A power grid optimization method and system based on topology

Abstract

This invention relates to the field of power communication network technology, specifically to a power grid optimization method and system based on topology. The specific implementation process includes: monitoring the voltage phase angle change rate and queue congestion of the distribution network bus; establishing a two-layer heterogeneous topology model using test probe messages combined with circuit breaker opening and closing times; deducing the power flow state during fault-triggered self-healing, calculating the bus voltage phase angle drift, and conducting a loop-closed impact test; marking paths determined to have deviated from the fault as pseudo-connected communication domains and eliminating corresponding candidate switching actions to separate the safe operating domain; selecting the action combination with the smallest drift as the self-healing execution sequence and converting it into control action commands for issuance. This invention, through a two-layer heterogeneous topology model and a loop-closed impact test mechanism, achieves coordinated optimization of electrical state and communication resources, effectively avoiding control failures caused by communication unavailability or timing mismatch, and improving the stability and security of distribution network fault recovery.

Description

Technical Field

[0001] This invention relates to the field of power communication network technology, specifically to a power grid optimization method and system based on topology. Background Technology

[0002] With the rapid development of smart grid and energy internet technologies, in order to cope with the volatility brought about by distributed power source access and improve power supply reliability, the power grid dispatch center needs to frequently perform optimization operations such as network reconfiguration, fault isolation, and load transfer. Existing technologies mainly construct mathematical models based on the physical power grid topology (such as substation nodes and transmission line connections), collect electrical quantities (voltage, current, power) and use Kirchhoff's laws for state estimation, thereby formulating switching action sequences or power allocation schemes. In this process, the power communication network, as the infrastructure supporting the observability and controllability of the power grid, is responsible for transmitting wide-area measurement data and remote control commands. Its connectivity and transmission quality are prerequisites for ensuring the accurate execution of the aforementioned power grid optimization schemes.

[0003] However, existing technologies have inherent flaws in practical applications. When formulating power supply and distribution optimization strategies, communication networks are often treated merely as static transmission channels, neglecting their inherent topological vulnerabilities and the dynamic constraints of their physical resources (such as bandwidth, latency, and routing hop count). The physical topology of the power grid and the logical topology of the communication network exhibit a complex heterogeneous, non-one-to-one correspondence, and they often share physical corridors (such as OPGW optical cables), exhibiting strong spatial correlation. When the power grid undergoes topology changes due to faults or optimization requirements, without considering the cooperative adaptability of the communication topology, phenomena such as optimal electrical paths but unreachable communication paths or control commands timed out due to communication congestion can easily occur. This neglect of the risks of heterogeneous topology coupling leads to power grid optimization schemes theoretically satisfying electrical constraints, but in actual implementation, control failures occur due to a lack of effective information support, severely restricting the safety of the power supply and distribution system.

[0004] In summary, during the optimization of power supply and distribution system operation, existing technologies are unable to overcome the scheduling problems caused by the heterogeneity of physical power grid and communication network topology, and cannot simultaneously meet electrical operation indicators and communication topology reliability constraints, thus leading to control failure due to neglecting the dynamic constraints of communication resources.

[0005] To address this, a power grid optimization method and system based on topology is proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a power grid optimization method and system based on topology structure for topology optimization of power distribution network outage recovery.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A topology-based power grid optimization method includes:

[0009] Real-time monitoring of the bus voltage phase angle mutation rate of the physical grid layer of the distribution network and the queue congestion of the power communication layer; by sending test probe messages and combining the opening and closing time of the circuit breaker, a two-layer heterogeneous topology model containing the timing coupling constraints of communication transmission and mechanical action is established.

[0010] When a fault in the distribution network triggers self-healing, the power flow state of each candidate switch during fault isolation and power transfer is deduced based on the two-layer heterogeneous topology model. Based on the power flow state, the bus voltage phase angle drift is calculated for each candidate switch action and mapped onto the power communication layer for a loop-closing impact test.

[0011] When the bus voltage phase angle drift is determined to be a deviation failure, it is marked as a communication pseudo-connectivity domain and the corresponding candidate switch action on the physical power grid layer is eliminated, and a safe operating domain that meets the closed-loop impact test is separated; within the safe operating domain, the candidate switch action with the smallest bus voltage phase angle drift is selected and combined into a self-healing execution sequence;

[0012] Based on the aforementioned timing coupling constraints, the self-healing execution sequence is transformed into control action commands for the physical power grid layer, and then sent to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network.

[0013] Preferably, the specific implementation process of establishing a two-layer heterogeneous topology model that includes timing coupling constraints between communication transmission and mechanical action by real-time monitoring of the bus voltage phase angle change rate of the physical grid layer of the distribution network and the queue congestion of the power communication layer, through sending test probe messages and combining the opening and closing times of circuit breakers, includes:

[0014] In the physical power grid layer, bus phase angle time-series data is continuously received, and time alignment and abrupt change feature extraction are performed on the bus phase angle time-series data to obtain the bus voltage phase angle abrupt change rate; in the power communication layer, the communication link queue status information carrying measurement data and control commands is collected to obtain the queue congestion degree; the end-to-end transmission delay is obtained based on the test probe message, and the bus voltage phase angle abrupt change rate, queue congestion degree and the opening and closing time of the corresponding circuit breaker are associated and bound to establish a two-layer heterogeneous topology model that includes electrical state changes and communication coordination relationships.

[0015] Preferably, when a fault in the distribution network triggers self-healing, the specific implementation process of deduceing the power flow state of each candidate switch during fault isolation and power transfer, based on the aforementioned two-layer heterogeneous topology model, includes:

[0016] When a fault in the distribution network triggers self-healing, candidate switching actions covering fault isolation paths and backup power supply paths are generated based on the electrical state changes and communication coordination relationships in the two-layer heterogeneous topology model. Under the candidate switching actions, the voltage phase angle change trend of each bus in the physical grid layer is expanded by combining the current load distribution information and power access status. At the same time, based on the transmission delay and queue status of the corresponding control path in the power communication layer, the consistency verification of the information synchronization feasibility of the candidate switching actions during execution is performed to form the power flow state corresponding to each candidate switching action.

[0017] Preferably, the specific implementation process of calculating the bus voltage phase angle drift for each candidate switch action based on the power grid power flow state and mapping it to the power communication layer for loop-closing impact testing includes:

[0018] The voltage phase angle change trend of the bus associated with the candidate switch action is extracted from the power grid power flow state, and the cumulative offset characteristics of the phase angle change are calculated under a unified time reference to generate the bus voltage phase angle drift. The bus voltage phase angle drift is mapped to the power communication layer as the electrical response of the closed-loop impact test. By triggering the control signaling interaction corresponding to the candidate switch action in the power communication layer, the time delay stability and queue response consistency of the communication path during the closed-loop impact test period are detected, and the closed-loop impact evaluation result characterizing the candidate switch action under both electrical and communication conditions is generated.

[0019] Preferably, when the bus voltage phase angle drift is determined to be a deviation failure, it is marked as a communication pseudo-connectivity domain and the corresponding candidate switching action on the physical power grid layer is eliminated. The specific implementation process of separating the safe operating domain that meets the closed-loop impulse test includes:

[0020] In the closed-loop impact assessment results, candidate switching actions that deviate from the communication signaling feedback state of the bus voltage phase angle drift are identified, and the corresponding communication paths are marked as communication pseudo-connected domains. Based on the correlation of the communication pseudo-connected domains in the two-layer heterogeneous topology model, the affected candidate switching actions in the physical power grid layer are located and eliminated. The remaining candidate switching actions are screened according to communication reachability and phase angle response consistency to form a safe operating domain.

