Multi-agent cooperative power system dynamic security regulation system

Abstract

The application belongs to the technical field of electric power, and particularly relates to a kind of multi-agent coordinated electric power system dynamic security regulation system, by introducing state monitoring module, abnormal response module, capability evaluation module and priority adjustment module, the task redistribution mechanism based on load and communication capability weighted evaluation is realized, makes the remaining agent can quickly undertake the task of fault node according to own processing margin, and priority is guaranteed to the smooth execution of high energy consumption task.This mechanism not only improves the response speed of the system, but also enhances the stability and resource utilization efficiency of the overall operation.Therefore, compared with the prior art, the application shows higher task migration success rate and shorter recovery time in the fault scenario, effectively improves the safety and robustness of the electric power system under non-steady state conditions.

Description

Technical Field

[0001] This invention belongs to the field of power technology, specifically relating to a dynamic security control system for power systems with multi-agent collaboration. Background Technology

[0002] In current power systems, multi-agent cooperative control has been widely applied in key areas such as grid dispatching, load allocation, and fault recovery. Traditional multi-agent systems typically employ centralized or fixed-rule task allocation mechanisms. When a particular agent fails, the system often relies on pre-set backup nodes or static priority strategies for task transfer. However, this approach is slow to respond to sudden faults or load fluctuations and struggles to consider the real-time load status and communication capabilities of the remaining agents, easily leading to task transfer failures or localized system overload.

[0003] In existing technologies, a common approach is to periodically collect the state information of each agent and make task redistribution decisions at a central node. While this method has a certain degree of fault tolerance, it relies on the computing power of the central node and the stability of the communication link. In the event of node failure or significant network latency, it is prone to scheduling delays or even task loss. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic safety control system for power systems with multi-agent collaboration, which enables rapid and reasonable redistribution of tasks for remaining agents, avoids system performance degradation or task execution interruption, and solves the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a multi-agent collaborative dynamic security control system for power systems, comprising:

[0006] A status monitoring module used to monitor the operating status of each intelligent agent in real time and record the load rate and communication link stability of the intelligent agents;

[0007] When any agent is detected to be abnormal, an abnormal response module is used to obtain the task list and current resource usage parameters of the abnormal agent.

[0008] A capability assessment module that calculates the remaining processing capacity of each remaining agent based on the task list, wherein the remaining processing capacity is obtained by weighting the load rate and communication bandwidth;

[0009] Based on the remaining processing capacity, the task priorities of the remaining agents are dynamically adjusted, with priority allocation to the priority adjustment module for high-energy-consuming tasks.

[0010] Preferably, the real-time monitoring of the operational status of each intelligent agent includes:

[0011] Collect the CPU usage, memory usage, and task queue length of each agent to generate runtime load metrics;

[0012] Based on the aforementioned operational load index, calculate the fluctuation value of the agent within a continuous time window;

[0013] A communication quality score is constructed by receiving signal strength and response delay data sent by neighboring agents.

[0014] By combining fluctuation values ​​and communication quality scores, it is determined whether the current agent is in an abnormal edge state. If it exceeds a set threshold, it is marked as an unstable node.

[0015] Preferably, the task list and current resource usage parameters of the abnormal agent are obtained, including:

[0016] Upon receiving an unstable node marker, a task snapshot request is sent to the node to obtain the set of tasks that are currently executing and waiting to be executed;

[0017] Based on the task set, query the data volume and estimated remaining execution time for each task;

[0018] Summarize the data volume and estimated remaining execution time of the tasks, and calculate the total unprocessed load of the abnormal agent;

[0019] Based on the total pending workload and the agent's current available memory capacity, determine whether to trigger an emergency migration priority, and arrange the task migration order according to the data volume.

[0020] Preferably, the remaining processing capacity of each remaining agent is calculated based on the task list, including:

[0021] Obtain the current load status and available communication bandwidth for each remaining agent;

[0022] The available processing capacity is calculated based on the current load status, and a weighted processing capacity index is generated by combining the available processing capacity with the available communication bandwidth.

