A physical guidance-based communication and network coordination partitioning method

Abstract

The application provides a physical guidance-based communication and network coordination partitioning method, and belongs to the technical field of power system optimization, and comprises the following steps: when the physical layer state of a coupled system changes, determining a physical layer change event based on a pre-constructed power distribution network physical layer complex network model; determining an optimal cross-domain transmission mapping based on the physical layer change event and a pre-constructed dual-domain coupled complex network model; evolving a current communication layer topology based on the optimal cross-domain transmission mapping and a pre-constructed sensing coordination communication layer complex network model to obtain an evolved communication layer structure; and converting topology reconstruction events and the like into communication layer structure adjustment bases directly by introducing the cross-domain optimal transmission mapping based on the physical layer state change, so that the physical layer can actively guide and dynamically link the communication layer, and the communication topology can evolve in real time and self-adaptively according to the physical state, and the dual-domain structure can be kept in highly coordinated matching.

Description

Technical Field

[0001] This invention belongs to the field of power system optimization technology, and specifically relates to a physical-guided communication and distribution network collaborative partitioning method. Background Technology

[0002] With the continuous advancement of new power system construction, the penetration rate of distributed photovoltaic, wind power, energy storage, flexible loads, and edge control terminals in active distribution networks is constantly increasing. The operation mode of distribution networks is evolving from the traditional unidirectional radial network to a complex system with deep coupling of power and information. In the process of achieving autonomous control, fault self-healing, regional coordinated regulation, and flexible resource optimization, active distribution networks increasingly rely on communication networks to provide status acquisition, control command transmission, edge coordination, and sensing support capabilities. Therefore, the coupling and partitioning of the distribution physical network and the communication network has become a key issue in improving system operating efficiency and robustness.

[0003] Existing active distribution network zoning methods mostly rely on electrical topology, power flow distribution, switch connections, or electrical sensitivity for zoning. They typically assume the communication network is a static support system and lack a mechanism to proactively drive communication layer structural reconfiguration based on changes in the physical layer's operational status. In scenarios involving topology reconfiguration, fault isolation, source-load fluctuations, and changes in control boundaries, the communication layer's link organization, node affiliation, master-slave control structure, and boundary interfaces cannot be dynamically adjusted in a timely manner with changes in the physical layer. This leads to a mismatch between the communication and physical layers, affecting the effectiveness of autonomous zoning control. Summary of the Invention

[0004] To address the problems in the background technology, this invention proposes a physical guidance-based collaborative partitioning method for communication and distribution networks.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A physical-guided collaborative partitioning method for communication and distribution networks includes: When the physical layer state of the coupled system changes, the physical layer change event is determined based on the pre-built complex network model of the distribution network physical layer. Determine the optimal cross-domain transport mapping based on the physical layer change events; Based on the optimal cross-domain transmission mapping and the pre-constructed complex network model of the synesthetic and cooperative communication layer, the current communication layer topology is evolved to obtain the evolved communication layer structure. Obtain the initial physical layer structure, and based on the evolved communication layer structure, divide the physical nodes and communication nodes of the coupled system to obtain the initial partition.

[0006] Preferably, the determination of physical layer change events based on a pre-constructed complex network model of the distribution network physical layer specifically includes: Obtain the current physical layer node state and the previous physical layer node state, and determine the physical layer state increment; Based on the physical layer state increment, the intensity of the physical layer change is determined, and its expression is as follows: In the formula, Indicates the intensity of physical layer changes. The change extracts the weight matrix. Denotes the Euclidean norm. Indicates the state increment; Determine whether the intensity of the physical layer change is greater than or equal to a preset change threshold. If so, determine the physical layer change event based on the intensity of the physical layer change.

[0007] Preferably, determining the optimal cross-domain transport mapping based on the physical layer change event specifically includes: Based on the physical layer change events and the pre-constructed dual-domain coupled complex network model, the event guidance cost is determined. Based on the event guidance cost and the predetermined importance distribution constraints of physical nodes and communication nodes, the optimal cross-domain transmission mapping is determined.

[0008] Preferably, determining the event guidance cost based on the physical layer change event and the pre-constructed dual-domain coupled complex network model specifically includes: Obtain the original importance of the physical nodes, and determine the normalized importance of the physical nodes based on the original importance of the physical nodes; Obtain the original importance of the communication nodes, and determine the normalized importance of the communication nodes based on the original importance of the communication nodes; Based on the physical layer change events, determine the local event characteristics of the physical nodes; Based on a pre-built dual-domain coupled complex network model, the coupling strength between physical nodes and communication nodes, the spatial distance between physical nodes and communication nodes, and the maximum spatial distance between physical nodes and communication nodes are determined. Based on the normalized importance of physical nodes, the normalized importance of communication nodes, the coupling strength between physical and communication nodes, the spatial distance between physical and communication nodes, the maximum spatial distance between physical and communication nodes, the local event characteristics of physical nodes, and the pre-determined service demand characteristics of communication nodes, the event guidance cost is determined, and its expression is:

[0009] In the formula, These are the cost combination coefficients, This represents the spatial distance between physical nodes and communication nodes. This represents the maximum spatial distance between physical nodes and communication nodes. Represents the local event characteristics of physical nodes. This indicates the service demand characteristics of communication nodes.

