Communication and power distribution collaborative low-interaction compression networking and partitioning method, device and medium

Abstract

This invention provides a method, device, and medium for low-interaction compressed networking and partitioning of communication and power distribution systems, belonging to the field of power system optimization technology. The method includes: acquiring basic data of the coupled system; constructing a complex network model of the distribution network physical layer based on distribution network physical node data, and constructing a complex network model of the communication layer based on communication node data and a cross-domain mapping matrix; partitioning the coupled system based on the distribution network physical layer complex network model and the communication layer complex network model to obtain initial partitions; when the coupled system experiences dynamic disturbances, generating candidate partitioning strategies for each partition using a stochastic stable game method, and obtaining intermediate partitioning schemes based on the candidate partitioning strategies and the initial partitions; verifying the intermediate partitioning schemes to obtain the final partitions; by using a stochastic stable game method to generate candidate partitioning strategies for each partition, stable strategy optimization can be carried out based on partition utility when the coupled system experiences dynamic disturbances.

Description

Technical Field

[0001] This invention belongs to the field of power system optimization technology, and specifically relates to methods, equipment and media for low-interaction compressed networking and partitioning of communication and power distribution. Background Technology

[0002] With the high proportion of distributed power sources, flexible loads, energy storage devices, and edge monitoring and control terminals integrated into distribution networks, the operation of these networks exhibits significant dynamism, randomness, and regional coupling. Simultaneously, communication networks undertake critical tasks such as uploading measurement information, synchronizing zone status, issuing control commands, and handling fault linkages. Their link latency, bandwidth usage, link reliability, and scheduling strategies directly impact distribution control performance. In scenarios involving coordinated control of communication networks and distribution networks, distribution zone partitioning is no longer merely a simple electrical topology division problem, but a dual-network collaborative optimization problem simultaneously influenced by electrical coupling, communication interaction burden, and control latency constraints.

[0003] In existing technologies, most communication and distribution collaborative zoning methods only focus on topology clustering around the electrical layer of the distribution network, with the communication layer merely serving as a passive adaptation link, failing to achieve deep collaborative optimization of electrical characteristics and communication performance. These methods have significant drawbacks in practical operation: when the number of cross-zone boundary nodes within the system is large, continuous state synchronization, boundary coordination, and control command interaction between numerous boundary nodes are required, directly leading to complex cross-zone communication interactions and excessive communication resource consumption; furthermore, in dynamic scenarios such as load fluctuations, changes in link quality, or migration of boundary control objects, the zoning topology is prone to frequent switching, resulting in repeated boundary jitter and frequent reconfiguration of control strategies, ultimately leading to insufficient system stability. Summary of the Invention

[0004] To address the problems in the background technology, this invention proposes a method, device, and medium for low-interaction compression networking and partitioning of communication and power distribution collaboration.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention proposes a method for low-interaction compressed networking and partitioning in coordination between communication and power distribution, comprising: S10. Obtain the basic data of the coupled system; wherein, the basic data includes distribution network physical node data, communication node data and cross-domain mapping matrix; S20. Construct a complex network model of the physical layer of the distribution network based on the physical node data of the distribution network, and construct a complex network model of the communication layer based on the communication node data and the cross-domain mapping matrix. S30. Based on the complex network model of the physical layer and the complex network model of the communication layer of the power distribution network, the coupled system is partitioned to obtain the initial partition; S40. When the coupled system experiences dynamic disturbances, a random stable game method is used to generate candidate partitioning strategies for each partition, and an intermediate partitioning scheme is obtained based on the candidate partitioning strategies and the initial partition. S50. Verify the intermediate partitioning scheme to obtain the final partition.

[0006] Preferably, the distribution network physical node data includes: distribution network physical node set, branch set, line admittance, equivalent electrical distance, physical node state characteristic matrix, and power flow correlation coefficient; The communication node data includes: a set of communication nodes, a set of communication links, communication link bandwidth, link latency, link reliability, a communication node state feature matrix, and cooperative service coefficients.