[0021] Preferably, within the safe operating domain, the specific implementation process of selecting the candidate switch action with the smallest bus voltage phase angle drift and combining them into a self-healing execution sequence includes:

[0022] In the safe operation domain, the bus voltage phase angle drift corresponding to each candidate switch action is compared and analyzed to determine the candidate switch actions whose bus voltage phase angle drift is in the stable range; combined with the temporal coupling constraint relationship in the two-layer heterogeneous topology model, the execution order of the candidate switch actions is coordinated and sorted; under the conditions of satisfying communication reachability and smooth transition of physical power grid state, the candidate switch actions are combined according to the sorting result to form a self-healing execution sequence with complete execution logic.

[0023] Preferably, the specific implementation process of converting the self-healing execution sequence into control action commands for the physical power grid layer according to the time-series coupling constraints, and then distributing them to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network includes:

[0024] Based on the sequential relationship of each switch action in the self-healing execution sequence and the corresponding communication path status, the control actions are time-series calibrated and encapsulated to generate control action instructions. The control action instructions are mapped to reachable transmission paths in the power communication layer and scheduled and sent to the corresponding switches and taps according to the time-series coupling constraints. During the execution of the control action instructions, equipment status feedback information is continuously received, and the power supply status changes of the physical power grid layer are updated and confirmed to achieve optimized adjustment of the power supply structure of the distribution network.

[0025] A topology-based power grid optimization system includes:

[0026] The two-layer construction module monitors the bus voltage phase angle mutation rate of the physical grid layer of the distribution network and the queue congestion of the power communication layer in real time. By sending test probe messages and combining the opening and closing time of the circuit breaker, a two-layer heterogeneous topology model containing the timing coupling constraints of communication transmission and mechanical action is established.

[0027] The power outage simulation module, when a fault in the distribution network triggers self-healing, simulates the power flow state of each candidate switch during fault isolation and power transfer based on the dual-layer heterogeneous topology model; based on the power flow state, it calculates the bus voltage phase angle drift for each candidate switch action and maps it to the power communication layer for loop-closing impact test.

[0028] The safety separation module marks the bus voltage phase angle drift as a deviation failure when it is determined to be a communication pseudo-connectivity domain and removes the corresponding candidate switch action on the physical power grid layer, thus separating the safe operating domain that meets the closed-loop impact test; within the safe operating domain, the candidate switch action with the smallest bus voltage phase angle drift is selected and combined into a self-healing execution sequence.

[0029] The topology optimization module transforms the self-healing execution sequence into control action commands for the physical power grid layer based on the temporal coupling constraints, and sends them to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] 1. This invention constructs a two-layer heterogeneous topology model between the physical power grid layer and the power communication layer, which includes timing coupling constraints of communication transmission and mechanical actions. This expands the distribution network optimization process based solely on electrical topology into a collaborative optimization process of electrical and communication states. This improves the problem of feasible electrical paths but unreachable communication paths or control commands failing due to communication congestion, and enhances the executability of distribution network self-healing and power transfer schemes during operation.

[0032] 2. In the fault isolation and power transfer decision-making process, this invention introduces the bus voltage phase angle drift as a unified evaluation index to characterize the electrical impact of switch actions, and maps this index to the power communication layer to conduct a closed-loop impact test. This enables the early identification and elimination of potential communication pseudo-connectivity domains during the optimization decision-making stage, achieving pre-filtering of high-risk switch actions, effectively reducing the risk of false operation, failure to operate, and secondary faults caused by inconsistencies between communication and electrical states, and enhancing the safety margin of distribution network operation.

[0033] 3. This invention comprehensively screens candidate switch actions based on the safe operation domain and generates a self-healing execution sequence under the premise of satisfying communication reachability and smooth transition of physical power grid status. This makes the final control command highly matched with the communication resource status in terms of timing, which not only shortens the fault recovery time, but also reduces the system disturbance caused by frequent reconfiguration. It is beneficial to improve the stability, power supply continuity and overall operating efficiency of the distribution network in complex operating scenarios. Attached Figure Description

[0034] Figure 1 This is a flowchart of a power grid optimization method based on topology proposed in this invention;

[0035] Figure 2 This is a structural diagram of a power grid optimization system based on topology proposed in this invention;

[0036] Figure 3 This is a schematic diagram of the power supply restoration process in a power distribution network after a power outage, as proposed in this invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It must be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to constitute any limitation on the scope of protection of this invention. Therefore, all equivalent changes or modifications conceived by those skilled in the art based on the content disclosed in this invention without inventive effort should fall within the scope of protection claimed by this invention.

[0038] Reference Figures 1 to 3 This invention proposes a power grid optimization method and system based on topology, the technical solution of which is as follows:

[0039] Example 1:

[0040] Reference Figure 1 This embodiment proposes a power grid optimization method based on topology, including:

[0041] Real-time monitoring of the bus voltage phase angle mutation rate of the physical grid layer of the distribution network and the queue congestion of the power communication layer; by sending test probe messages and combining the opening and closing time of the circuit breaker, a two-layer heterogeneous topology model containing the timing coupling constraints of communication transmission and mechanical action is established.

[0042] When a fault in the distribution network triggers self-healing, the power flow state of each candidate switch during fault isolation and power transfer is deduced based on the two-layer heterogeneous topology model. Based on the power flow state, the bus voltage phase angle drift is calculated for each candidate switch action and mapped onto the power communication layer for a loop-closing impact test.

[0043] When the bus voltage phase angle drift is determined to be a deviation failure, it is marked as a communication pseudo-connectivity domain and the corresponding candidate switch action on the physical power grid layer is eliminated, and a safe operating domain that meets the closed-loop impact test is separated; within the safe operating domain, the candidate switch action with the smallest bus voltage phase angle drift is selected and combined into a self-healing execution sequence;

[0044] Based on the aforementioned timing coupling constraints, the self-healing execution sequence is transformed into control action commands for the physical power grid layer, and then sent to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network.

[0045] Furthermore, the specific implementation process of establishing a two-layer heterogeneous topology model that includes timing coupling constraints between communication transmission and mechanical action by real-time monitoring of the bus voltage phase angle change rate of the distribution network physical grid layer and the queue congestion of the power communication layer, through sending test probe messages and combining the opening and closing times of circuit breakers, includes:

[0046] In the physical power grid layer, bus phase angle time-series data is continuously received, and time alignment and abrupt change feature extraction are performed on the bus phase angle time-series data to obtain the bus voltage phase angle abrupt change rate; in the power communication layer, the communication link queue status information carrying measurement data and control commands is collected to obtain the queue congestion degree; the end-to-end transmission delay is obtained based on the test probe message, and the bus voltage phase angle abrupt change rate, queue congestion degree and the opening and closing time of the corresponding circuit breaker are associated and bound to establish a two-layer heterogeneous topology model that includes electrical state changes and communication coordination relationships.