[0023] The weighted processing capability index of all remaining agents is normalized to obtain the task assignment ratio allocation benchmark for each agent.

[0024] Preferably, based on the remaining processing capacity, the task priority of the remaining agents is dynamically adjusted, prioritizing the allocation of high-energy-consuming tasks, including:

[0025] The remaining agents are ranked by their weighted processing capacity index, and the agent with the highest processing capacity is selected as the master allocation node.

[0026] Obtain historical energy consumption records for each task in the abnormal agent and set task energy consumption levels;

[0027] Insert tasks into the task queue of the main allocation node in descending order of task energy consumption, and replace low-energy-consumption tasks in it;

[0028] Broadcast the updated task list to the remaining agents and update the local task scheduling flags synchronously.

[0029] Preferably, it also includes a task allocation module for constructing a task migration matrix, which splits the tasks of the abnormal agents according to priority and maps them to the remaining agents; specifically including the following steps:

[0030] The tasks are sorted according to their energy consumption levels to form an ordered task queue, with the task with the highest energy consumption placed at the top.

[0031] Based on the task assignment ratio of each agent, the upper limit of tasks that each agent can receive is determined.

[0032] The tasks in the ordered task queue are assigned to the remaining agents in turn, and the tasks are mapped to the corresponding agent execution list, provided that the upper limit is not exceeded.

[0033] Generate a task migration matrix, where the matrix elements represent the allocation relationship between tasks and agents.

[0034] Preferably, it also includes a time coordination module for determining the execution time window for each task in the remaining agents; specifically including the following steps:

[0035] Obtain the set of estimated completion times in the current task queue of the remaining agents, which represents the end time of the last task of each agent;

[0036] Based on the task migration matrix, extract the new task sequence assigned to each agent and calculate the estimated execution time of each task;

[0037] Set a start time window for each new task, with the start time no earlier than the end time of the current last task plus the minimum interval time;

[0038] The starting time window is superimposed with the starting time of subsequent tasks to generate a complete task schedule.

[0039] Preferably, it also includes a topology optimization module for updating the communication topology of the power system and establishing redundant paths between the remaining agents and the target node; specifically including the following steps:

[0040] Obtain the set of current communication paths between the remaining agents and the task target node;

[0041] Based on the communication quality score, available nodes with scores higher than a set threshold are selected as candidate relay points.

[0042] Two nodes in different directions are selected from the candidate relay points to construct a second communication path, forming an alternative transmission channel parallel to the original path;

[0043] Simultaneously activate the original path and the second communication path, and control the distribution of data packets on the two paths through the traffic allocation ratio.

[0044] Preferably, it also includes a feasibility verification module for verifying the feasibility of the task migration matrix. If the total processing capacity of the remaining agents is lower than a preset threshold, the backup node is triggered to start. Specifically, this includes the following steps:

[0045] Based on the task migration matrix, the total data volume of the tasks assigned to each agent is counted, and the overall load requirement is calculated in combination with the task energy consumption level.

[0046] Obtain the weighted total processing capacity of all remaining agents and compare it with the overall load requirement to determine if there is a resource gap;

[0047] If a resource gap exists, a resource gap signal is generated to trigger a backup node access request;

[0048] Activate the least recently used standby node based on the resource shortage signal and add it to the task acceptance list.

[0049] Preferably, it also includes an execution monitoring module, used to synchronously execute the task migration plan and continuously monitor the load changes of the remaining agents, dynamically adjusting the task allocation ratio; specifically including the following steps:

[0050] According to the task migration matrix, each task is distributed to the corresponding agent to start execution, and the initial allocation timestamp is recorded;

[0051] During execution, the current load status and communication quality information of each agent are periodically collected;

[0052] Calculate the deviation between the current load status and the pre-allocated baseline. If the deviation exceeds a set threshold, an adjustment mechanism is activated.