[0007] Preferably, the step of partitioning the coupled system based on the complex network model of the distribution network physical layer and the complex network model of the communication layer to obtain an initial partition specifically includes: S301. Based on the pre-constructed partitioning objective function, the coupled system is partitioned to obtain a preliminary partitioning scheme; S302. Based on the preliminary partitioning scheme, identify the boundary nodes and obtain the boundary node set; S303. Based on the set of boundary nodes, calculate the comprehensive importance score for each boundary node; S304. Based on the comprehensive importance score of each boundary node, select proxy nodes and construct a boundary aggregation matrix; S305. Based on the boundary aggregation matrix, compress the boundary nodes; S306. Based on the compressed boundary nodes, calculate the global deterministic time delay boundary of the preliminary partitioning scheme; S307. Determine whether the global deterministic delay boundary is greater than the preset delay threshold. If so, repeat steps S301-307 until an initial partition with a delay less than the preset delay threshold is obtained.

[0008] Preferably, the construction of the boundary aggregation matrix specifically includes: Based on the proxy nodes, determine the non-proxy nodes; Compress the non-proxy nodes to construct a boundary aggregation matrix.

[0009] Preferably, the step of generating candidate partitioning strategies for each partition using a stochastic stable game method specifically includes: A policy set is generated for each partition; wherein the policy set includes: boundary migration, proxy node update, and partition topology switching; Each strategy is evaluated using a utility function to obtain the utility level of each strategy; the expression for the utility function is as follows: In the formula, Indicates the first The utility function of each partition, Indicates the degree of benefit of local power distribution modules, Indicates the local communication module degree benefit, Indicates cross-domain consistency benefits, This indicates the latency cost of the partition. Indicates the interaction complexity cost, Indicates topology switching cost, Indicates the oscillation penalty item, These are the utility weighting coefficients; The policy update probability is calculated based on the utility level of each policy to obtain candidate policies; Determine whether the utility level of the candidate strategy is greater than a pre-set hysteresis threshold. If so, then the candidate strategy is selected as the candidate partitioning strategy.

[0010] Preferably, the expression for the partitioning objective function is:

[0011] In the formula, For the modularity of the power distribution network, For the modularity of communication networks, As a cross-domain mapping consistency indicator, , , These are the corresponding weight coefficients, and .

[0012] Preferably, the verification of the intermediate partitioning scheme to obtain the final partition specifically involves: verifying the intermediate partitioning scheme based on a pre-constructed joint partitioning optimization objective function and constraints to obtain the final partition; wherein, the expression of the joint partitioning optimization objective function is:

[0013] In the formula, For the modularity of the power distribution network, For the modularity of communication networks, As a cross-domain mapping consistency indicator, To reduce the complexity of cross-partition communication, For the global deterministic time delay boundary, As a global oscillation indicator, , , , , , These are the weight coefficients of the joint partitioning optimization objective function.

[0014] Preferably, the constraints include: deterministic delay boundary constraints, revenue hysteresis constraints, intra-mode stability constraints, and average dwell time constraints.

[0015] In a second aspect, 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 communication and power distribution collaborative low-interaction compression networking and partitioning method of the first aspect when executing the computer instructions.

[0016] Thirdly, the present invention also proposes a computer-readable storage medium storing computer instructions thereon, which, when executed, can realize the communication and power distribution collaborative low-interaction compression networking and partitioning method of the first aspect.

[0017] The beneficial effects of this invention are: The method of this invention generates candidate partitioning strategies for each partition by employing a stochastic stable game approach. When dynamic disturbances occur in the coupled system, it can optimize stable strategies based on partition utility, effectively avoiding frequent switching of partition topology due to small fluctuations, ensuring the stability and consistency of the partition update process, and thus solving the technical problems of easy jitter of partition boundaries and unstable system operation in dynamic scenarios. It realizes deep coupling and collaborative optimization of the distribution network physical layer and communication network layer at the partition level, so that the partitioning results not only meet the electrical coupling requirements, but also meet the requirements of low communication interaction and deterministic time delay control, making it more suitable for communication and control collaboration scenarios in new power distribution systems.

[0018] 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 can be realized and obtained by means of the structures pointed out in the description and the drawings. Attached Figure Description

[0019] 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.

[0020] Figure 1 A flowchart of the communication and power distribution collaborative low-interaction compression networking and partitioning method of the present invention is shown. Detailed Implementation

[0021] 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.

