodically acquires new time-series operational data for its node and executes a periodic update mechanism. At different stages—morning load increases, midday photovoltaic boosts, evening photovoltaic declines, and nighttime energy storage charging and discharging—the agent's periodically updated comprehensive time-series state information continuously reflects the grid's operational trends, enabling the control strategy to evolve and adapt over time. Compared to traditional fixed-parameter control methods, this invention allows each node to autonomously adjust its operating strategy based on periodic data, ensuring the overall distribution network maintains higher sensitivity and adaptability over long-term operation.

[0133] Long-term operational observations in this region show that, after implementing this invention, the distribution network can maintain a more stable operating state under the influence of various variables such as weather changes, load fluctuations, and unstable output from photovoltaic and wind power. The intelligent agent, leveraging the temporal information captured by the improved graph echo state network and the comprehensive temporal state information efficiently exchanged with neighboring nodes, enables control behavior to not only possess local autonomy but also demonstrate cross-node collaborative consistency. Whether in rapid response during disturbance phases or continuous adjustment during cyclical operation, this invention ensures the entire distribution network maintains reliable and stable operating characteristics under different operating conditions, fully demonstrating its practical value and potential for widespread application in complex distribution network scenarios.

[0134] Table 1. Performance comparison test data between the method of this invention and the traditional distributed control method.

[0135] Comparison Index Traditional Distributed Control Traditional Robust Control Improved ESN Method Lifting Amplitude Comparison (Traditional Distributed Control) Lifting Amplitude Comparison (Traditional Robust Control) Voltage Deviation Mean (%) 3.8 3.2 1.4 63.2% 56.3% Voltage Out-of-limit Times (times / day) 14 10 3 78.6% 70.0% Power Flow Reversal Recovery Time (second) 4.6 4.2 1.9 58.7% 54.8% Node Synchronization Deviation under Disturbance (ms) 29 24 11 62.1% 54.2% Control Command Stability Score (0-1) 0.63 0.71 0.89 +41.3% +25.4%

[0136] As can be seen from Table 1, the method of the application is significantly better than the traditional distributed control method and the traditional robust control method in multiple key performance indicators. In terms of voltage deviation, the application reduces the average deviation from 3.8% of the traditional distributed control to 1.4%, a decrease of more than 60%, effectively avoiding voltage instability caused by new energy fluctuations. In terms of voltage limit times, the method of the application reduces the limit times from 14 times to 3 times due to the introduction of the neighborhood dynamic echo state vector splicing mechanism, so that the agent can more accurately identify regional voltage trend changes, thereby greatly improving power supply quality. In the power flow reversal scene, the traditional method generally relies on static parameters or single node data, and the recovery speed is slow, while the application accelerates the state coordination between nodes through the reserve pool time evolution under dynamic topology constraints, and shortens the recovery time to less than half of the traditional distributed control. In terms of disturbance synchronization, the application significantly reduces node deviation, and the multi-node collaboration ability is comprehensively enhanced. Comprehensive analysis shows that the main reasons for performance improvement include: the reserve pool weight construction based on the graph structure avoids the defect that the traditional method is difficult to capture regional correlation; the dynamic echo state vector enables the agent to have adaptive ability in the time dimension; the composition of the comprehensive time state information makes the control instruction generation no longer rely on single node inference, but on the collaborative evolution based on local topology, which fundamentally improves the overall stability.

[0137] The above is only the preferred specific embodiment of the application, but the protection scope of the application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical solution and inventive concept of the application within the technical range disclosed by the application, which should be covered within the protection scope of the application.

Claims

1. A multi-agent based power distribution network stability control method, characterized in that, The method comprises the following steps: Deploying intelligent agents at each operating node of the power distribution network, collecting voltage amplitude, phase angle, active power, reactive power of the node and connection relationship between nodes, and constructing graph data; Establishing an improved graph echo state network for each intelligent agent, setting input weight, reservoir pool weight and state update parameters, and forming a structured input sequence; Inputting the input sequence into the improved graph echo state network of the corresponding intelligent agent, performing time evolution, obtaining the current reservoir pool state, and forming a dynamic echo state vector; Each intelligent agent sends the dynamic echo state vector to the neighboring intelligent agents and receives the dynamic echo state vector sent by the neighboring intelligent agents to splice and form comprehensive time sequence state information; Each intelligent agent analyzes the voltage, active power, reactive power and correlation relationship change characteristics of the node and the neighboring nodes according to the comprehensive time sequence state information, and generates a local stable control instruction; When a disturbance occurs in the power distribution network, the time sequence operation data of the disturbance is obtained, the disturbance update dynamic echo state vector is generated, the disturbance update comprehensive time sequence state information is constructed, and the disturbance update local stable control instruction is generated; Each intelligent agent periodically obtains the periodic time sequence operation data in the normal operation period and continuously performs dynamic echo state vector update, comprehensive time sequence state information construction and local stable control instruction generation.

2. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the graph data comprises: Assigning a unique node index to each operating node of the power distribution network, and arranging all node indexes in a fixed order to form a fixed node index sequence; According to the fixed node index sequence, the connection relationship between any two nodes is mapped to an adjacency relationship matrix according to the corresponding relationship of the node index; According to the fixed node index sequence, all node pairs connected by lines are recorded in the order of node index as an edge set; Organize the fixed node index sequence, the adjacency relationship matrix and the edge set, and associate the voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence of each node with the corresponding node index to form the graph data.

3. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the structured input sequence comprises: Normalizing the voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence; According to the fixed node index sequence in the graph data, the normalized voltage amplitude sequence, phase angle sequence, active power sequence and reactive power sequence of each node are arranged in the order of fixed node index to form the time sequence operation input structure of the corresponding node; According to the fixed node index sequence, the input weight is mapped to the time sequence operation of the node in the order of node index; According to the adjacency relationship matrix, the reservoir pool weight is constructed so that any position in the reservoir pool is only connected to the neighboring nodes of the corresponding node through weight; According to the preset internal decay coefficient, input mapping coefficient, normalization coefficient and maximum change limit coefficient, the state update parameters are set to enable the state update process to perform time evolution under topological constraints within the reservoir pool. The voltage amplitude sequence, the phase angle sequence, the active power sequence and the reactive power sequence arranged in time sequence and normalized are sequentially input to the improved graph echo state network of the corresponding agent in node index order to form a structured input sequence of the improved graph echo state network.

4. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the dynamic echo state vector includes: The structured input sequence is input to the improved graph echo state network of the corresponding agent in fixed node index order; According to the input weight, the structured input sequence at the current time is mapped to the corresponding position in the reservoir pool in fixed node index order, so that the corresponding position obtains the input driving quantity at the current time; According to the reservoir pool weight, the reservoir pool state at the last time is attenuated according to the internal attenuation coefficient, and then the attenuated reservoir pool state and the reservoir pool state at the last time of all the adjacent nodes having a connection relationship with the attenuated reservoir pool state are superimposed according to the connection relationship represented by the adjacency relationship matrix to form an intermediate state at the current time; According to the state update parameter, the input driving quantity and the intermediate state are added in a fixed proportion to obtain the final state of the reservoir pool at the current time, and the final state is arranged in fixed node index order to form a dynamic echo state vector.

5. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the comprehensive time sequence state information includes: After each agent completes the generation of the dynamic echo state vector at the corresponding node, the dynamic echo state vector of the node at the current time is structurally arranged, and each agent sends the arranged dynamic echo state vector of the node to all the adjacent agents having a connection relationship with the agent; Each agent receives the dynamic echo state vectors of the adjacent nodes generated at the current time from all the adjacent agents having a connection relationship with the agent, and preliminarily collects the received dynamic echo state vectors of the adjacent nodes, and arranges the preliminarily collected dynamic echo state vectors of the adjacent nodes in a pre-set index order according to the fixed node index order; Each agent checks the structure of the arranged dynamic echo state vectors of the adjacent nodes, so that each dynamic echo state vector of the adjacent nodes is consistent with the dynamic echo state vector of the node in length, order and node order identification; Each agent splices the dynamic echo state vector of the node and all the dynamic echo state vectors of the adjacent nodes arranged and checked according to a fixed splicing format; Each agent arranges the spliced data structure for consistency, sets the arranged spliced structure as the comprehensive time sequence state information at the current time, and stores the comprehensive time sequence state information in the internal time sequence information buffer.

6. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the local stability control instruction includes: Each agent acquires the comprehensive time sequence state information at the current time, and structurally analyzes the dynamic echo state vector of the node and the dynamic echo state vectors of the adjacent nodes in the comprehensive time sequence state information to obtain the structurally analyzed comprehensive time sequence state information; Each agent extracts the voltage amplitude change feature, the active power change feature, the reactive power change feature and the node association relationship change feature of the node at the current time according to the structurally analyzed comprehensive time sequence state information, and combines the change features into a time sequence running change set of the node in a fixed order; The intelligent agent extracts voltage amplitude variation characteristics, active power variation characteristics, reactive power variation characteristics and node association relationship variation characteristics of all adjacent nodes at the current time according to the integrated time sequence state information after structure analysis, and arranges the variation characteristics into an adjacent time sequence operation variation set in the order of fixed node index; Each intelligent agent fuses the time sequence operation variation set of the node with the adjacent time sequence operation variation set according to a fixed combination format, so that the fused time sequence operation variation structure can reflect the local topological structure relationship and the synchronous time sequence operation state of the node and the adjacent nodes; Each intelligent agent generates a node-level adjustment amount according to the fused time sequence operation variation structure; Each intelligent agent sets the node-level adjustment amount as a local stability control instruction at the current time, and executes the local stability control instruction through the corresponding execution device of the node.

7. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the disturbance updated local stability control instruction includes: When the intelligent agent detects new energy output fluctuation, load change or line disturbance, the intelligent agent obtains disturbance time sequence operation data of the node, and arranges the disturbance time sequence operation data into a disturbance update input sequence in the order of fixed node index; The disturbance update input sequence is input into the improved graph echo state network, and a disturbance update reserve pool state is formed; The disturbance update reserve pool state is arranged in the order of fixed node index to form a disturbance update dynamic echo state vector of the node, which is sent to all adjacent intelligent agents having a connection relationship with the node, and all adjacent node disturbance update dynamic echo state vectors generated at the same disturbance moment are received; All adjacent node disturbance update dynamic echo state vectors are arranged in the order of fixed node index; The disturbance update dynamic echo state vector of the node is spliced with the arranged adjacent node disturbance update dynamic echo state vectors in a fixed combination format to form disturbance update integrated time sequence state information; According to the disturbance update integrated time sequence state information, voltage amplitude variation characteristics, active power variation characteristics, reactive power variation characteristics and node association relationship variation characteristics of the node and the adjacent nodes at the disturbance moment are extracted, and the variation characteristics are arranged into a disturbance update operation amount set; According to the disturbance update operation amount set, a node-level adjustment amount is generated, which is used as a disturbance update local stability control instruction, and the disturbance update local stability control instruction is executed through the corresponding execution device.

8. The multi-agent based power distribution network stability control method of claim 1, wherein, The generation of the local stability control instruction includes: When the intelligent agent detects new energy output fluctuation, load change or line disturbance, the intelligent agent obtains disturbance time sequence operation data of the node, and arranges the disturbance time sequence operation data into a disturbance update input sequence in the order of fixed node index; Each intelligent agent inputs the periodic update input sequence into the improved graph echo state network, and forms a periodic update reserve pool state, and arranges the periodic update reserve pool state in the order of fixed node index to form a periodic update dynamic echo state vector of the node; Each intelligent agent sends the periodic update dynamic echo state vector of the node to all adjacent intelligent agents having a connection relationship with the node, and receives periodic update dynamic echo state vectors of all adjacent nodes; The intelligent agent arranges all the neighborhood period update dynamic echo state vectors in order of fixed node index, and splices the arranged neighborhood period update dynamic echo state vectors and the node period update dynamic echo state vector according to a fixed combination format; The intelligent agent extracts the voltage amplitude change characteristics, active power change characteristics, reactive power change characteristics and node association relationship change characteristics of the node and the neighborhood nodes according to the period update comprehensive time sequence state information, and arranges the extracted change characteristics into a period update operation quantity set; The intelligent agent generates a period update node-level adjustment quantity according to the period update operation quantity set, takes the period update node-level adjustment quantity as a period update local stability control instruction, and executes the period update local stability control instruction through a corresponding execution device; After the execution of the period update local stability control instruction is completed, the intelligent agent distinguishes the period update input sequence, the period update reserve pool state, the period update dynamic echo state vector, the period update comprehensive time sequence state information and the period update node-level adjustment quantity from the corresponding contents generated in the disturbance update process, so that the period update control chain and the disturbance update control chain are independent in structure and can cooperatively maintain overall stable operation in a double-track manner during the operation of the power distribution network.

Images