[0047] Specifically, the distribution network comprises multiple ring network power supply lines, several automated switches, and a supporting power communication network. The communication network employs a hybrid fiber optic and wireless networking approach to carry wide-area measurement data and remote control commands. At the physical power grid layer, voltage phase angle timing data from each bus node is continuously received through synchronous phasor measurement devices deployed in the distribution automation system. This bus phase angle timing data is calibrated using a unified time reference, and its sampling period is consistent with the clock synchronization mechanism of the communication system, ensuring the comparability of cross-layer data. During data processing, time alignment processing is performed on the bus phase angle timing data to eliminate the time drift caused by equipment sampling deviations or communication delays. Furthermore, the phase angle change trend within a continuous time window is analyzed to extract abrupt changes that reflect rapid changes in electrical state, thereby obtaining the bus voltage phase angle abrupt change rate. Actual operating data shows that under normal steady-state operating conditions, the bus voltage phase angle abrupt change rate remains at a low level, while it exhibits a significant increase during switch operation or fault occurrence, thus serving as an important state variable characterizing the dynamic response of the power grid.

[0048] In the power communication layer, the communication network management unit collects the queue status information of the communication links carrying measurement data and control commands. This queue status information includes link buffer occupancy levels, queuing time, and instantaneous load changes, thus characterizing the queue congestion degree of the communication links. Through statistical analysis of the queue status over multiple communication cycles, different operating states such as idle, slightly congested, or heavily congested communication links can be distinguished. The queue congestion degree is calculated comprehensively based on the communication link buffer occupancy level, queuing time, and instantaneous load changes, and is classified according to thresholds configured by operation and maintenance. For example, when both buffer occupancy and queuing time are below a preset first threshold, the communication link is determined to be in an idle or slightly congested state; when either indicator exceeds a preset second threshold, the communication link is determined to be in a heavily congested state. Specific thresholds can be flexibly configured by those skilled in the art according to the design bandwidth and operational requirements of different communication networks.

[0049] In real-world deployment scenarios, when a fault occurs in the power distribution network and triggers the parallel issuance of multiple control commands, the congestion of some communication links increases significantly. If this is not taken into account, it can easily lead to excessive delays in control commands or disordered execution order.

[0050] Based on this, by periodically sending test probe messages at the power communication layer, end-to-end transmission delay information of each control path is obtained. These test probe messages use the same bearer path and service priority as the actual control commands, thus accurately reflecting the transmission capacity of the communication link under current load conditions. Combining the opening and closing time parameters of each circuit breaker, load switch, and tap changer in the distribution network, the communication transmission delay and mechanical action response time are uniformly described, allowing the entire process of control commands from generation and transmission to equipment action completion to be characterized on the same time scale. The bus voltage phase angle mutation rate, queue congestion degree, and the opening and closing time of the corresponding circuit breaker are correlated and bound, forming a state correlation relationship spanning the physical power grid layer and the power communication layer. In this correlation, electrical state changes reflect the impact of switch actions on the power grid operating state, while communication state parameters constrain the reachability and timing reliability of control commands during actual execution. Based on the above correlation, a two-layer heterogeneous topology model containing timing coupling constraints of communication transmission and mechanical action is constructed. This model can simultaneously describe the non-one-to-one correspondence between the electrical topology and the communication control path. The specific construction of the two-layer heterogeneous topology model adopts a data structure combining graph theory matrices and association tables. The physical power grid layer is defined as a weighted undirected graph containing bus nodes and line edge sets, while the power communication layer is defined as a directed graph containing routing nodes and fiber optic or wireless link edge sets. The two are associated through a device mapping table, which records the correspondence between the unique identifiers (e.g., device IDs) of physical layer switching devices and the network logical addresses (e.g., IP addresses and port numbers) of communication layer control terminals. Temporal coupling constraints in the model are implemented by adding dynamic time weights to the edge sets. These weights are numerically equal to the sum of the real-time queuing delay, transmission delay, and the inherent mechanical action time of the corresponding circuit breaker in the communication link. This total weight is calculated in real time and compared with the power grid protection action time limit or transient stability limit time. Once the total weight exceeds the aforementioned limit time, the corresponding path is marked as a temporally infeasible path in the topology model, thus achieving the fusion of physical and communication constraints in mathematical expression.

[0051] This embodiment introduces communication resource state constraints into distribution network optimization, avoiding the shortcomings of making decisions solely based on electrical topology while ignoring communication feasibility. This effectively improves control failures caused by communication link congestion or latency uncertainties. The two-layer heterogeneous topology model provides a reliable data foundation and constraints for subsequent loop-closing impact assessment, safe operating domain selection, and self-healing execution sequence generation, enabling the distribution network to have higher security, stability, and execution determinism during fault recovery and operation reconfiguration.

[0052] Furthermore, when a fault in the distribution network triggers self-healing, the specific implementation process of deduceing the power flow state of each candidate switch during fault isolation and power transfer, based on the aforementioned two-layer heterogeneous topology model, includes:

[0053] When a fault in the distribution network triggers self-healing, candidate switching actions covering fault isolation paths and backup power supply paths are generated based on the electrical state changes and communication coordination relationships in the two-layer heterogeneous topology model. Under the candidate switching actions, the voltage phase angle change trend of each bus in the physical grid layer is expanded by combining the current load distribution information and power access status. At the same time, based on the transmission delay and queue status of the corresponding control path in the power communication layer, the consistency verification of the information synchronization feasibility of the candidate switching actions during execution is performed to form the power flow state corresponding to each candidate switching action.

[0054] Specifically, when a fault in the distribution network triggers self-healing, the scope of the fault's impact is first determined using the electrical state changes and communication coordination relationships contained in the two-layer heterogeneous topology model. Then, operable switches adjacent to the faulty section are identified in the physical power grid layer. In this two-layer heterogeneous topology model, the physical power grid layer describes the electrical connections between buses, feeders, and switches, while the power communication layer describes the control paths, transmission delays, and queue states corresponding to each switch. These two layers are linked through timing coupling constraints. Based on this, a set of candidate switch actions covering fault isolation paths and backup power supply paths is generated. This set includes both tripping actions to isolate the faulty section and closing actions to restore power to non-faulty sections. This approach avoids the problem of overlooking communication reachability constraints when relying solely on static topology searches.

[0055] After generating the candidate switch actions, the operating status of each bus in the physical power grid layer is analyzed by combining the current load distribution information and power supply access status. The load distribution information comes from real-time measurement data of the distribution automation system and can reflect the actual load level of each feeder and bus; the power supply access status includes the main transformer operating condition and the grid connection status of distributed power sources. Under the assumptions of the candidate switch actions, the trend of the change in the phase angle of the involved bus voltage is continuously expanded to deduce the impact of different switch operation combinations on the power flow distribution of the power grid. In a preferred embodiment, in a test scenario of a distribution network in an urban area, when a single feeder fails and different power transfer path schemes are implemented, the change amplitude of the phase angle of each bus voltage is significantly different. This change trend can be used to characterize the degree of impact of switch actions on power grid stability.

[0056] In parallel, consistency checks are performed on the control paths corresponding to candidate switch actions at the power communication layer. These consistency checks are based on the communication transmission delay and queue status information already bound in the two-layer heterogeneous topology model, evaluating the feasibility of information synchronization during the execution of control commands. Specifically, based on the expected execution order of each candidate switch action, it is checked whether the corresponding communication path can complete command transmission within the required time window under the current network load conditions, and whether there are timing conflicts caused by queue congestion when multiple control commands are issued concurrently. When the communication links corresponding to some backup power supply paths are in a high-congestion state, although the electrical layer has the conditions for power transfer, the communication consistency check results show that its control timing has uncontrollable risks, thus marking it as a non-preferred solution. By jointly integrating the analysis results of the bus voltage phase angle change trend at the physical power grid layer with the information synchronization feasibility check results at the power communication layer, a power flow state corresponding to each candidate switch action is formed. This power flow state not only reflects the power distribution and phase angle change characteristics at the electrical operation layer but also includes the reachability and timing consistency constraints at the communication execution layer, enabling subsequent decisions to be made within a unified state space.