[0022] Reference Figure 1 As shown, a method for low-interaction compression networking and partitioning in collaboration between communication and power distribution specifically includes the following steps: S10. Obtain the basic data of the coupled system; wherein, the basic data includes distribution network physical node data, communication node data and cross-domain mapping matrix; S20. Construct a complex network model of the physical layer of the distribution network based on the physical node data of the distribution network, and construct a complex network model of the communication layer based on the communication node data and the cross-domain mapping matrix. S30. Based on the complex network model of the physical layer and the complex network model of the communication layer of the power distribution network, the coupled system is partitioned to obtain the initial partition; S40. When the coupled system experiences dynamic disturbances, a random stable game method is used to generate candidate partitioning strategies for each partition, and an intermediate partitioning scheme is obtained based on the candidate partitioning strategies and the initial partition. S50. Verify the intermediate partitioning scheme to obtain the final partition.

[0023] By employing a stochastic stable game method to generate candidate partitioning strategies for each partition, stable strategy optimization can be carried out based on partition utility when dynamic disturbances occur in the coupled system. This effectively avoids frequent switching of partition topology due to small fluctuations, ensuring the stability and consistency of the partition update process. In turn, it solves the technical problems of easy jitter of partition boundaries and unstable system operation in dynamic scenarios. It realizes deep coupling and collaborative optimization of the distribution network physical layer and communication network layer at the partition level, so that the partitioning results not only meet the electrical coupling requirements, but also meet the requirements of low communication interaction and deterministic time delay control, making it more suitable for communication and control collaboration scenarios in new power distribution systems.

[0024] In step S10 above, the coupling system is the communication and power distribution coupling system that needs to be partitioned.

[0025] Distribution network physical node data refers to data that describes the topology, line parameters, and real-time operating status of all physical devices (such as buses, switches, distributed power sources, energy storage, loads, controllers, etc.) in the distribution network.

[0026] Communication node data refers to data that describes the topology, link status, and real-time operating status of all communication devices (such as base stations, switches, routers, edge gateways, etc.) in a communication network.

[0027] A cross-domain mapping matrix is ​​a matrix that describes the service associations between physical nodes and communication nodes in a distribution network. Its expression is: In the formula, This represents the mapping matrix between power distribution nodes and communication nodes. Indicates the total number of physical nodes in the distribution network. Indicates the total number of communication nodes; Represents communication node responsible node Measurement, control, or forwarding services; This indicates that there is no direct service relationship; when a distribution node is served collaboratively by multiple communication nodes, Pick The inner real number represents the service association strength.

[0028] Specifically, the physical node data of the distribution network includes: the set of physical nodes of the distribution network, the set of branches, the line admittance, the equivalent electrical distance, the physical node state characteristic matrix, and the power flow correlation coefficient.

[0029] The communication node data includes: the set of communication nodes, the set of communication links, the bandwidth of communication links, the link delay, the link reliability, the communication node state characteristic matrix, and the cooperative service coefficient.

[0030] Furthermore, the distribution network physical node set refers to the set of nodes composed of the unique identifiers of all physical devices that require zoned management, extracted from the distribution network equipment ledger. Each physical device (such as busbars, switches, distributed power sources, energy storage, loads, controllers, etc.) corresponds to a unique node number.

[0031] A branch set refers to a set of all branches that directly connect two physical nodes, extracted from the distribution network wiring diagram or topology database. Each branch (such as an overhead line, cable, or transformer) records the node numbers at both ends.

[0032] Line admittance refers to the admittance value of each branch obtained from the power grid parameter table or equipment nameplate, which is used to quantify the static electrical connection strength between two physical nodes.

[0033] Equivalent electrical distance refers to the value obtained based on the power grid topology and line impedance, using the shortest path algorithm or electrical distance calculation formula, and is used to reflect the actual distance between two physical nodes on the circuit.

[0034] The physical node state characteristic matrix refers to a matrix composed of real-time operating state data (such as voltage amplitude, voltage phase angle, active power, reactive power, etc.) of each physical node obtained from the distribution network SCADA system or phasor measurement unit. Each row corresponds to a physical node, and each column corresponds to a state characteristic.

[0035] The power flow correlation coefficient refers to the value obtained by calculating the Pearson correlation coefficient using a sliding window based on the active power (or reactive power) time series extracted from the physical node state feature matrix. It is used to measure the correlation of power fluctuations between two physical nodes during dynamic operation.

[0036] A communication node set refers to a set of nodes composed of unique identifiers of all communication devices (such as base stations, switches, routers, and edge gateways) extracted from the communication network management system or network planning documents; each communication device corresponds to a unique node number.

[0037] A communication link set refers to the set of all physical or logical links (such as optical fibers or wireless channels) that directly connect two communication nodes, extracted from the communication network topology; each link records the numbers of its two end nodes.