[0010] Preferably, the expression for the importance distribution constraint of physical nodes and communication nodes is: In the formula, dimension A column vector of all 1s. dimension A column vector of all 1s. and These represent the importance distribution vectors of physical nodes and communication nodes, respectively; The expression for determining the optimal cross-domain transport mapping is:

[0011] In the formula, Indicates time The optimal cross-domain transport mapping, This indicates that the transfer mapping matrix takes values ​​within the feasible region defined by the importance distribution constraint. Represents the total number of physical nodes. Indicates the total number of communication nodes. Represents the physical node at time t and The structural distance between them represents the communication nodes at time t. and The structural distance between them represents the physical nodes. Communication Node The mapping weights represent the physical nodes. To communication node The mapping weights represent the physical nodes. Mapped to communication node The cost of guiding events.

[0012] Preferably, the step of evolving the current communication layer topology based on the optimal cross-domain transmission mapping and the pre-constructed complex network model of the sensory cooperative communication layer to obtain the evolved communication layer structure specifically includes: Based on a pre-constructed complex network model of the synergistic communication layer, the hidden state of the communication layer at the previous time step is determined. Based on the previous hidden state of the communication layer, the local event characteristics of the physical layer, and the optimal cross-domain transmission mapping, the current hidden state of the communication layer is determined; wherein, the local event characteristics of the physical layer are determined by the local event characteristics of the physical nodes; Obtain the communication connection strength at the previous moment, and determine the current communication connection strength based on the communication connection strength at the previous moment, the current hidden state of the communication layer, and the pre-determined spatial distance between communication nodes and the maximum spatial distance between communication nodes; Based on the current communication connection strength, the evolved communication layer structure is obtained.

[0013] Preferably, the step of obtaining the initial physical layer structure and, based on the evolved communication layer structure, dividing the physical nodes and communication nodes of the coupled system to obtain an initial partition specifically includes: Obtain the initial physical layer structure, and determine the physical layer partition modularity based on the initial physical layer structure; Based on the evolved communication layer structure, the modularity of the communication layer partition is determined; The initial partition is determined based on the physical layer partition modularity, the communication layer partition modularity, and the predetermined constraints.

[0014] Preferably, the constraints include: partition balancing constraints and time smoothing constraints; The expression for determining the initial partition is:

[0015] In the formula, Describes the joint modularity objective function. These represent the weight coefficients for physical layer modularity, communication layer modularity, and cross-domain consistency modularity, respectively. This represents the equilibrium penalty coefficient. This represents the time smoothing penalty coefficient. Indicates cross-domain consistency modularity. Represents the sum of the elements of the optimal transfer matrix. This indicates a partition balancing constraint. This indicates a time smoothing constraint.

[0016] Preferably, the method further includes: correcting the initial partition to obtain the final partition.

[0017] Preferably, the step of correcting the initial partition to obtain the final partition specifically includes: Traverse all edges in the communication layer and filter out candidate boundary edges; Calculate the curvature of each candidate boundary edge; Based on the curvature of each candidate boundary edge, the partition boundaries are corrected to obtain the final partition.

[0018] The beneficial effects of this invention are: The method of this invention introduces cross-domain optimal transmission mapping based on physical layer state changes, which directly transforms events such as topology reconstruction, fault isolation, source-load fluctuations, and control boundary changes into the basis for communication layer structure adjustment. This enables the physical layer to actively guide and dynamically link the communication layer, allowing the communication topology to adaptively evolve with the physical state in real time. This ensures that the dual-domain structure maintains a high degree of coordination and matching, thereby overcoming the problems of passive adaptation, static lag, and mismatch between partitioning results and physical operating states in traditional methods. This significantly improves the adaptive adjustment capability, real-time control performance, and operational robustness of the coupled system.