[0057] This embodiment achieves simultaneous simulation of electrical operation feasibility and communication execution reliability, effectively improving the problem of self-healing schemes failing in actual execution due to neglecting communication constraints. It can identify potential unexecutable switching action combinations in advance, reduce invalid reconfiguration attempts, thereby shortening the distribution network fault recovery time and improving the stability and security of the power supply restoration process.

[0058] Furthermore, based on the power flow state, the specific implementation process of calculating the bus voltage phase angle drift for each candidate switch action and mapping it to the power communication layer for a closed-loop impact test includes:

[0059] The voltage phase angle change trend of the bus associated with the candidate switch action is extracted from the power grid power flow state, and the cumulative offset characteristics of the phase angle change are calculated under a unified time reference to generate the bus voltage phase angle drift. The bus voltage phase angle drift is mapped to the power communication layer as the electrical response of the closed-loop impact test. By triggering the control signaling interaction corresponding to the candidate switch action in the power communication layer, the time delay stability and queue response consistency of the communication path during the closed-loop impact test period are detected, and the closed-loop impact evaluation result characterizing the candidate switch action under both electrical and communication conditions is generated.

[0060] Specifically, based on the power flow state, for each candidate switch action, the bus nodes with electrical connections to that candidate switch action at the physical power grid layer are extracted, and the voltage phase angle change trend of the bus under the assumed conditions of the candidate switch action is obtained. The voltage phase angle change trend originates from the power flow state decomposition results, and its time axis is calibrated with the power communication layer using a unified time reference to ensure consistency in cross-layer state analysis. Based on this, continuous cumulative analysis is performed on the voltage phase angle change trend to extract cumulative offset features that reflect the overall phase angle shift, thereby generating the bus voltage phase angle drift corresponding to the candidate switch action.

[0061] The bus voltage phase angle drift is used as an electrical response indicator for the closed-loop impact test and is introduced into the power communication layer for cross-layer mapping. The electrical response is correlated with the control signaling interaction process corresponding to the candidate switch actions, enabling the communication layer to perceive the timing sensitivity corresponding to the electrical state changes when executing control command interactions. In the power communication layer, the control signaling interaction process corresponding to the candidate switch actions is triggered, and the operational status of the control commands during the transmission and feedback phases is monitored, with a focus on detecting the transmission delay stability and queue response consistency of the communication path during the closed-loop impact test period. Statistical analysis of multiple signaling interaction processes can identify whether there are sudden congestion, delay jitter, or inconsistent feedback issues in the communication link under conditions of rapid changes in electrical state. In a preferred embodiment, when candidate switch actions with large bus voltage phase angle drift are mapped to the power communication layer, some communication paths experience increased queue backlog and delay fluctuations during the closed-loop impact test, leading to significant asynchrony in control signaling feedback. In contrast, candidate switch actions with smaller bus voltage phase angle drift have relatively stable transmission characteristics in their corresponding communication paths under the same test conditions, and can complete command issuance and status feedback within the expected time window.

[0062] Based on the joint analysis of the electrical and communication responses, a closed-loop impact assessment result is generated, characterizing the candidate switch actions under both electrical and communication conditions. Specifically, in the execution and quantitative evaluation of the closed-loop impact test, a series of high-frequency test data packets are continuously sent to the target communication path to simulate the burst characteristics of traffic during the issuance of actual control commands. Delay stability is quantified using the standard deviation index in statistics, i.e., calculating the standard deviation of the round-trip time of the test packets to determine whether there is severe jitter in the communication path. Queue response consistency is evaluated by comparing the cross-correlation coefficient between the transmission rate curve and the reception rate curve; the lower the coefficient, the more severe the queue backlog. The specific rule for determining deviation failure is: if and only if the bus voltage phase angle drift is in the high-risk range, and simultaneously the communication delay standard deviation or test packet loss rate is greater than 1%, this state is considered a deviation failure, indicating a severe mismatch between electrical requirements and communication capabilities. The closed-loop impact assessment result can comprehensively reflect the electrical impact risks and communication execution risks that may be caused by each candidate switch action during actual execution, providing a reliable basis for subsequent safety operation domain separation and optimization decisions.

[0063] This embodiment introduces a loop-closing impact test mechanism driven by bus voltage phase angle drift during the power grid optimization process, realizing the synchronous evaluation of electrical operation impact and communication execution stability. It effectively improves the problem of ignoring the impact of communication timing when only evaluating the loop-closing risk from an electrical perspective, thereby reducing the maloperation and failure caused by cross-layer inconsistency during self-healing and power transfer, and improving the safety and feasibility of the distribution network operation optimization scheme.

[0064] Furthermore, when the bus voltage phase angle drift is determined to be a deviation failure, it is marked as a communication pseudo-connectivity domain and the corresponding candidate switching action on the physical power grid layer is eliminated. The specific implementation process of separating the safe operating domain that meets the closed-loop impulse test includes:

[0065] In the closed-loop impact assessment results, candidate switching actions that deviate from the communication signaling feedback state of the bus voltage phase angle drift are identified, and the corresponding communication paths are marked as communication pseudo-connected domains. Based on the correlation of the communication pseudo-connected domains in the two-layer heterogeneous topology model, the affected candidate switching actions in the physical power grid layer are located and eliminated. The remaining candidate switching actions are screened according to communication reachability and phase angle response consistency to form a safe operating domain.

[0066] Specifically, based on the closed-loop impact assessment results, a joint comparative analysis is performed on the bus voltage phase angle drift and communication signaling feedback status corresponding to each candidate switch action. The bus voltage phase angle drift is used to characterize the impact of the candidate switch action on the electrical operation status of the power grid, while the communication signaling feedback status reflects the actual execution of control commands in the power communication layer, including whether the command issuance is completed within the expected time window and whether the equipment status feedback is consistent with the control logic. When the bus voltage phase angle drift corresponding to a candidate switch action shows obvious dynamic offset characteristics, while its communication signaling feedback status exhibits abnormalities such as delay, loss, or disordered sequence, it is determined that there is a risk of deviation failure between the electrical and communication layers for this candidate switch action. Based on this, the communication path corresponding to this candidate switch action is identified as a pseudo-connectivity domain. The pseudo-connectivity domain does not refer to a physical link interruption, but rather to a situation where, under the operating scenario and time constraints, the communication path remains connected in the topology, but fails to meet the control command's requirements for delay stability and feedback consistency. When a fault occurs in the distribution network and triggers concurrent control of multiple switches, some communication links experience amplified queuing delays due to a sudden increase in load, causing control commands to fail to complete the interaction within the required time. Such links are identified and marked as pseudo-connected communication domains.