[0038] Communication link bandwidth refers to the maximum data transmission rate that the link's physical or configuration layer can support. Furthermore, available bandwidth refers to the value obtained by real-time monitoring of the remaining bandwidth of each link through a network management system, used to quantify the upper limit of data transmission capacity between two communication nodes.

[0039] Link latency refers to the value obtained by actively probing (such as ping, TWAMP) or collecting the one-way or round-trip latency of each link through the network management system. It is used to quantify the time delay of data transmission between two communication nodes.

[0040] Link reliability refers to a value calculated based on historical statistics (such as packet loss rate and number of interruptions over a period of time). It is usually between 0 and 1 and is used to quantify the stability of data transmission between two communication nodes.

[0041] The communication node status feature matrix is ​​a matrix composed of the real-time operating status (such as CPU load, memory usage, queue length, location information, etc.) of each communication node obtained from the network management system. Each row corresponds to a communication node, and each column corresponds to a status feature.

[0042] The collaborative service coefficient is used to reflect the degree of overlap in service coverage between two communication nodes.

[0043] In complex network theory, a network is usually represented by a graph, which contains four elements: V: Node set, representing all individuals or entities in the network.

[0044] E: Edge set, representing the direct connections between nodes.

[0045] W: Edge weight, representing the tightness or strength of the connection between nodes. The larger the weight, the closer the relationship between the two nodes.

[0046] X: The characteristics of the node itself, representing the node's own attributes or state, such as voltage, power, location, load rate, etc.

[0047] Based on this, the present invention constructs complex network models of the physical layer and communication layer of the power distribution network, respectively, and introduces time variables. This characterizes the changes in network parameters and node states over time. Therefore, we can obtain: The expression for the complex network model of the physical layer of the distribution network is:

[0048] In the formula, Indicates time The physical layer of the power distribution network is a complex network. Represents the set of physical nodes in the distribution network. Indicates the total number of physical nodes in the distribution network. Represents the set of branches. This represents the physical layer weighted adjacency matrix (a matrix composed of the weights of all physical layer edges). Represents the physical node state feature matrix. This represents the dimension of the physical node state features.

[0049] in, In the formula, matrix elements Indicates time Next physical node 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, This represents the maximum line admittance. Represents a node and At any moment The correlation coefficient of the trend, Represents a node and The equivalent electrical distance, This indicates the avoidance of extremely small positive numbers with a denominator of zero.

[0050] The expression for the complex network model at the communication layer is:

[0051] 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. This represents the weighted adjacency matrix of the communication layer (a matrix composed of the weights of all communication layer edges). Represents the state feature matrix of the communication node. This represents the dimension of the state characteristics of the communication node.

[0052] in, 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, and satisfy... , Indicates the bandwidth of the communication link. Indicates the maximum available bandwidth. Indicates link reliability. Indicates link latency. Indicates the maximum allowable delay. Indicates the collaborative service coefficient. and Representing communication nodes and At any moment The set of physical nodes covered or served. and All through cross-domain mapping matrix Sure, Indicates the number of elements in the set. This indicates the avoidance of extremely small positive numbers with a denominator of zero.

[0053] In step S30 above, based on the complex network model of the distribution network physical layer and the complex network model of the communication layer, the coupled system is partitioned to obtain an initial partition, specifically as follows: S301. Based on the pre-constructed partitioning objective function, the coupled system is partitioned to obtain a preliminary partitioning scheme; The expression for the partitioning objective function is:

[0054] In the formula, For the modularity of the power distribution network, For the modularity of communication networks, As a cross-domain mapping consistency indicator, , , These are the corresponding weight coefficients, and

[0055] Furthermore, In the formula, For the total edge weight of the distribution network, For nodes The weighting degree, Represents a node The tag of the corresponding section. For the Kronecker function, This represents the total number of physical nodes in the distribution network. For nodes The weighted degree, which is the sum of the weights of all associated branches of that node. For nodes The weighting degree, for physical node at time and The overall electrical coupling weight between them.

[0056] In the formula, Represents the total edge weight of the communication network. Represents communication node The weighting degree, Represents communication node The tag of the corresponding section. for Time communication node and Comprehensive communication and coordination weights between them For communication nodes The weighted degree is the sum of the weights of all associated links of that node. For communication nodes The weighting degree, This represents the total number of communication nodes.

[0057] In the formula, It is a cross-domain mapping consistency index used to characterize the degree to which a distribution node and its serving communication node fall into the same partition.