[0053] Based on the deviation, the tasks exceeding the expected amount are reassigned to other agents with lower workloads, and the task mapping is updated.

[0054] Technical effects and advantages of the present invention: The multi-agent collaborative dynamic security control system for power systems proposed in this invention has the following advantages compared with the prior art:

[0055] This invention introduces a status monitoring module, an anomaly response module, a capability assessment module, and a priority adjustment module to implement a task redistribution mechanism based on a weighted evaluation of load and communication capabilities. This mechanism enables remaining agents to quickly take over tasks from faulty nodes based on their processing capacity, while prioritizing the successful execution of high-energy-consuming tasks. This mechanism not only improves the system's response speed but also enhances overall operational stability and resource utilization efficiency. Therefore, compared to existing technologies, this invention exhibits a higher task migration success rate and shorter recovery time in fault scenarios, effectively improving the safety and robustness of power systems under unsteady-state conditions. Attached Figure Description

[0056] Figure 1 This is a block diagram of the multi-agent collaborative dynamic security control system for power systems according to the present invention;

[0057] Figure 2 This is a diagram showing the intelligent agent load analysis of the present invention;

[0058] Figure 3 This is the matrix allocation diagram for the present invention. Detailed Implementation

[0059] 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, and not all embodiments. The specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention. 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.

[0060] This invention provides, for example Figure 1 The illustrated multi-agent collaborative dynamic security control system for power systems includes the following modules:

[0061] The status monitoring module is used to monitor the operating status of each agent in real time and record the load rate and communication link stability of the agents; it includes the following steps:

[0062] Collect the central processing unit utilization of each agent Memory usage and task queue length Generate operating load indicators , where i represents the i-th agent; this formula combines the agent's core resources (CPU and memory) with the scale of the tasks it is processing to form a comprehensive operational load assessment value.

[0063] Based on operating load indicators Calculate the fluctuation value Δ of the agent within a continuous time window. Δ It is the range of change in the operating load of an intelligent agent over a period of time, used to measure whether its operating state is stable.

[0064] A communication quality score is constructed by receiving signal strength and response delay data from neighboring agents. , where j represents the agent adjacent to i; For receiving signal strength indication, This is the round-trip time; the formula quantifies link quality by giving higher weight to high signal strength and low latency by dividing the Received Signal Strength Indication (RSSI) by the round-trip time (RTT). +1 is added to prevent division-by-zero errors.

[0065] Combined with fluctuation value Δ and communication quality score Determine whether the current agent is in an abnormal edge state. If Δ > or < If so, it is marked as an unstable node; The threshold for operating load fluctuation; This sets the lower limit threshold for communication quality scoring. Two independent judgment criteria are set to determine whether the agent is on the verge of anomalies, based on both operational stability and communication stability.

[0066] The anomaly response module is used to obtain the task list and current resource usage parameters of any abnormal agent when an anomaly is detected; it includes the following steps:

[0067] Upon receiving an unstable node marker, a task snapshot request is sent to that node to obtain a set of task identifiers that are currently executing and awaiting execution. When a node is marked as unstable, it actively requests its current task list in order to prepare for task migration.

[0068] Based on task identifier set Query the amount of data corresponding to each task. and the estimated remaining execution time ; This indicates the size of the data that the task needs to process. Both reflect the expected completion time of the task and together determine the migration priority of the task.

[0069] Summary data volume and the estimated remaining execution time Calculate the total unprocessed load of the abnormal intelligent agent. ; Taking into account both the amount of task data and the time consumed, this represents the overall workload of the abnormal node, used to determine whether it can be taken over by other nodes.

[0070] Combined with total unprocessed load With the agent's current available memory Determine whether an emergency migration priority has been triggered. > Then the task migration order will be arranged in descending order of data volume.

[0071] The capability assessment module is used to calculate the remaining processing capacity of each remaining agent based on the task list. The remaining processing capacity is obtained by weighting the load rate and communication bandwidth. The module includes the following steps:

[0072] Get the current load rate of each remaining agent. and available communication bandwidth Both factors together determine the agent's capacity to handle loads. Based on the load rate... Calculate its available processing capacity. ; This indicates the degree to which an agent can continue to accept new tasks; a higher value indicates a greater capacity to take on additional tasks.