[0019] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description and the drawings. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart of a physical-guided communication and distribution network collaborative partitioning method according to the present invention is shown; Figure 2 A schematic diagram of the communication and distribution network structure of the present invention is shown. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Reference Figure 1 As shown, a dual-domain collaborative partitioning method for communication and distribution networks based on physical layer changes actively drives the process, specifically including the following steps: S10. When the physical layer state of the coupled system changes, the physical layer change event is determined based on the pre-built complex network model of the distribution network physical layer. S20. Determine the optimal cross-domain transport mapping based on the physical layer change events; S30. Based on the optimal cross-domain transmission mapping and the pre-constructed complex network model of the synesthetic and cooperative communication layer, the current communication layer topology is evolved to obtain the evolved communication layer structure. S40. Obtain the initial physical layer structure, and based on the evolved communication layer structure, divide the physical nodes and communication nodes of the coupled system to obtain the initial partition; S50. Correct the initial partition to obtain the final partition.

[0024] By introducing cross-domain optimal transmission mapping based on physical layer state changes, events such as topology reconstruction, fault isolation, source-load fluctuations, and control boundary changes are directly transformed into the basis for communication layer structure adjustment. This enables the physical layer to actively guide and dynamically link the communication layer, allowing the communication topology to adaptively evolve with the physical state in real time. This ensures that the dual-domain structure maintains a high degree of coordination and matching, thereby overcoming the problems of passive adaptation, static lag, and mismatch between partitioning results and physical operating states in traditional methods. This significantly improves the adaptive adjustment capability, real-time control performance, and operational robustness of the coupled system.

[0025] Within the framework of complex network theory, time-varying networks can be characterized by a set of nodes, a set of edges, a weighted adjacency matrix, and a node state feature matrix. The node set represents various entities in the system, the edge set reflects the topological relationships between entities, the weighted adjacency matrix quantifies the coupling strength between nodes, and the node state feature matrix describes the operational attributes and dynamic changes of each entity.

[0026] To address the time-varying, coupled, and dynamically evolving characteristics of active distribution networks, this invention first constructs a complex physical layer network model of the time-varying active distribution network to fully characterize the network's topology, electrical coupling relationships, and operating state. The time variable... The model can be introduced to characterize the dynamic changes in network structure and node state under scenarios such as topology reconstruction and source load fluctuation.

[0027] Therefore, in step S10 above, the complex network model of the distribution network physical layer is in The expression for time is:

[0028] In the formula, express The complex physical layer of the power distribution network is constantly being actively managed. Represents the set of physical nodes. Represents the total number of physical nodes. Represents the set of physical edges. Represents the physical layer weighted adjacency matrix. Represents the physical node state feature matrix. This represents the dimension of the physical node state features.

[0029] Furthermore, the physical layer edge weights comprehensively reflect the tightness of electrical connections between nodes. By weighting and fusing line admittance, power flow correlation, and equivalent electrical distance, the strength of electrical coupling can be more accurately characterized. Its expression is:

[0030] In the formula, express physical node at time and The overall electrical coupling weight between them These are the physical layer edge weight combination coefficients, and they satisfy... , Represents a node and In-line admittance, Indicates the line admittance amplitude. Represents a node and At any moment The correlation coefficient of the trend, Indicates the equivalent electrical distance. This indicates the avoidance of extremely small positive numbers with a denominator of zero.

[0031] The reliable operation of an active distribution network highly depends on the transmission support and sensing coverage capabilities of the communication network. The communication network also possesses time-varying and dynamic reconfiguration characteristics. To achieve precise matching with the physical layer, the communication layer model, based on the physical layer, further introduces sensing coverage relationships to characterize the service and data collection range of communication nodes to physical nodes. Therefore, in step S30 above, the complex network model of the sensing-cooperative communication layer... The expression for time is:

[0032] In the formula, Indicates time Complex networks at the communication layer Represents a set of communication nodes. Indicates the total number of communication nodes. Represents a set of communication links. Represents the weighted adjacency matrix of the communication layer. Represents the state feature matrix of the communication node. The dimension of the communication node state features. This represents the set of perception coverage relationships.

[0033] Furthermore, the communication layer edge weights need to comprehensively consider transmission performance and perception coordination, and unify the modeling of bandwidth, reliability, latency, and perception overlap to fully reflect the cooperative service capabilities of the communication link. Its expression is:

[0034] In the formula, Indicates time Next communication node and Comprehensive communication and coordination weights between them These are the edge weight combination coefficients of the communication layer, satisfying... ; Indicates link bandwidth. Indicates the maximum available bandwidth. Indicates link reliability. Indicates link latency. Indicates the maximum allowable delay. Represents the perceived coordination coefficient. and Representing communication nodes , At any moment The set of physical nodes covered or served. This indicates the avoidance of extremely small positive numbers with a denominator of zero.