[0058] S302. Based on the preliminary partitioning scheme, identify the boundary nodes and obtain the boundary node set; Based on the preliminary partitioning scheme of the coupled system, boundary nodes among the physical and communication nodes of the distribution network are identified, providing a node set basis for subsequent boundary node importance assessment and proxy node selection. Boundary nodes are determined according to their partition affiliation, and their expression is as follows: In the formula, Set of boundary nodes and These are the physical nodes and communication nodes of the distribution network. For the collection of physical branches of the distribution network, A set of communication network links. Represents a node The corresponding partition label, Represents a node The corresponding partition label, Represents a node With nodes They belong to different partitions. When a node has a topological relationship that is directly connected to nodes in other partitions, the node is determined to be a cross-cluster boundary node and included in the boundary node set. middle.

[0059] S303. Based on the set of boundary nodes, calculate the comprehensive importance score for each boundary node; Based on the established set of cross-cluster boundary nodes, a comprehensive importance score is applied to each boundary node to select highly representative proxy nodes. Specifically, the expression for calculating the comprehensive importance score of boundary nodes is: In the formula, Represents boundary nodes The overall importance score, Represents the normalized boundary degree. This represents the normalized betweenness centrality of a node in a distribution network. This represents the normalized betweenness centrality of a node in a communication network. This represents the normalized latency sensitivity of a node. , , , These are the scoring weights, and they satisfy... .

[0060] S304. Based on the comprehensive importance score of each boundary node, select proxy nodes and construct a boundary aggregation matrix; Specifically, the expression for selecting the proxy node is: In the formula, For a set of proxy nodes, For boundary nodes, For the set of boundary nodes, For the first The overall importance score of each boundary node. The preset importance threshold is set when the overall importance score of the boundary node is greater than or equal to... At that time, the node is selected as the proxy node, which undertakes the functions of aggregating and interacting with the boundary states.

[0061] Constructing the boundary aggregation matrix specifically includes: After determining the proxy nodes, the remaining boundary nodes are the non-proxy boundary nodes. Mapping and compression are performed on these non-proxy nodes, and the expression is as follows: In the formula, For the first Non-agent boundary nodes The corresponding proxy node, For nodes to proxy node Communication hop count, , They are nodes With proxy nodes The state vector, It is a 2-norm. For nodes to proxy node Link quality, , , These are the mapping cost weighting coefficients, used to balance the influence of various mapping factors.

[0062] Based on the mapping relationship from non-proxy boundary nodes to proxy nodes described above, a boundary aggregation matrix is ​​constructed. The expression for the boundary aggregation matrix is ​​as follows: In the formula, Represents the boundary aggregation matrix; when the boundary nodes Mapped to proxy node hour, ,otherwise .

[0063] S305. Based on the boundary aggregation matrix, compress the boundary nodes; Specifically, the state information of boundary nodes and the interaction relationships between nodes are compressed, and the expression is as follows:

[0064]

[0065] In the formula, For the original boundary node state vector, For the compressed proxy node state vector, Represents the compressed proxy interaction matrix, This represents the original boundary interaction matrix.

[0066] By using a node compression mechanism, the original many-to-many interaction relationships between a large number of boundary nodes are compressed into a small number of interaction relationships between proxy nodes, which significantly reduces the scale of cross-partition message exchange and communication interaction complexity, and improves the efficiency of communication resource utilization.

[0067] S306. Based on the compressed boundary nodes, calculate the global deterministic time delay boundary of the preliminary partitioning scheme; After compressing the boundary nodes using the boundary aggregation matrix, the main body of cross-regional interaction in the coupled system has been simplified to proxy nodes, effectively reducing the amount of interactive data and communication complexity. However, cross-regional information interaction still needs to be carried out between the compressed proxy nodes. Distribution network collaborative control has strict requirements for the real-time and deterministic nature of data transmission. If the cross-regional communication delay is uncontrollable or exceeds the allowable range, it will directly affect the system's operational stability and control accuracy. Therefore, it is necessary to construct a deterministic delay boundary for the cross-regional communication path to clarify the maximum delay upper limit of cross-regional interaction between proxy nodes, quantitatively evaluate the real-time performance of the current partitioning and boundary compression scheme, and provide a basis for judging whether the partitioning scheme meets the requirements.