[0067] Based on the correlation of communication pseudo-connectivity in the two-layer heterogeneous topology model, its influence range in the physical power grid layer is located. In the two-layer heterogeneous topology model, communication paths and physical switching devices are bound by a control mapping relationship, thus allowing accurate tracing of the specific switching devices affected by the communication pseudo-connectivity and their corresponding candidate switching actions. Through this correlation and location process, candidate switching actions in the physical power grid layer that have a control dependency relationship with the communication pseudo-connectivity are removed from the original candidate set, thereby avoiding the inclusion of operations with unreliable communication execution in subsequent decisions. After removing high-risk candidate switching actions, the remaining candidate switching actions are further screened. The screening process comprehensively considers two dimensions: communication reachability and phase angle response consistency. Communication reachability characterizes the stable transmission capability of control commands under the current network state, while phase angle response consistency reflects whether the phase angle change of the electrical layer bus voltage remains smooth and consistent with the expected operating state. Only when a candidate switching action meets the safety requirements in both of the above dimensions is it included in the final safe operation domain. Specifically, the identification and removal of communication pseudo-connectivity employs topology reverse tracing. First, in the communication layer topology, all links with unsatisfactory latency stability or queue response indicators are marked as pseudo-connected links. Then, using the mapping table in the two-layer heterogeneous topology model, all physical switching devices that must pass through the aforementioned pseudo-connected links for control commands are searched in reverse. A depth-first search strategy is used to traverse the communication paths, removing all affected switching devices from the candidate action set. For the remaining candidate actions, connectivity checks are further performed. Only if the communication path corresponding to the action remains logically connected after removing pseudo-connected domains, and the corresponding phase angle response prediction value is within the safe range, is the action retained. The final set of retained actions constitutes the safe operation domain, ensuring that any action within the domain is executable at both the electrical and communication levels.

[0068] This embodiment introduces a communication pseudo-connectivity domain identification and safe operation domain separation mechanism during the distribution network self-healing and power transfer process. This effectively improves the control failure problem caused by relying solely on physical topology connectivity while neglecting communication execution reliability. It can eliminate potentially unexecutable switching operation combinations in advance during the decision-making stage, reducing the probability of maloperation and repeated reconfiguration, thereby improving the determinism and operational safety of the distribution network fault recovery process and enhancing the overall performance of distribution network operation optimization.

[0069] Furthermore, within the safe operating domain, the specific implementation process of selecting the candidate switch action with the smallest bus voltage phase angle drift and combining them into a self-healing execution sequence includes:

[0070] In the safe operation domain, the bus voltage phase angle drift corresponding to each candidate switch action is compared and analyzed to determine the candidate switch actions whose bus voltage phase angle drift is in the stable range; combined with the temporal coupling constraint relationship in the two-layer heterogeneous topology model, the execution order of the candidate switch actions is coordinated and sorted; under the conditions of satisfying communication reachability and smooth transition of physical power grid state, the candidate switch actions are combined according to the sorting result to form a self-healing execution sequence with complete execution logic.

[0071] Specifically, within the safe operating domain, the bus voltage phase angle drift corresponding to each candidate switch action is compared and analyzed. The bus voltage phase angle drift is derived from the closed-loop impact assessment results and is used to characterize the dynamic impact of different switch actions on the electrical state of the power grid. Through comparative analysis, candidate switch actions with bus voltage phase angle drift within the stable range can be identified. These actions cause less disturbance to the power flow distribution and phase angle changes during execution, which is more conducive to maintaining a smooth transition in the distribution network's operating state. When comparing and analyzing the bus voltage phase angle drift corresponding to candidate switch actions, the bus voltage phase angle drift is correlated with the existing protection settings, transient stability limits, and phase angle over-limit criteria specified in the operating procedures of the distribution network. When the bus voltage phase angle drift is lower than the preset criteria used to trigger out-of-step and over-limit alarms, and the change amplitude remains within the allowable fluctuation range over multiple sampling periods, the state corresponding to the drift is determined to be within the stable range, thus identifying candidate switch actions within the stable range. The specific value of this stable range is configured according to different power grid topologies and operating specifications. In a preferred embodiment, test results on a fault recovery scenario of a power distribution network in a certain urban area show that the transfer path scheme with a smaller bus voltage phase angle drift has a more stable voltage recovery process after execution, and the load fluctuation amplitude is significantly lower than other schemes, thus having higher operational safety.

[0072] Based on the timing coupling constraints described in the two-layer heterogeneous topology model, the candidate switch actions are coordinated and ordered according to their execution sequence. These timing coupling constraints comprehensively consider the control command transmission delay of the power communication layer, equipment response time, and the order in which each switch action in the physical power grid layer affects the operating state. This ensures that the ordering not only meets electrical logic requirements but also the time consistency requirements of communication execution. This coordinated ordering process avoids electrical surge amplification or communication command conflicts caused by improper switch action execution order. When multiple switches need to complete opening and closing operations sequentially, a reasonable execution order can significantly reduce the communication load peak caused by concurrent control and improve the success rate of control command execution. Under the conditions of satisfying communication reachability and smooth transition of the physical power grid state, the ordered candidate switch actions are combined in a predetermined order to form a self-healing execution sequence with complete execution logic. Specifically, the bus voltage phase angle drift is used as the first priority, and the actions within the safe operating domain are initially ordered according to the drift amount from smallest to largest. If multiple actions have similar drift amounts, a secondary ordering is performed using the bandwidth margin of the communication path as the second priority. After determining the sequence, a time window coordination mechanism is introduced to handle dependencies and conflicts. For multiple switching actions that need to pass through the same communication aggregation node or share bandwidth resources, a protection time interval that covers the maximum queuing delay is automatically inserted between their execution times. The length of this interval is dynamically calculated based on the current queue congestion level, ensuring that the instruction for the next action is sent only after the feedback signal of the previous action has been confirmed. This logically forms a self-healing execution sequence with complete timing and collision avoidance capabilities. The self-healing execution sequence has fully considered the dependencies between actions and the execution interval requirements during its generation process, enabling the sequence to continuously and stably complete the entire process of fault isolation and power restoration after being sent to the field equipment.

[0073] This embodiment introduces a bus voltage phase angle drift-driven action optimization and timing coordination mechanism on the basis of the safe operation domain, which ensures that the generated self-healing execution sequence is both electrically stable and communicatively executable. It effectively improves the problem that although the self-healing decision result meets the electrical constraints, the execution process is unstable, thereby improving the overall reliability and operating efficiency of distribution network fault self-healing and operation optimization.

[0074] Furthermore, the specific implementation process of optimizing the power distribution network by converting the self-healing execution sequence into control action commands for the physical power grid layer based on the aforementioned timing coupling constraints, and then distributing them to the corresponding switches and taps through the power communication layer, includes:

[0075] Based on the sequential relationship of each switch action in the self-healing execution sequence and the corresponding communication path status, the control actions are time-series calibrated and encapsulated to generate control action instructions. The control action instructions are mapped to reachable transmission paths in the power communication layer and scheduled and sent to the corresponding switches and taps according to the time-series coupling constraints. During the execution of the control action instructions, equipment status feedback information is continuously received, and the power supply status changes of the physical power grid layer are updated and confirmed to achieve optimized adjustment of the power supply structure of the distribution network.