[0035] Reference Figure 2 As shown, after constructing the complex network model of the distribution network physical layer and the complex network model of the sensor-coordinated communication layer, it is necessary to unify the two into a whole. Therefore, in step S20 above, the expression of the dual-domain coupled complex network model is:

[0036] In the formula, Indicates time A complex network model with dual-domain coupling. This represents the fundamental coupling matrix between physical nodes and communication nodes. Represents physical nodes With communication nodes The coupling strength, Let be the coupling combination coefficients, satisfying , Indicates an overlay relationship. Indicates the control or data acquisition relationship. Indicates spatial proximity.

[0037] By constructing a complex dual-domain network model that unifies and couples the power distribution physical layer, communication layer, and sensing coverage relationship, a unified representation of the structural, service, and spatial relationships between power networks and communication networks is achieved.

[0038] Furthermore, in step S10 above, based on the pre-constructed complex network model of the distribution network physical layer, physical layer change events are determined, specifically including the following steps: S101. Obtain the current physical layer node state and the previous physical layer node state, and determine the physical layer state increment. Specifically, obtain the current moment from the coupled system. The physical layer node state and the previous time step The physical layer node state. The expression for calculating the physical layer state increment is: In the formula, Indicates the sampling time interval. Indicates the state increment. Indicates the current time The physical layer global state vector is composed of the physical node state feature matrix. After vectorization, This represents the global state vector of the physical layer at the previous moment.

[0039] S102. Determine the intensity of physical layer changes based on physical layer state increments; Specifically, based on the physical layer state increment, the overall intensity of physical layer changes can be quantified, thereby enabling the identification of physical layer change events. The expression for calculating the intensity of physical layer changes is as follows: In the formula, Indicates the intensity of physical layer changes. The change extracts the weight matrix. This represents the Euclidean norm.

[0040] S103. Determine whether the intensity of the physical layer change is greater than or equal to the preset change threshold. If so, determine the physical layer change event based on the intensity of the physical layer change. Specifically, when the following conditions are met When this occurs, a significant change event is determined to have occurred at the physical layer, thus requiring the identification of the physical layer change event. Among these, This indicates the preset threshold for change.

[0041] Furthermore, the expression for physical layer change events is determined as follows: In the formula, Represents a vector of physical layer change events. Represents the event encoding matrix, This represents the dimension of the event features.

[0042] In step S20 above, determining the optimal cross-domain transport mapping based on the physical layer change event specifically includes the following steps: S201. Based on the physical layer change events and the pre-constructed dual-domain coupled complex network model, determine the event guidance cost; S2011. Based on physical layer change events, determine the local event characteristics of physical nodes; Physical layer changes can alter the local event characteristics of physical nodes, which in turn can affect the construction of the event guidance cost matrix. Different changes correspond to different event feature vectors, which can lead to differences in the feature matching costs between physical nodes and different communication nodes. This can change the solution of the optimal cross-domain transmission mapping, and ultimately guide the evolution of the communication topology and the partitioning of the physical and communication layers.

[0043] S2012. Obtain the original importance of physical nodes and the original importance of communication nodes, and determine the normalized importance of physical nodes and the normalized importance of communication nodes based on the original importance of physical nodes and the original importance of communication nodes, respectively. Specifically, the expression for calculating the normalized importance of physical nodes is: In the formula, Represents physical nodes Normalized importance Indicates the original importance of the physical node. This represents the sum of the original importance of all physical nodes.

[0044] Specifically, the expression for calculating the normalized importance of communication nodes is: In the formula, Represents communication node Normalized importance Indicates the original importance of the communication node. This represents the sum of the original importance of all communication nodes.

[0045] S2013. Based on a pre-built dual-domain coupled complex network model, determine the coupling strength between physical nodes and communication nodes, the spatial distance between physical nodes and communication nodes, and the maximum spatial distance between physical nodes and communication nodes. The coupling strength between physical nodes and communication nodes is calculated by the basic coupling matrix in the dual-domain coupled complex network model. The spatial distance between physical nodes and communication nodes is calculated by Euclidean distance from the node location information in the dual-domain coupled complex network model. The maximum spatial distance between physical nodes and communication nodes is determined by the farthest communication service radius pre-set by the dual-domain coupled complex network model.