[0073] Combined with available processing capacity With communication bandwidth Generate weighted indicators α is the processing weight factor and β is the communication weight factor, which allows the system to dynamically adjust the task allocation strategy according to actual needs.

[0074] Weighted index of all remaining agents After normalization, a baseline for task allocation among each agent is obtained. All... Perform a normalization operation to make the sum equal to 1, thus forming the task acceptance ratio of each node.

[0075] The priority adjustment module is used to dynamically adjust the task priorities of the remaining agents based on the remaining processing capacity, prioritizing the allocation of high-energy-consuming tasks; it includes the following steps:

[0076] Weighted index of remaining agents The system sorts the data and selects the agent with the highest processing capacity as the primary allocation node. Tasks are then prioritized for allocation to the node with the most abundant processing resources to reduce overall scheduling time.

[0077] Obtain historical energy consumption records for each task in the abnormal agent and set task energy consumption levels. ,in Let t be the historical average completion time of task t; The higher the value, the more resources the task consumes per unit of time, i.e., "high energy consumption".

[0078] Arrange the task queue of the primary allocation node according to task energy consumption level. Insert tasks from high to low power consumption, replacing low-power tasks, and keep the total load on the queue below a certain level. Keep the load on the master node below its available capacity, while ensuring that high-value tasks are executed first.

[0079] The updated task list is broadcast to the remaining agents, and local task scheduling flags are updated synchronously to ensure that high-energy-consuming tasks are executed first. Broadcasting ensures that all nodes have priority in task execution. Figure 1 To avoid conflict.

[0080] The task allocation module is used to construct a task migration matrix, which splits the tasks of the abnormal agents according to priority and maps them to the remaining agents; it includes the following steps:

[0081] Based on the task's energy consumption level The tasks are sorted to form an ordered task queue. ,in This indicates the task with the highest energy consumption; prioritize high-energy-consuming tasks to reduce overall scheduling complexity.

[0082] Based on the task assignment ratio of each agent Determine the upper limit of tasks that each agent can receive. ; This indicates the total amount of tasks that the node should undertake, allocated based on its weighted capacity to ensure a balanced distribution of tasks.

[0083] The tasks in task queue Q are assigned to the remaining agents in turn, satisfying the following conditions: Under the premise of [specific conditions], tasks are mapped to the corresponding agent execution list; tasks are assigned item by item to ensure that the sum of the data volume of each task does not exceed the carrying capacity of the node.

[0084] Generate a task transition matrix M[i][j], where M[i][j] = 1 indicates that task i is assigned to agent j, and 0 otherwise. This matrix guides task scheduling in subsequent execution phases. Recording task allocation in matrix form facilitates subsequent scheduling and monitoring.

[0085] The time coordination module is used to determine the execution time window for each task in the remaining agents; it includes the following steps:

[0086] Get the set of estimated completion times in the current task queue of the remaining agents. , used to indicate the end time of the last task of each agent;

[0087] Based on the task transition matrix M[i][j], extract the new task sequence assigned to each agent and calculate the estimated execution time of each task. Set a start time window for each new task. , where δ is the minimum interval time, to ensure that it does not overlap with the original task;

[0088] Start time window The start times of subsequent tasks are sequentially superimposed to generate a complete task schedule, which is used to coordinate the orderly execution of each task on the agent.

[0089] The topology optimization module is used to update the communication topology of the power system and establish redundant paths between the remaining agents and the target node; it includes the following steps:

[0090] Obtain the set of current communication paths between the remaining agents and the task target node. Where i represents the agent number and j represents the target node number; based on communication quality scoring Select available nodes with scores higher than a set threshold as candidate relay points;

[0091] Select two nodes v and w in different directions from the candidate relay points to construct the second path. This creates an alternative transmission channel that runs parallel to the original path; It is the first alternative path from an agent to the target node; It is the second alternative path, which also goes from the same agent to the same target node, but passes through another set of possible relay nodes.