[0076] Specifically, based on the self-healing execution sequence, the sequential relationship of each switch action is analyzed, and combined with the communication path status corresponding to each switch action, the timing of the control actions is finely calibrated. The timing calibration process uses the communication transmission delay, queue congestion, and mechanical response time of the circuit breaker and tap changer established in the two-layer heterogeneous topology model as constraints to assign a reasonable triggering order and execution time window to each control action. Subsequently, the timing-calibrated control actions are encapsulated into instructions, uniformly encapsulating the operation type of the switch or tap changer, target device identifier, execution sequence marker, and necessary verification information into standardized control action instructions, ensuring good transmissibility and resolvability in the power communication network. The control action instructions are mapped to reachable transmission paths in the power communication layer. The reachability judgment is based on the identification result of pseudo-connected components in the communication network, ensuring that the selected communication path can stably carry the transmission of control instructions under the current network load and operating state. Then, the control action instructions are scheduled and issued according to the timing coupling constraints, ensuring that multiple instructions satisfy the predetermined execution order during issuance while avoiding instantaneous congestion in the communication link. The generation and encapsulation of the control action commands comply with standard power automation communication protocol specifications (such as IEC61850 or IEC60870-5-104). The generated command data frame strictly includes the unique logical address of the target device, the operation type code (open / closed), the sequence check code, and a high-precision execution timestamp. The command issuance process is established on a reliable transport layer connection and employs an application layer closed-loop feedback mechanism. During the execution of the control action commands, device status feedback information from switches and taps is continuously received. This device status feedback information includes the switch open / closed status, tap position changes, and the voltage recovery status of the associated bus. The real-time acquired device status feedback information is compared and analyzed with the expected execution results, and the power supply status changes at the physical power grid layer are dynamically updated and confirmed. When the device execution status is detected to be consistent with the expected result, subsequent control actions in the self-healing execution sequence continue; when abnormal feedback or inconsistent status is detected, subsequent command issuance is promptly stopped and the abnormal status is recorded, thereby preventing the abnormal status from further spreading in the power grid.

[0077] This embodiment achieves a reliable conversion of self-healing execution sequences into physical grid layer control action commands, and incorporates communication timing constraints throughout the entire process of control command generation, issuance, and execution. This effectively mitigates the problem of optimization decision-making schemes being susceptible to communication instability or timing mismatches during actual execution. Power restoration and structural optimization are completed while ensuring control command reachability and execution consistency, shortening distribution network fault handling time, improving power supply continuity and operational stability, and thus enhancing the overall technical effectiveness of distribution network power supply optimization.

[0078] Example 2:

[0079] This embodiment provides a topology-based power grid optimization system, referring to... Figure 2 The system includes a two-layer construction module, a power failure simulation module, a safety separation module, and a topology optimization module.

[0080] The two-layer construction module monitors the bus voltage phase angle mutation rate of the physical grid layer of the distribution network and the queue congestion of the power communication layer in real time. By sending test probe messages and combining the opening and closing time of the circuit breaker, a two-layer heterogeneous topology model containing the timing coupling constraints of communication transmission and mechanical action is established.

[0081] The power outage simulation module, when a fault in the distribution network triggers self-healing, simulates the power flow state of each candidate switch during fault isolation and power transfer based on the dual-layer heterogeneous topology model; based on the power flow state, it calculates the bus voltage phase angle drift for each candidate switch action and maps it to the power communication layer for loop-closing impact test.

[0082] The safety separation module marks the bus voltage phase angle drift as a deviation failure when it is determined to be a communication pseudo-connectivity domain and removes the corresponding candidate switch action on the physical power grid layer, thus separating the safe operating domain that meets the closed-loop impact test; within the safe operating domain, the candidate switch action with the smallest bus voltage phase angle drift is selected and combined into a self-healing execution sequence.

[0083] The topology optimization module transforms the self-healing execution sequence into control action commands for the physical power grid layer based on the temporal coupling constraints, and sends them to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network.

[0084] Furthermore, the dual-layer construction module continuously connects to the distribution automation system and the communication network management system during system operation, acquiring operational data from the physical power grid layer and the power communication layer, respectively. At the physical power grid layer, it receives bus voltage phase angle timing data from synchronous measurement devices and distribution terminals via interfaces, and performs unified time reference alignment processing on the bus voltage phase angle timing data to eliminate the influence of clock deviations between different acquisition devices. After time alignment, the phase angle change process within a continuous time window is analyzed, and phase angle mutation information reflecting rapid changes in electrical state is extracted, thus forming a description result of the bus voltage phase angle mutation rate. At the power communication layer, it connects to the communication network management system to collect communication link operation status information carrying measurement data and control commands. The operation status information includes the communication link queue occupancy level, instantaneous load changes, and queuing delay characteristics during control command transmission. By comprehensively analyzing the above information, a queue congestion characterization result reflecting the operational pressure of the communication link is formed. End-to-end transmission delay information of the control path is obtained by periodically sending test probe messages in the power communication layer. The test probe messages use the same transmission path and scheduling priority as the actual control commands, thus accurately reflecting the transmission performance of the communication network under current load conditions. Combining the opening and closing time parameters of devices such as circuit breakers, load switches, and tap changers, the two-layer construction module provides a unified description of communication transmission delay and equipment mechanical response time, giving a clear time profile of the entire process from control command generation to execution completion. The bus voltage phase angle mutation rate, communication link queue congestion, and the opening and closing times of corresponding devices are correlated and bound to construct a two-layer heterogeneous topology model that includes timing coupling constraints between communication transmission and mechanical actions. Structurally, this two-layer heterogeneous topology model simultaneously preserves the electrical connection relationships of the physical power grid layer and the control path relationships of the power communication layer, coupling the two through control mapping and timing constraints, enabling the model to accurately reflect the synergistic relationship between electrical state changes and communication execution capabilities.

[0085] Furthermore, the power outage simulation module receives operating status information from the dual-layer construction module and the current distribution network during system operation. When a fault occurs in the distribution network and triggers the self-healing process, it identifies candidate switch actions associated with the faulty section based on the physical grid layer structure relationship described in the dual-layer heterogeneous topology model. Under the assumption that each candidate switch action is executed sequentially, it performs a power flow state simulation for the fault isolation and power transfer process. This simulation process comprehensively considers the current load distribution, power supply access status, and line operating parameters, thereby forming a power flow state description result corresponding to each candidate switch action. After obtaining the power flow state, for each candidate switch action, it extracts the bus nodes that have an electrical relationship with that action in the physical grid layer and obtains the voltage phase angle change trend of the bus under the corresponding power flow state. The voltage phase angle change trend is processed under a unified time reference. Through cumulative analysis of the phase angle change process, a bus voltage phase angle drift that reflects the degree of electrical impact is generated. The bus voltage phase angle drift is mapped as an electrical response index to the power communication layer, triggering the control signaling interaction process corresponding to the candidate switch action. The control signaling interaction process employs a communication path and scheduling strategy consistent with actual control execution, monitoring the operational status of control commands during transmission and feedback phases. During the closed-loop impact test period, the stability of transmission delay and the consistency of queue responses are the primary focus of detection, used to assess whether the communication network can reliably support the issuance and feedback of control commands under rapidly changing electrical conditions. Based on the joint detection results of the electrical and communication responses, closed-loop impact assessment results characterizing candidate switch actions under both electrical and communication conditions are generated.