[0046] S2014. Based on the normalized importance of physical nodes, the normalized importance of communication nodes, the coupling strength between physical nodes and communication nodes, the spatial distance between physical nodes and communication nodes, the maximum spatial distance between physical nodes and communication nodes, the local event characteristics of physical nodes, and the pre-determined service demand characteristics of communication nodes, determine the event guidance cost. Specifically, an event-guided cost from the physical layer to the communication layer is constructed based on the coupling strength between physical nodes and communication nodes, normalized importance, predetermined spatial distance, local event characteristics of physical nodes, and service demand characteristics of communication nodes. This cost is used to quantify the matching degree of cross-domain mapping. More specifically, the coupling strength between physical nodes and communication nodes reflects the inherent closeness of their relationship; normalized importance ensures that physical nodes and communication nodes with similar importance are matched, improving the rationality of the mapping; spatial distance reflects the actual accessibility of communication support and transmission cost constraints; local event characteristics of physical nodes reflect the current state and scope of influence of the physical layer; and service demand characteristics of communication nodes characterize the support capabilities and service positioning of the communication side. These factors, from the four levels of static association, importance balance, spatial constraints, and dynamic event adaptation, constitute a complete evaluation basis for the mapping cost, thus enabling a comprehensive and accurate quantification of the matching degree of mapping from the physical layer to the communication layer.

[0047] The expression for calculating the event-guided cost is:

[0048] In the formula, These are the cost combination coefficients, This represents the spatial distance between physical nodes and communication nodes. This represents the maximum spatial distance between physical nodes and communication nodes. Represents the local event characteristics of physical nodes. This indicates the service requirement characteristics of the communication node. These characteristics reflect the types of services and support capabilities that the communication node can provide, and are pre-defined based on control requirements.

[0049] S202. Based on the event guidance cost and the predetermined importance distribution constraints of physical nodes and communication nodes, determine the optimal cross-domain transmission mapping; Specifically, the structural distance between physical nodes characterizes the degree of topological association and electrical coupling between physical layer nodes, and is calculated using the physical layer weighted adjacency matrix; the structural distance between communication nodes characterizes the link connectivity and cooperative service strength between communication layer nodes, and is calculated using the communication layer weighted adjacency matrix. Under the constraints of event guidance cost and structural distance, the optimal cross-domain transmission mapping from the physical domain to the communication domain is obtained, thereby enabling adaptive matching of dual-domain topological features.

[0050] The cross-domain transmission matrix is ​​used to quantify the mapping relationship between physical nodes and communication nodes, and its expression is: In the formula, Represents physical nodes To communication node Mapping weights.

[0051] The optimal cross-domain transmission mapping feasible region is used to limit the reasonable range of the mapping, ensuring that the mapping process satisfies the importance distribution constraints of physical nodes and communication nodes. Its expression is: In the formula, dimension A column vector of all 1s; dimension A column vector of all 1s; and These represent the importance distribution vectors of physical nodes and communication nodes, respectively.

[0052] The expression for calculating the optimal cross-domain transport mapping is:

[0053] In the formula, Indicates time The optimal cross-domain transport mapping, This indicates that the transfer mapping matrix takes values ​​within the feasible region defined by the importance distribution constraint. Represents the total number of physical nodes. Indicates the total number of communication nodes. Represents the physical node at time t and The structural distance between them represents the communication nodes at time t. and The structural distance between them represents the physical nodes. Communication Node The mapping weights represent the physical nodes. To communication node The mapping weights represent the physical nodes. Mapped to communication node The event-guided cost. Further, the first term represents the Gromov-Wasserstein structural mismatch cost, the second term represents the event-guided cost term, and the third term represents the entropy regularization term. Indicates the weight of the event guide item. This represents the entropy regularity coefficient.

[0054] S30. Based on the optimal cross-domain transmission mapping and the pre-constructed complex network model of the synesthetic and cooperative communication layer, the current communication layer topology is evolved to obtain the evolved communication layer structure. S301. Based on a pre-constructed complex network model of the synergistic communication layer, determine the hidden state of the communication layer at the previous time step. Specifically, the complex network model of the sensor-cooperative communication layer includes communication node states, link connection relationships, and node characteristic information. Based on this model, the structure, dimensions, and previous hidden state values ​​of the communication layer hidden state matrix can be directly extracted and determined. The communication layer hidden state is used to comprehensively characterize the operational status, service capabilities, and topological association characteristics of each communication node. The expression for the communication layer hidden state matrix is:

[0055] In the formula, This represents the hidden state matrix of the communication layer. Represents communication node The hidden state vector. This represents the dimension of the hidden state.

[0056] The hidden state matrix of the communication layer at the previous moment can be determined from the hidden state matrix of the communication layer. This serves as the initial basis for updating the hidden state at the current moment.

[0057] S302. Based on the hidden state of the communication layer at the previous moment, the local event characteristics of the physical layer, and the optimal cross-domain transmission mapping, determine the hidden state of the communication layer at the current moment. Among them, the physical layer local event features are determined by the physical node local event features; the physical layer local event feature matrix is ​​composed of the combination of all physical node local event features.