[0092] The original path and the second path are activated simultaneously, and the distribution of data packets on the two paths is controlled by the traffic allocation ratios α and (1-α) to improve transmission reliability.

[0093] The feasibility verification module is used to verify the feasibility of the task migration matrix. If the total processing capacity of the remaining agents is lower than a preset threshold, the backup node is triggered to start. This includes the following steps:

[0094] Based on the task transition matrix M[i][j], the total amount of data assigned to each agent is calculated. And in combination with the task's energy consumption level Calculate overall load requirements ;

[0095] Obtain the weighted sum of the metrics of all remaining agents. and overall load requirements Compare and calculate the difference. ;

[0096] If the difference ΔQ>0, it is determined that the current resources are insufficient, and a resource gap signal is generated to trigger the backup node access request;

[0097] Based on the resource shortage signal, activate the standby node k that has been in a recently unused state, add it to the task acceptance list, and reassess whether the overall system capacity is sufficient. .

[0098] The execution monitoring module is used to synchronously execute the task migration plan and continuously monitor the load changes of the remaining agents, dynamically adjusting the task allocation ratio; it includes the following steps:

[0099] According to the task migration matrix M[i][j], each task is distributed to the corresponding agent to start execution, and the initial allocation timestamp is recorded. During execution, the current load rate of each agent is periodically collected. and communication quality score ; Calculate the current load rate and the pre-allocated baseline deviation ,like If the deviation exceeds ε (the set threshold), the adjustment mechanism is activated; based on the deviation... The tasks that exceed the expected workload are reallocated to other agents with lower workloads, the task mapping is updated, and the total number of tasks remains unchanged.

[0100] To more clearly illustrate the operation method of the aforementioned multi-agent collaborative power system dynamic security control system, specific embodiments will be described below:

[0101] I. Background Setting

[0102] Five intelligent agents (Agent1 to Agent5) are deployed in a regional power grid, responsible for distributed load control, equipment status monitoring, and emergency response tasks. At a certain moment, Agent3 suddenly fails. The system needs to quickly reassign its tasks to other normal nodes and ensure that high-energy-consuming tasks are executed first.

[0103] II. System Flow Execution Instructions

[0104] 1. Status monitoring module in operation

[0105] The collected data is as follows:

[0106] Agent1 0.4 0.3 2 Agent2 0.6 0.5 3 Agent3 0.9 0.8 5 Agent4 0.3 0.2 2 Agent5 0.5 0.4 3

[0107] Operating load indicators: ;

[0108] Calculation results:

[0109] Agent1 (0.4+0.3)*2=1.4 Agent2 (0.6+0.5)*3=3.3 Agent3 (0.9+0.8)*5=8.5 Agent4 (0.3+0.2)*2=1.0 Agent5 (0.5+0.4)*3=2.7

[0110] Fluctuation value Judgment: Assume ΔW threshold =3.0; Agent3's =8.5 is significantly greater than , marked as an unstable node.

[0111] Communication quality score Assume Agent3 communicates with other nodes as follows:

[0112] 1 80 10 2 60 20 4 30 30 5 40 25

[0113] Calculation formula: ;

[0114] Calculation results:

[0115] 1 80 / 11=7.27 2 60 / 21=2.86 4 30 / 31=0.97 5 40 / 26=1.54

[0116] Set communication quality lower limit =2.0, discovered This further confirms that Agent3 is on the verge of anomaly.

[0117] Conclusion: Agent3 has been identified as an unstable node and has entered the abnormal response phase.

[0118] 2. Exception response module starts

[0119] Get Agent3's task list: ;

[0120] Query the data volume and estimated remaining time for each task:

[0121] t1 500 20 t2 300 15 t3 400 10

[0122] Calculate the total workload to be processed:

[0123] =(500*20)+(300*15)+(400*10)=10000+4500+4000=18500.