[0086] Furthermore, the safety separation module receives the set of candidate switching actions and their corresponding loop-closing impact assessment results from the power outage simulation module during system operation. The loop-closing impact assessment results simultaneously include the electrical response characteristics of the bus voltage phase angle drift and the time delay stability and queue response status of control signaling in the power communication layer, used to characterize the comprehensive safety of candidate switching actions under actual execution conditions. Based on the above assessment results, the electrical and communication states are jointly determined under a unified analysis framework to avoid decision-making biases caused by risk assessment based on only a single dimension. Candidate switching actions with significant deviations between the bus voltage phase angle drift and the communication signaling feedback status are identified. At the electrical level, this deviation manifests as a non-stationary phase angle change range, while at the communication level, it manifests as inconsistent control signaling feedback timing or significant transmission delay fluctuations. For the aforementioned candidate switching actions, the corresponding communication path in the power communication layer is marked as a pseudo-connected domain, characterizing the characteristic that although the path is topologically reachable, it cannot stably support the execution of control commands under the current operating state. Based on the correlation of the pseudo-connected communication domain in the two-layer heterogeneous topology model, candidate switching actions with control dependencies on the communication path are traced and located in the physical power grid layer. These control dependencies are determined by the control mapping and temporal coupling relationships in the two-layer heterogeneous topology model, and are used to clarify the impact range of communication anomalies on physical operation execution. Hidden execution risks in the communication layer are accurately projected to the physical power grid layer, thereby removing affected candidate switching actions from the original candidate set. After removing risky actions, a safety reconstruction analysis is performed on the remaining candidate switching actions. This reconstruction analysis comprehensively considers communication reachability and the consistency of bus voltage phase angle response, evaluates the stability of candidate switching actions under continuous execution conditions, and forms a safe operating domain that meets the constraints of the closed-loop impact test.

[0087] Furthermore, the topology optimization module receives candidate switch actions within the safe operating domain, along with their corresponding bus voltage phase angle drift characteristics, load distribution information, and communication reachability status. The load distribution information originates from real-time feeder load, voltage monitoring data, and distributed power source access status collected by the distribution automation system, reflecting the actual power supply demand after fault recovery. By introducing this multi-source operating information, the load unevenness and local overload problems caused by refactoring solely based on static topology relationships can be avoided. Based on candidate switch actions within the safe operating domain, the power supply coverage and load transfer effects of the distribution network under different operating structures are deduced. The deduction process continuously evaluates the power supply capacity of each feeder and the bus phase angle change trend at the physical grid layer, and combines this with the stability of the corresponding control path in the power communication layer to select operating structures that maintain stability under both electrical response and communication execution conditions. The selected operating structures are then refined and adjusted, including optimizing the state combinations of some tie switches and sectionalizing switches to minimize power supply distance and reduce energy loss while meeting safety constraints. During the adjustment process, the changes in the phase angle drift of the bus voltage are continuously monitored to ensure that the optimized topology does not introduce new electrical instability factors. After the topology optimization is completed, the optimized switch state combination and its corresponding operating structure are output as the final reference result of the self-healing execution sequence. This is then converted into control action commands and sent to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network.

[0088] This embodiment eliminates the static, ideal transmission channel of the communication network during system operation. Instead, it introduces the network as a crucial constraint affecting power grid optimization decisions into the unified model, effectively improving the problem of electrical optimization failing due to limited communication resources during actual execution.

[0089] Introducing a cross-layer loop closure impact assessment mechanism during the distribution network optimization process enables the system to identify electrical impacts and communication execution risks in advance. This effectively improves the problem of neglecting communication timing constraints when performing loop closure analysis based solely on electrical topology, thereby reducing the probability of self-healing operation failures and secondary disturbances, and enhancing the safety and reliability of the distribution network self-healing and power supply optimization process.

[0090] In the process of self-healing control of distribution network, explicit identification and isolation of communication pseudo-connection risks are realized, which effectively improves the misjudgment problem caused by communication topology surface connectivity but unreliable operating status, and enables self-healing execution decisions to be generated more robustly under two-layer constraints, thereby improving the safety, reliability and controllability of distribution network power supply restoration process.

[0091] Based on ensuring both communication and electrical safety constraints, the operation structure of the distribution network is refined and optimized. This effectively improves the problem that fault self-healing only focuses on rapid recovery while neglecting the operational quality after recovery. It enables the distribution network to maintain a more stable operating state after completing self-healing, thereby improving the overall power supply optimization effect and operational reliability.

[0092] Example 3:

[0093] This embodiment deploys the aforementioned topology-based power grid optimization method and system entirely within the distribution network of a power company. This distribution network includes multiple ring network feeders, several sectionalizing switches and tie switches, and is connected to distributed power sources and loads. (Refer to...) Figure 3 The power distribution automation terminal and the power communication network enable collaborative sensing and control between the physical power grid layer and the power communication layer.

[0094] In the physical power grid layer, bus voltage phase angle timing data from substations and feeder terminals are received in real time. This data is then aligned using a unified time reference to extract the bus voltage phase angle abrupt change rate, reflecting operational disturbance characteristics. Simultaneously, in the power communication layer, communication link queue status information carrying measurement data and control commands is collected. End-to-end transmission delay characteristics are obtained by sending test probe messages. Combined with the opening and closing time information of on-site circuit breakers and switching equipment, a two-layer heterogeneous topology model is established, incorporating timing coupling constraints between communication transmission behavior and mechanical actions. This model clarifies the correlation between electrical state changes and communication execution conditions, providing a foundation for subsequent self-healing decisions.

[0095] During an actual operation, a fault occurred on a feeder in the distribution network due to external force, triggering the self-healing process after the protection device activated. Based on a two-layer heterogeneous topology model, the fault isolation and power transfer simulation process was initiated. Under model constraints, multiple sets of candidate switch actions covering fault isolation paths and backup power supply paths were generated. Combining the current load distribution and distributed generation access status, the trend of bus voltage phase angle changes at the physical grid layer under each candidate scheme was analyzed. Simultaneously, based on the real-time transmission delay and queue status of the corresponding control path in the power communication layer, the consistency verification of information synchronization feasibility during the execution of candidate switch actions was performed, thus forming a power flow state description corresponding to each candidate scheme.

[0096] Based on this, a closed-loop impact assessment is further performed on each candidate switch action. The bus voltage phase angle change trend associated with the candidate switch action is extracted from the power flow state, and the cumulative offset characteristic of the phase angle change within the execution cycle is calculated to generate the bus voltage phase angle drift. Subsequently, the bus voltage phase angle drift is mapped as an electrical response feature to the power communication layer. In the communication layer, the control signaling interaction process corresponding to the candidate switch action is triggered, and the latency stability and queue response consistency of the communication path during this period are detected. Through this method, the system obtains a closed-loop impact assessment result reflecting the execution risk of the candidate switch action under both electrical and communication conditions.

[0097] Based on the closed-loop impact assessment results, candidate schemes with potential risks are identified and eliminated. Candidate switching actions with significant deviations between the bus voltage phase angle drift and the communication signaling feedback state are marked as abnormal, and their corresponding communication paths are identified as pseudo-connected domains. Further, based on the control correlation of these pseudo-connected domains in the two-layer heterogeneous topology model, candidate switching actions affected by them in the physical power grid layer are located and eliminated, thereby separating a safe operating domain that satisfies both electrical stability and communication executability constraints. Within this safe operating domain, the candidate switching action with the smallest bus voltage phase angle drift is selected as the candidate switching action. Combined with the temporal coupling constraints described in the two-layer heterogeneous topology model, the execution order of the candidate switching actions is coordinated and sorted. Under the premise of satisfying communication reachability and smooth transition of the physical power grid state, the candidate switching actions are combined into a complete self-healing execution sequence.

[0098] During the self-healing execution phase, the system transforms the self-healing execution sequence into control action commands for the physical power grid layer based on the aforementioned timing coupling constraints. These control action commands undergo timing calibration and command encapsulation during generation and are mapped to stable and reachable transmission paths in the power communication layer, then distributed to the corresponding switches and tap changers in a predetermined order. During execution, the system continuously receives device status feedback information, dynamically confirms changes in the power supply status of the physical power grid layer, and fine-tunes the execution rhythm as necessary to ensure the continuity and safety of the power supply structure adjustment process.