[0068] The expression for the upper bound of the deterministic delay of a single communication link is: In the formula, Indicates link upper bound of deterministic delay, Indicates the propagation delay, Indicates processing delay, Indicates control message length, Indicates link service rate, Indicates the arrival burst degree parameter, Let represent the average velocity of the arriving flow, and satisfy . , This indicates the upper bound of the link scheduling wait delay.

[0069] The expression for the path-level deterministic delay boundary is: In the formula, Representing a path Deterministic delay boundary, Indicates link Belongs to path .

[0070] The expression for the global deterministic delay boundary is: In the formula, Indicates partition With partitions The maximum time delay boundary among all candidate paths; This represents the global deterministic time delay boundary.

[0071] S307. Determine whether the global deterministic delay boundary is greater than the preset delay threshold. If so, repeat steps S301-307 until an initial partition with a delay less than the preset delay threshold is obtained.

[0072] The above steps yield an initial partition that satisfies electrical coupling, communication coordination, cross-domain consistency, and latency constraints at the current moment. However, the power distribution network and communication network coupling system is dynamically changing. For example, fluctuations in distributed power output and load changes alter electrical coupling relationships; changes in communication link load and signal interference change link latency and available bandwidth; and the addition of new equipment and the removal of old equipment change node affiliations. If repartitioning is performed for every change, it will lead to frequent and significant changes in the partitioning scheme, repeated reconfiguration of control strategies, and repeated reconfiguration of communication links, ultimately causing system instability.

[0073] In response to disturbances (load fluctuations, communication changes, topology changes) in the operation of the coupled system, each partition can adopt different coping behaviors. Therefore, a set of strategies is generated for each partition, treating each partition as a game player. The expression for the set of strategies for each partition is:

[0074] In the formula, Indicates the first The strategy for each partition, Indicates boundary migration strategy, Indicates the proxy node update strategy, This indicates the partition topology switching strategy.

[0075] Furthermore, to evaluate the effectiveness of different strategies in each partition, a utility function can be used to comprehensively consider the benefits and costs of strategy execution, thereby achieving a quantitative assessment of the strategy's merits. Its expression is:

[0076] In the formula, Indicates the first The utility function of each partition, Indicates the degree of benefit of local power distribution modules, Indicates the local communication module degree benefit, Indicates cross-domain consistency benefits, This indicates the latency cost of the partition. Indicates the interaction complexity cost, Indicates topology switching cost, Indicates the oscillation penalty item, These are the utility weighting coefficients; where, In the formula, Indicates the first The switching cost coefficient of each partition, Indicates characteristic functions; In the formula, Indicates the length of the history memory window, Indicates the first The weights corresponding to each historical moment, and satisfying the following conditions: .

[0077] In a dynamic game involving multiple interconnected partitions, the probability of choosing a strategy at the next moment is determined based on the utility level of each strategy, allowing the system to gradually approach a stable equilibrium. The expression for the strategy update probability is:

[0078] In the formula, Indicates the first Candidate strategies for each partition, Indicates except the first The remaining partitions, besides the one partition, are at time... Strategy combination, Indicates the first The strategy set for each partition, Indicates the inverse temperature parameter. This indicates a summation traversal. Each strategy in the process.

[0079] Furthermore, to avoid triggering partition switching due to minor utility improvements, a hysteresis constraint is set. Policy adjustments are only allowed when the increase in returns is sufficiently significant. The expression for the hysteresis constraint is:

[0080] In the formula, Indicates the first The hysteresis threshold for each partition's return. Only when the return improvement of the candidate partition strategy is not less than... Partition switching is only permitted under certain conditions.

[0081] After obtaining candidate partitioning strategies through stochastic stable game theory, although frequent handovers have been suppressed at the decision-making level, partitioning topology switching inherently transforms the entire power distribution and communication coupling system into a type of time-delayed dynamic system. At this point, relying solely on game gains cannot guarantee asymptotic stability at the system state level. Therefore, a final partitioning strategy is needed that simultaneously satisfies the optimization objective, time delay boundaries, and dynamic stability.

[0082] A time-delay switching system model of the coupled system is established to represent the dynamic characteristics of the coupled system and the impact of network-induced delay during topology switching. Its expression is as follows: In the formula, Represents the system state vector, Indicates partition topology mode index, The state matrix representing the current partition topology mode, Represents the time delay state matrix, This indicates network-induced latency.

[0083] To ensure that the coupled system always meets the real-time requirements, its time delay behavior is constrained, and its expression is as follows: In the formula, Indicates the preset delay threshold, This represents the upper bound of the rate of change of time delay.