[0058] Specifically, the hidden state is dynamically updated using a continuous-time evolution equation at the communication layer. This is achieved by integrating topology diffusion characteristics, physical layer event-driven mechanisms, optimal transmission mapping guidance, and the influence of the communication node's own state, thus obtaining the current hidden state at the communication layer. The specific expression for the continuous-time evolution equation at the communication layer is as follows:

[0059] In the formula, This represents the rate of change of the hidden state in the communication layer. Represents the topological diffusion coefficient. Represents the Laplace matrix of the communication layer graph. Represents a diagonal matrix of communication layer degrees. Represents the physical layer drive gain coefficient. This represents the transpose of the optimal transfer matrix. Represents the physical layer local event feature matrix. The parameter matrix representing the mapping from events to hidden states. Represents the state-driving coefficients of the communication node itself. Represents the feature mapping parameter matrix of communication nodes. This represents the hidden state decay coefficient.

[0060] The hidden state of the communication layer at the current moment can be obtained by numerical integration using the continuous-time evolution equation of the communication layer.

[0061] S303. Obtain the communication connection strength of the previous moment, and determine the current communication connection strength based on the communication connection strength of the previous moment, the hidden state of the communication layer at the current moment, and the pre-determined spatial distance between communication nodes and the maximum spatial distance between communication nodes. The spatial distance between communication nodes and the maximum spatial distance are both determined by a pre-constructed complex network model of the sensory collaborative communication layer. The spatial distance between communication nodes is calculated using Euclidean distance based on the node location information within the complex network model of the sensory collaborative communication layer, while the maximum spatial distance is directly determined by the communication service radius pre-set by the complex network model of the sensory collaborative communication layer.

[0062] Specifically, based on the communication connection strength at the previous moment, and combining the hidden state interactions of nodes with spatial distance constraints, the current connection strength between communication nodes is calculated, and its expression is: In the formula, Represents communication node and The connection strength between them This represents the fusion coefficient between the original topology and the evolved topology. This represents the Sigmoid function. This represents the hidden state interaction matrix. Indicates the link maintenance cost weight. This represents the cost of link reconfiguration.

[0063] S304. Based on the current communication connection strength, obtain the evolved communication layer structure; Specifically, the current communication connection strengths between all communication nodes are organized in matrix form to form the evolved communication layer structure matrix, the expression of which is:

[0064] In the formula, This represents the evolved communication layer structure matrix.

[0065] By establishing a continuous-time evolution equation for the communication layer, the affiliation of communication nodes, link connections, and local topology can be dynamically updated as the physical layer state changes, thereby improving the communication layer structure's responsiveness to changes in the operating state of the active distribution network.

[0066] S40. Obtain the initial physical layer structure, and based on the evolved communication layer structure, divide the physical nodes and communication nodes of the coupled system to obtain the initial partition; S401. Obtain the initial physical layer structure and determine the physical layer partition modularity based on the initial physical layer structure; The expression for the physical layer partition modularity is:

[0067] In the formula, Represents the physical layer modularity matrix. This represents the weighted degree vector of the physical layer nodes. Represents the total edge weight of the physical layer. Indicates the modularity of the physical layer partition. Represents the matrix trace operation. This represents the physical layer partitioning indicator matrix. Furthermore, for any physical node... Its sum of membership degrees in all K partitions is 1.

[0068] S402. Based on the evolved communication layer structure, determine the modularity of the communication layer partitions; The expression for the modularity of the communication layer partition is:

[0069] In the formula, Represents the module degree matrix of the communication layer. This represents the weighted degree vector of the communication layer nodes. Indicates the total edge weight of the communication layer. Indicates the modularity of the communication layer partition. This represents the communication layer partitioning indicator matrix. Furthermore, for any communication node... Its sum of membership degrees in all K partitions is 1.

[0070] S403. Based on the physical layer partition modularity, the communication layer partition modularity, and predetermined constraints, an initial partition is obtained; wherein, the constraints include: partition balancing constraints and time smoothing constraints. Specifically, the initial partition is determined based on the joint modularity objective function, the expression of which is:

[0071] In the formula, Describes the joint modularity objective function. These represent the weight coefficients for physical layer modularity, communication layer modularity, and cross-domain consistency modularity, respectively. This represents the equilibrium penalty coefficient. This represents the time smoothing penalty coefficient. Indicates cross-domain consistency modularity. Represents the sum of the elements of the optimal transfer matrix. This indicates a partition balancing constraint. This indicates a time smoothing constraint.

[0072] By constructing a joint modularity objective function, physical layer modularity, communication layer modularity, and cross-domain consistency are integrated into the same optimization framework, achieving global coordinated optimization of dual-domain collaborative partitioning.

[0073] As a preferred embodiment of the present invention, the above method further includes: S50, correcting the initial partition to obtain the final partition.