[0099] This embodiment effectively improves the problem of the disconnect between self-healing decision-making and communication execution in the distribution network by using two-layer heterogeneous collaborative modeling, loop impact assessment, safety separation, self-healing sequence generation and control execution. This allows the self-healing operation to be safely implemented under the dual constraints of electrical stability and communication reliability, thereby improving the power supply restoration efficiency, operational stability and engineering feasibility of the distribution network.

[0100] It should be clarified that the embodiments described above are merely exemplary and are intended to aid in understanding the present invention, not to limit it. Those skilled in the art can make various changes and modifications after grasping the core ideas of the present invention. Therefore, the scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A power grid optimization method based on topology, characterized in that, include: The specific implementation process of establishing a two-layer heterogeneous topology model that includes timing coupling constraints between communication transmission and mechanical action by real-time monitoring of the bus voltage phase angle change rate of the physical grid layer and the queue congestion of the power communication layer, through sending test probe messages and combining the opening and closing times of circuit breakers, includes: continuously receiving bus phase angle timing data in the physical grid layer, and performing time alignment and change feature extraction on the bus phase angle timing data to obtain the bus voltage phase angle change rate; collecting the communication link queue status information carrying measurement data and control commands in the power communication layer to obtain the queue congestion; obtaining the end-to-end transmission delay based on the test probe messages, and associating and binding the bus voltage phase angle change rate, queue congestion, and the opening and closing times of the corresponding circuit breakers to establish a two-layer heterogeneous topology model that includes electrical state changes and communication coordination relationships; When a fault in the distribution network triggers self-healing, the power flow state of each candidate switch action during fault isolation and power transfer is deduced based on the aforementioned two-layer heterogeneous topology model. The specific implementation process of calculating the bus voltage phase angle drift for each candidate switch action and mapping it to the power communication layer for a closed-loop impact test, based on the power flow state, includes: extracting the voltage phase angle change trend of the bus associated with the candidate switch action from the power flow state, calculating the cumulative offset characteristics of the phase angle change under a unified time reference, and generating the bus voltage phase angle drift; mapping the bus voltage phase angle drift as the electrical response of the closed-loop impact test to the power communication layer; and detecting the time delay stability and queue response consistency of the communication path during the closed-loop impact test period by triggering control signaling interaction corresponding to the candidate switch action at the power communication layer, thereby generating a closed-loop impact assessment result characterizing the candidate switch action under both electrical and communication conditions. When the bus voltage phase angle drift is determined to be a deviation failure, it is marked as a communication pseudo-connectivity domain and the corresponding candidate switch action on the physical power grid layer is eliminated, and a safe operating domain that meets the closed-loop impact test is separated; within the safe operating domain, the candidate switch action with the smallest bus voltage phase angle drift is selected and combined into a self-healing execution sequence; Based on the aforementioned timing coupling constraints, the self-healing execution sequence is transformed into control action commands for the physical power grid layer, and then sent to the corresponding switches and taps through the power communication layer to optimize the power supply of the distribution network.

2. The power grid optimization method based on topology according to claim 1, characterized in that, When a fault in the distribution network triggers self-healing, the specific implementation process of deduce the power flow state of each candidate switch during fault isolation and power transfer, based on the aforementioned two-layer heterogeneous topology model, includes: When a fault in the distribution network triggers self-healing, candidate switching actions covering fault isolation paths and backup power supply paths are generated based on the electrical state changes and communication coordination relationships in the two-layer heterogeneous topology model. Under the candidate switching actions, the voltage phase angle change trend of each bus in the physical grid layer is expanded by combining the current load distribution information and power access status. At the same time, based on the transmission delay and queue status of the corresponding control path in the power communication layer, the consistency verification of the information synchronization feasibility of the candidate switching actions during execution is performed to form the power flow state corresponding to each candidate switching action.

3. The power grid optimization method based on topology according to claim 1, characterized in that, When the bus voltage phase angle drift is determined to be a deviation failure, it is marked as a communication pseudo-connectivity domain and the corresponding candidate switching action on the physical power grid layer is eliminated. The specific implementation process of separating the safe operating domain that meets the closed-loop impulse test includes: In the closed-loop impact assessment results, candidate switching actions that deviate from the communication signaling feedback state of the bus voltage phase angle drift are identified, and the corresponding communication paths are marked as communication pseudo-connected domains. Based on the correlation of the communication pseudo-connected domains in the two-layer heterogeneous topology model, the affected candidate switching actions in the physical power grid layer are located and eliminated. The remaining candidate switching actions are screened according to communication reachability and phase angle response consistency to form a safe operating domain.

4. The power grid optimization method based on topology according to claim 1, characterized in that, Within the safe operating domain, the specific implementation process of selecting the candidate switch action with the smallest bus voltage phase angle drift and combining them into a self-healing execution sequence includes: In the safe operation domain, the bus voltage phase angle drift corresponding to each candidate switch action is compared and analyzed to determine the candidate switch actions whose bus voltage phase angle drift is in the stable range; combined with the temporal coupling constraint relationship in the two-layer heterogeneous topology model, the execution order of the candidate switch actions is coordinated and sorted; under the conditions of satisfying communication reachability and smooth transition of physical power grid state, the candidate switch actions are combined according to the sorting result to form a self-healing execution sequence with complete execution logic.

5. The power grid optimization method based on topology according to claim 1, characterized in that, The specific implementation process of optimizing the power distribution network by converting the self-healing execution sequence into control action commands for the physical power grid layer based on the aforementioned timing coupling constraints, and then distributing them to the corresponding switches and taps through the power communication layer, includes: Based on the sequential relationship of each switch action in the self-healing execution sequence and the corresponding communication path status, the control actions are time-series calibrated and encapsulated to generate control action instructions. The control action instructions are mapped to reachable transmission paths in the power communication layer and scheduled and sent to the corresponding switches and taps according to the time-series coupling constraints. During the execution of the control action instructions, equipment status feedback information is continuously received, and the power supply status changes of the physical power grid layer are updated and confirmed to achieve optimized adjustment of the power supply structure of the distribution network.

6. A topology-based power grid optimization system, said system being applied to a topology-based power grid optimization method as described in any one of claims 1 to 5, characterized in that, include: The two-layer construction module monitors the bus voltage phase angle mutation rate of the physical grid layer of the distribution network and the queue congestion of the power communication layer in real time. By sending test probe messages and combining the opening and closing time of the circuit breaker, a two-layer heterogeneous topology model containing the timing coupling constraints of communication transmission and mechanical action is established. The power outage simulation module, when a fault in the distribution network triggers self-healing, simulates the power flow state of each candidate switch during fault isolation and power transfer based on the dual-layer heterogeneous topology model; based on the power flow state, it calculates the bus voltage phase angle drift for each candidate switch action and maps it to the power communication layer for loop-closing impact test. The safety separation module marks the bus voltage phase angle drift as a deviation failure when it is determined to be a communication pseudo-connectivity domain and removes the corresponding candidate switch action on the physical power grid layer, thus separating out the safe operating domain that meets the closed-loop impact test. Within the safe operating domain, the candidate switch action with the smallest bus voltage phase angle drift is selected and combined into a self-healing execution sequence; The topology optimization module converts the self-healing execution sequence into control action commands for the physical power grid layer based on the temporal coupling constraints, and sends them to the corresponding switches and taps through the power communication layer to perform topology optimization on the distribution network.

Images