[0084] For a coupled system operating under any fixed partition topology mode s, to ensure the asymptotic stability of the system state within this mode, a stability criterion for the coupled system within the mode is constructed, the expression of which is:

[0085] In the formula, For a symmetric positive definite matrix, Indicates the first State matrix under partition topology mode Indicates the first The delay state matrix under a partitioned topology pattern. Furthermore, .

[0086] When a coupled system switches from one partitioned topology mode to another, both the system structure and energy function undergo abrupt changes. To prevent a sharp increase in energy during the switch that could lead to system instability, jump variable constraints are applied to the Lyapunov function values ​​before and after the mode switch. This limits the energy increase to an allowable range, thus ensuring a stable and smooth partitioned topology switch process. Specifically, when the coupled system switches from partitioned topology mode s to mode r, to prevent a sharp increase in system energy that could lead to instability, jump variable constraints are applied to the Lyapunov function of the coupled system. The expression for this constraint is: In the formula, Representation pattern Lyapunov functions under Representation pattern Lyapunov functions under and Representing the time before and after the switch, Indicates from pattern To mode The upper bound of the Lyapunov jump variable.

[0087] To avoid frequent and rapid switching between multiple topology modes in the coupled system, and to ensure stable system operation through the switching mechanism, a constraint on the average residence time of the coupled system is introduced, the expression of which is: In the formula, Indicates average length of stay, Represents the upper bound of all pattern jump variables. Indicates the in-mode attenuation rate.

[0088] After constructing the oscillation index, time-delay model, and stability constraints for the coupled system, all optimization and cost terms can be completely solved. To quantify the global oscillation level of the coupled system, a global oscillation index is defined, whose expression is: In the formula, Indicates topology switching cost, This indicates a penalty for oscillation.

[0089] Based on this, a complete joint partitioning optimization objective function is used to make a global decision on the partitioning scheme of the coupled system, the expression of which is: In the formula, For the modularity of the power distribution network, For the modularity of communication networks, As a cross-domain mapping consistency indicator, The complexity of cross-partition communication interaction (after compression). For the global deterministic time delay boundary, As a global oscillation indicator, , , , , , These are the weighting coefficients, For the first The set of proxy nodes in each partition The total number of collaborative partitions, For the first The number of proxy nodes in each partition For partitioning With partitions The frequency of interaction between them per unit of time.

[0090] Ultimately, the candidate partition is output as the final communication and power distribution partitioning result only when the candidate partition simultaneously satisfies the joint objective optimality, deterministic delay boundary constraints, revenue hysteresis constraints, intra-mode stability constraints, and average dwell time constraints; otherwise, the original partition topology remains unchanged to suppress partition topology switching oscillations.

[0091] Based on the same inventive concept as the above method, this invention also proposes a communication and power distribution collaborative low-interaction compressed networking and partitioning system, comprising: The data acquisition module is used to acquire basic data of the coupled system; wherein, the basic data includes distribution network physical node data, communication node data and cross-domain mapping matrix; The model building module is used to build a complex network model of the physical layer of the distribution network based on the physical node data of the distribution network, and to build a complex network model of the communication layer based on the communication node data and the cross-domain mapping matrix. The initial partitioning module is used to partition the coupled system based on the complex network model of the physical layer and the complex network model of the communication layer of the power distribution network to obtain the initial partitions; The candidate partitioning module is used to generate candidate partitioning strategies for each partition using a stochastic stable game method when the coupled system experiences dynamic disturbances, and to obtain intermediate partitioning schemes based on the candidate partitioning strategies and the initial partitions. The verification module is used to verify the intermediate partitioning scheme and obtain the final partition.

[0092] 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 communication and power distribution collaborative low-interaction compression networking and partitioning method when executing the computer instructions.

[0093] Based on the same inventive concept as the above method, the present invention also proposes a computer-readable storage medium storing computer instructions thereon, which, when executed, can realize the above-mentioned communication and power distribution collaborative low-interaction compression networking and partitioning method.

[0094] 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.

[0095] 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.