[0074] S501. Traverse all edges in the communication layer and filter out candidate boundary edges; Specifically, based on the initial partition membership information of the communication nodes, a boundary determination threshold is set, and communication links whose partition membership product falls within the threshold are selected to form a candidate boundary edge set, the expression of which is: In the formula, Represents the set of candidate boundary edges. and Representing communication nodes , The partition membership degree row vector, The threshold for boundary determination, and .

[0075] S502. Calculate the curvature of each candidate boundary edge; For any candidate boundary edge Calculate its Ollivier-Ricci curvature Its expression is: In the formula, Represents the discrete Ricci curvature of the candidate boundary edge. Represents the first-order Wasserstein distance. and Representing communication nodes , The probability measure constructed from the neighborhood weights. Represents a node and The shortest path distance.

[0076] The expression for calculating the probability measure is as follows:

[0077] In the formula, Indicates from node to its neighboring nodes The probability quality of the allocation, express At this moment, communication nodes With nodes The evolution of communication connection weights between them Represents a node The sum of the connection weights between it and all its neighboring nodes after evolution.

[0078] S503. Based on the curvature of each candidate boundary edge, correct the partition boundaries to obtain the final partition; Specifically, the discrete Ricci curvature is used as the basis for consistency correction. A curvature consistency correction term is constructed and a joint objective function is introduced. The corrected objective function is maximized to complete the local optimization of the partition boundary. The final collaborative partitioning result of the physical layer and the communication layer is output according to the principle of maximum membership.

[0079] The expression for the curvature consistency correction term is:

[0080] The joint modularity objective function after introducing curvature correction is:

[0081] The final partitioning solution expression is:

[0082]

[0083]

[0084] In the formula, Represents physical nodes The final partition number, Represents communication node The final partition number, This indicates retrieving the partition index corresponding to the highest membership degree.

[0085] By introducing a discrete Ricci curvature local consistency correction mechanism on the candidate boundary edges, the accuracy of fuzzy boundary recognition and the stability of the partition structure are improved, ultimately enhancing the accuracy and engineering practicality of the dual-domain collaborative partitioning results.

[0086] Based on the same inventive concept as the above method, this invention also proposes a communication and distribution network collaborative partitioning system based on physical layer active guidance, comprising: The physical layer change determination module is used to determine physical layer change events based on a pre-built complex network model of the distribution network physical layer when the physical layer state of the coupled system changes. The optimal cross-domain transport mapping determination module is used to determine the optimal cross-domain transport mapping based on physical layer change events. The evolution module is used to evolve the current communication layer topology based on the optimal cross-domain transmission mapping and a pre-built complex network model of the synesthetic and collaborative communication layer, so as to obtain the evolved communication layer structure. The partitioning module is used to obtain the initial physical layer structure and, based on the evolved communication layer structure, divide the physical nodes and communication nodes of the coupled system to obtain the initial partitions.

[0087] Based on the same inventive concept as the above method, the present invention also proposes a device including a memory and a processor, wherein the memory stores computer instructions that can be executed on the processor, and the processor executes the above-described physical-guided communication and distribution network collaborative partitioning method when executing the computer instructions.

[0088] Based on the same inventive concept as the above method, the present invention also proposes a computer-readable storage medium storing computer instructions, which, when executed, can realize the above-mentioned physical guidance-based communication and distribution network collaborative partitioning method.

[0089] Any references to memory, storage, database, or other media used in the embodiments provided in this invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0090] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0091] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for collaborative partitioning of communication and distribution networks based on physical guidance, characterized in that, include: When the physical layer state of the coupled system changes, the physical layer change event is determined based on the pre-built complex network model of the distribution network physical layer. Determine the optimal cross-domain transport mapping based on the physical layer change events; Based on the optimal cross-domain transmission mapping and the pre-constructed complex network model of the synesthetic and cooperative communication layer, the current communication layer topology is evolved to obtain the evolved communication layer structure. Obtain the initial physical layer structure, and based on the evolved communication layer structure, divide the physical nodes and communication nodes of the coupled system to obtain the initial partition.

2. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 1, characterized in that, The determination of physical layer change events based on a pre-constructed complex network model of the distribution network physical layer specifically includes: Obtain the current physical layer node state and the previous physical layer node state, and determine the physical layer state increment; Based on the physical layer state increment, the intensity of the physical layer change is determined, and its expression is as follows: In the formula, Indicates the intensity of physical layer changes. The change extracts the weight matrix. Denotes the Euclidean norm. Indicates the state increment; Determine whether the intensity of the physical layer change is greater than or equal to a preset change threshold. If so, determine the physical layer change event based on the intensity of the physical layer change.

3. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 1, characterized in that, The determination of the optimal cross-domain transport mapping based on the physical layer change event specifically includes: Based on the physical layer change events and the pre-constructed dual-domain coupled complex network model, the event guidance cost is determined. Based on the event guidance cost and the predetermined importance distribution constraints of physical nodes and communication nodes, the optimal cross-domain transmission mapping is determined.

4. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 3, characterized in that, The determination of the event guidance cost based on the physical layer change event and the pre-constructed dual-domain coupled complex network model specifically includes: Obtain the original importance of the physical nodes, and determine the normalized importance of the physical nodes based on the original importance of the physical nodes; Obtain the original importance of the communication nodes, and determine the normalized importance of the communication nodes based on the original importance of the communication nodes; Based on the physical layer change events, determine the local event characteristics of the physical nodes; Based on a pre-built dual-domain coupled complex network model, the coupling strength between physical nodes and communication nodes, the spatial distance between physical nodes and communication nodes, and the maximum spatial distance between physical nodes and communication nodes are determined. Based on the normalized importance of physical nodes, the normalized importance of communication nodes, the coupling strength between physical and communication nodes, the spatial distance between physical and communication nodes, the maximum spatial distance between physical and communication nodes, the local event characteristics of physical nodes, and the pre-determined service demand characteristics of communication nodes, the event guidance cost is determined, and its expression is: In the formula, These are the cost combination coefficients, This represents the spatial distance between physical nodes and communication nodes. This represents the maximum spatial distance between physical nodes and communication nodes. Represents the local event characteristics of physical nodes. This indicates the service demand characteristics of communication nodes.

5. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 4, characterized in that, The expression for the importance distribution constraint of physical nodes and communication nodes is: In the formula, dimension A column vector of all 1s dimension A column vector of all 1s and These represent the importance distribution vectors of physical nodes and communication nodes, respectively; The expression for determining the optimal cross-domain transport mapping is: In the formula, Indicates time The optimal cross-domain transport mapping, This indicates that the transfer mapping matrix takes values ​​within the feasible region defined by the importance distribution constraint. Represents the total number of physical nodes. Indicates the total number of communication nodes. Represents the physical node at time t and The structural distance between them represents the communication nodes at time t. and The structural distance between them represents the physical nodes. Communication Node The mapping weights represent the physical nodes. To communication node The mapping weights represent the physical nodes. Mapped to communication node The cost of guiding events.

6. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 1, characterized in that, Based on the optimal cross-domain transmission mapping and the pre-constructed complex network model of the sensor-cooperative communication layer, the current communication layer topology is evolved to obtain the evolved communication layer structure, specifically including: Based on a pre-constructed complex network model of the synergistic communication layer, the hidden state of the communication layer at the previous time step is determined. Based on the previous hidden state of the communication layer, the local event characteristics of the physical layer, and the optimal cross-domain transmission mapping, the current hidden state of the communication layer is determined; wherein, the local event characteristics of the physical layer are determined by the local event characteristics of the physical nodes; Obtain the communication connection strength at the previous moment, and determine the current communication connection strength based on the communication connection strength at the previous moment, the current hidden state of the communication layer, and the pre-determined spatial distance between communication nodes and the maximum spatial distance between communication nodes; Based on the current communication connection strength, the evolved communication layer structure is obtained.

7. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 1, characterized in that, The process of obtaining the initial physical layer structure and, based on the evolved communication layer structure, dividing the physical nodes and communication nodes of the coupled system to obtain an initial partition specifically includes: Obtain the initial physical layer structure, and determine the physical layer partition modularity based on the initial physical layer structure; Based on the evolved communication layer structure, the modularity of the communication layer partition is determined; The initial partition is determined based on the physical layer partition modularity, the communication layer partition modularity, and the predetermined constraints.

8. The method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 7, characterized in that, The constraints include: partition balancing constraints and time smoothing constraints; The expression for determining the initial partition is: In the formula, Describes the joint modularity objective function. These represent the weight coefficients for physical layer modularity, communication layer modularity, and cross-domain consistency modularity, respectively. This represents the equilibrium penalty coefficient. This represents the time smoothing penalty coefficient. Indicates cross-domain consistency modularity. Represents the sum of the elements of the optimal transfer matrix. This indicates a partition balancing constraint. This indicates a time smoothing constraint.

9. A method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 1, characterized in that, The method further includes: correcting the initial partition to obtain the final partition.

10. A method for collaborative partitioning of communication and distribution networks based on physical guidance according to claim 9, characterized in that, The process of correcting the initial partition to obtain the final partition specifically includes: Traverse all edges in the communication layer and filter out candidate boundary edges; Calculate the curvature of each candidate boundary edge; Based on the curvature of each candidate boundary edge, the partition boundaries are corrected to obtain the final partition.

Images