[0124] Agent3 currently has 15000 units of available memory (M_free=15000).

[0125] Compare > → Trigger emergency migration mechanism, press The order of sorting tasks from largest to smallest is: t1>t3>t2.

[0126] 3. Operation of the competency assessment module

[0127] Obtain the load rate and bandwidth of the remaining agents:

[0128] Agent1 0.3 100 Agent2 0.5 80 Agent4 0.2 120 Agent5 0.4 90

[0129] Calculate available processing margin : ;

[0130] Agent1 0.7 Agent2 0.5 Agent4 0.8 Agent5 0.6

[0131] Weighted Indicators Let α = 0.5, β = 0.5:

[0132] Agent1 0.50.7+0.5100=0.35+50=50.35 Agent2 0.50.5+0.580=0.25+40=40.25 Agent4 0.50.8+0.5120=0.4+60=60.4 Agent5 0.50.6+0.590=0.3+45=45.3

[0133] After normalization, the baseline for the bearing ratio of each node is obtained:

[0134] Agent1 50.35 / 196.3=0.257 Agent2 40.25 / 196.3=0.205 Agent4 60.4 / 196.3=0.308 Agent5 45.3 / 196.3=0.230

[0135] 4. Priority adjustment module operation

[0136] Sort tasks by energy consumption level:

[0137] Historical average completion time as follows:

[0138] t1 25 t2 18 t3 12

[0139] calculate :

[0140] t1 500 / 25=20 t2 300 / 18=16.7 t3 400 / 12=33.3

[0141] After sorting: t3>t1>t2.

[0142] Select Agent4, which has the highest processing capacity, as the primary allocation node and insert high-energy-consuming tasks.

[0143] 5. Task allocation module running

[0144] Ordered task queue According to the proportion of contracting :

[0145] Agent1 50.35*18500=931475 Agent2 40.25*18500=744625 Agent4 60.4*18500=1117400 Agent5 45.3*18500=838050

[0146] Assign tasks in sequence:

[0147] t3:D=400 → Assigned to Agent4;

[0148] t1:D=500 → Assigned to Agent1;

[0149] t2:D=300 → Assigned to Agent5;

[0150] The task transition matrix M[i][j] is generated as follows:

[0151] t1 1 0 0 0 t2 0 0 0 1 t3 0 0 1 0

[0152] 6. Time coordination module in operation

[0153] Agent4 current task completion time set ={12:00,12:30}→max =12:30;

[0154] Minimum interval δ = 5 minutes;

[0155] New task start time window =12:35, subsequent tasks will be arranged accordingly.

[0156] 7. Topology optimization module running

[0157] Original path →5 is unavailable, construct redundant path #imgpt119#;

[0158] Two directional nodes, v=Agent1 and w=Agent4, are selected to form alternative paths and improve transmission reliability.

[0159] 8. Feasibility verification module operation

[0160] #imgpt120#=400+500+300=1200;

[0161] Overall load requirements:

[0162] Q_total=Σ(D_i_j*E_t)=40033.3+50020+300*16.7=13320+10000+5010=28330;

[0163] #imgpt121#=196.3;

[0164] #imgpt122#=28330-196.3=28133.7>0;

[0165] Trigger backup node k to connect and reassess carrying capacity.

[0166] 9. Run the monitoring module.

[0167] Record the initial allocation timestamp #imgpt123#=12:00;

[0168] The load changes are periodically collected. If the deviation exceeds ε=0.1, the task is reassigned.

[0169] like Figure 2 As shown, the following conclusions can be drawn from the data analysis results:

[0170] Agent load analysis:

[0171] Agent3's load metric (#imgpt125#=8.5) significantly exceeded the threshold (3.0), which was 2-8 times that of other agents, verifying its determination as an unstable node;

[0172] The loads of the other agents are all within a safe range, with Agent4 having the lowest load (#imgpt126#=1.0) and the greatest potential to handle the load.