[0096] 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 low-interaction compressed networking and partitioning in collaboration between communication and power distribution, characterized in that, include: S10. Obtain the basic data of the coupled system; wherein, the basic data includes distribution network physical node data, communication node data and cross-domain mapping matrix; S20. Construct a complex network model of the physical layer of the distribution network based on the physical node data of the distribution network, and construct a complex network model of the communication layer based on the communication node data and the cross-domain mapping matrix. S30. Based on the complex network model of the physical layer and the complex network model of the communication layer of the power distribution network, the coupled system is partitioned to obtain the initial partition; S40. When the coupled system experiences dynamic disturbances, a random stable game method is used to generate candidate partitioning strategies for each partition, and an intermediate partitioning scheme is obtained based on the candidate partitioning strategies and the initial partition. S50. Verify the intermediate partitioning scheme to obtain the final partition.

2. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 1, characterized in that, The distribution network physical node data includes: distribution network physical node set, branch set, line admittance, equivalent electrical distance, physical node state characteristic matrix, and power flow correlation coefficient; The communication node data includes: a set of communication nodes, a set of communication links, communication link bandwidth, link latency, link reliability, a communication node state feature matrix, and cooperative service coefficients.

3. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 1, characterized in that, The process of partitioning the coupled system based on the complex network model of the distribution network physical layer and the complex network model of the communication layer to obtain an initial partition includes: S301. Based on the pre-constructed partitioning objective function, the coupled system is partitioned to obtain a preliminary partitioning scheme; S302. Based on the preliminary partitioning scheme, identify the boundary nodes and obtain the boundary node set; S303. Based on the set of boundary nodes, calculate the comprehensive importance score for each boundary node; S304. Based on the comprehensive importance score of each boundary node, select proxy nodes and construct a boundary aggregation matrix; S305. Based on the boundary aggregation matrix, compress the boundary nodes; S306. Based on the compressed boundary nodes, calculate the global deterministic time delay boundary of the preliminary partitioning scheme; S307. Determine whether the global deterministic delay boundary is greater than the preset delay threshold. If so, repeat steps S301-307 until an initial partition with a delay less than the preset delay threshold is obtained.

4. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 3, characterized in that, The construction of the boundary aggregation matrix specifically includes: Based on the proxy nodes, determine the non-proxy nodes; Compress the non-proxy nodes to construct a boundary aggregation matrix.

5. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 1, characterized in that, The method of generating candidate partitioning strategies for each partition using a stochastic stable game specifically includes: A policy set is generated for each partition; wherein the policy set includes: boundary migration, proxy node update, and partition topology switching; Each strategy is evaluated using a utility function to obtain the utility level of each strategy; the expression for the utility function is as follows: In the formula, Indicates the first The utility function of each partition, Indicates the degree of benefit of local power distribution modules, Indicates the local communication module degree benefit, Indicates cross-domain consistency benefits, This indicates the latency cost of the partition. Indicates the interaction complexity cost, Indicates topology switching cost, Indicates the oscillation penalty item, These are the utility weighting coefficients; The policy update probability is calculated based on the utility level of each policy to obtain candidate policies; Determine whether the utility level of the candidate strategy is greater than a pre-set hysteresis threshold. If so, then the candidate strategy is selected as the candidate partitioning strategy.

6. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 3, characterized in that, The expression for the partitioning objective function is: In the formula, For the modularity of the power distribution network, For the modularity of communication networks, As a cross-domain mapping consistency indicator, , , These are the corresponding weight coefficients, and .

7. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 1, characterized in that, The verification of the intermediate partitioning scheme to obtain the final partitioning specifically involves: verifying the intermediate partitioning scheme based on a pre-constructed joint partitioning optimization objective function and constraints to obtain the final partitioning; wherein, the expression of the joint partitioning optimization objective function is: In the formula, For the modularity of the power distribution network, For the modularity of communication networks, As a cross-domain mapping consistency indicator, To reduce the complexity of cross-partition communication, For the global deterministic time delay boundary, As a global oscillation indicator, , , , , , These are the weight coefficients of the joint partitioning optimization objective function.

8. The communication and power distribution collaborative low-interaction compressed networking and partitioning method according to claim 7, characterized in that, The constraints include: deterministic delay boundary constraints, revenue hysteresis constraints, intra-mode stability constraints, and average dwell time constraints.

9. A device comprising a memory and a processor, wherein the memory stores computer instructions executable on the processor, characterized in that, When the processor executes the computer instructions, it performs the communication and power distribution collaborative low-interaction compression networking and partitioning method according to any one of claims 1 to 8.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the computer instructions are executed, the communication and power distribution collaborative low-interaction compression networking and partitioning method described in any one of claims 1 to 8 can be implemented.

Images