[0173] Communication quality assessment:

[0174] The communication quality score between Agent3 and Agent4 (#imgpt127#=0.97) is below the threshold (2.0), indicating a communication link risk. The communication quality with Agent1 is the best (#imgpt128#=7.27), and it can be used as the preferred migration path.

[0175] Task allocation scheme:

[0176] The high-energy-consuming task t3 (energy level 33.3) is assigned to Agent4, which has the largest processing capacity.

[0177] The task allocation matrix display achieves load balancing, preventing overload of individual nodes;

[0178] The emergency migration mechanism optimizes the execution order based on two criteria: size (t1>t3>t2) and energy consumption level (t3>t1>t2).

[0179] System optimization results:

[0180] By calculating the weighted index #imgpt130#, a balance between computing resources (α=0.5) and bandwidth resources (β=0.5) was achieved.

[0181] Topology optimization establishes dual redundant paths for Agent1 and Agent4, improving the system's fault tolerance.

[0182] The deviation threshold of ε=0.1 set in the monitoring module ensures timely dynamic adjustments.

[0183] like Figure 3 As shown, Figure 3The vertical axis (Y-axis) represents the queue of tasks to be migrated, arranged in descending order of energy consumption priority: t3→t1→t2 (energy consumption level: 33.3>20>16.7); the horizontal axis (X-axis) represents available agent nodes, arranged in descending order of acceptance rate: Agent4→Agent1→Agent5→Agent2 (acceptance rate: 30.8%>25.7%>23.0%>20.5%).

[0184] Hollow cube: Indicates the location of task assignment.

[0185] t3 is located at the top of the matrix (Y=0), reflecting the highest priority.

[0186] t3 allocation: Occupies the top position of Agent4 column. The hollow square indicates that the node has the maximum processing capacity (#imgpt131#=0.8), which is suitable for high-energy-consuming tasks (E_t=33.3).

[0187] t1 allocation: Located in the middle of Agent1 column, the hollow square reflects the selection of suboptimal nodes (#imgpt132#=0.7) and the balancing of bandwidth resources (#imgpt133#=100).

[0188] t2 allocation: Located at the bottom of Agent5 column, it indicates that the remaining tasks are allocated according to the acceptance ratio to avoid Agent2 being overloaded (#imgpt134#=0.5 has reached a medium load).

[0189] Through the complete process described above, the system successfully achieved rapid response to abnormal intelligent agents, task reallocation, resource rebalancing, and communication link optimization, effectively solving the problems of slow response and low resource utilization in traditional methods. Ultimately, this ensured the continuous operation of the power system under unsteady conditions and the priority execution of critical tasks.

[0190] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-agent collaborative dynamic security control system for power systems, characterized in that, include: A status monitoring module used to monitor the operating status of each intelligent agent in real time and record the load rate and communication link stability of the intelligent agents; When any agent is detected to be abnormal, an abnormal response module is used to obtain the task list and current resource usage parameters of the abnormal agent. A capability assessment module that calculates the remaining processing capacity of each remaining agent based on the task list, wherein the remaining processing capacity is obtained by weighting the load rate and communication bandwidth; Based on the remaining processing capacity, the task priority of the remaining agents is dynamically adjusted, and priority adjustment modules for high-energy-consuming tasks are allocated first. Real-time monitoring of the operational status of each intelligent agent, including: Collect the CPU usage, memory usage, and task queue length of each agent to generate runtime load metrics; Based on the aforementioned operational load index, calculate the fluctuation value of the agent within a continuous time window; A communication quality score is constructed by receiving signal strength and response delay data sent by neighboring agents. By combining fluctuation values ​​and communication quality scores, it is determined whether the current agent is in an abnormal edge state. If it exceeds a set threshold, it is marked as an unstable node. Retrieve the task list and current resource usage parameters of the abnormal agent, including: Upon receiving an unstable node marker, a task snapshot request is sent to the node to obtain the set of tasks that are currently executing and waiting to be executed; Based on the task set, query the data volume and estimated remaining execution time for each task; Summarize the data volume and estimated remaining execution time of the tasks, and calculate the total unprocessed load of the abnormal agent; Based on the total pending workload and the agent's current available memory capacity, determine whether to trigger an emergency migration priority, and arrange the task migration order according to the data volume.

2. The multi-agent collaborative dynamic security control system for power systems according to claim 1, characterized in that: The remaining processing capacity of each remaining agent is calculated based on the task list, including: Obtain the current load status and available communication bandwidth for each remaining agent; The available processing capacity is calculated based on the current load status, and a weighted processing capacity index is generated by combining the available processing capacity with the available communication bandwidth. The weighted processing capability index of all remaining agents is normalized to obtain the task assignment ratio allocation benchmark for each agent.

3. The multi-agent collaborative dynamic security control system for power systems according to claim 2, characterized in that: Based on the remaining processing capacity, the task priorities of the remaining agents are dynamically adjusted, prioritizing the allocation of high-energy-consuming tasks, including: The remaining agents are ranked by their weighted processing capacity, and the agent with the highest processing capacity is selected as the master allocation node. Obtain historical energy consumption records for each task in the abnormal agent and set task energy consumption levels; Insert tasks into the task queue of the main allocation node in descending order of task energy consumption, and replace low-energy-consumption tasks in it; Broadcast the updated task list to the remaining agents and update the local task scheduling flags synchronously.

4. The multi-agent collaborative dynamic security control system for power systems according to claim 3, characterized in that: It also includes a task allocation module, used to construct a task migration matrix, which splits the tasks of the malfunctioning agent according to priority and maps them to the remaining agents; specifically, it includes the following steps: The tasks are sorted according to their energy consumption levels to form an ordered task queue, with the task with the highest energy consumption placed at the top. Based on the task assignment ratio of each agent, the upper limit of tasks that each agent can receive is determined. The tasks in the ordered task queue are assigned to the remaining agents in turn, and the tasks are mapped to the corresponding agent execution list, provided that the upper limit is not exceeded. Generate a task migration matrix, where the matrix elements represent the allocation relationship between tasks and agents.

5. A multi-agent collaborative dynamic security control system for power systems according to claim 4, characterized in that: It also includes a time coordination module, used to determine the execution time window for each task in the remaining agents; specifically including the following steps: Obtain the set of estimated completion times in the current task queue of the remaining agents, which represents the end time of the last task of each agent; Based on the task migration matrix, extract the new task sequence assigned to each agent and calculate the estimated execution time of each task; Set a start time window for each new task, with the start time no earlier than the end time of the current last task plus the minimum interval time; The starting time window is superimposed with the starting time of subsequent tasks to generate a complete task schedule.

6. The multi-agent collaborative dynamic security control system for power systems according to claim 5, characterized in that: It also includes a topology optimization module, used to update the communication topology of the power system and establish redundant paths between the remaining agents and the target node; specifically, it includes the following steps: Obtain the set of current communication paths between the remaining agents and the task target node; Based on the communication quality score, available nodes with scores higher than a set threshold are selected as candidate relay points. Two nodes in different directions are selected from the candidate relay points to construct a second communication path, forming an alternative transmission channel parallel to the original path; Simultaneously activate the original path and the second communication path, and control the distribution of data packets on the two paths through the traffic allocation ratio.

7. A multi-agent collaborative dynamic security control system for power systems according to claim 6, characterized in that: It also includes a feasibility verification module, used to verify the feasibility of the task migration matrix. If the total processing capacity of the remaining agents is lower than a preset threshold, the backup node is triggered to start. Specifically, it includes the following steps: Based on the task migration matrix, the total data volume of the tasks assigned to each agent is counted, and the overall load requirement is calculated in combination with the task energy consumption level. Obtain the weighted total processing capacity of all remaining agents and compare it with the overall load requirement to determine if there is a resource gap; If a resource gap exists, a resource gap signal is generated to trigger a backup node access request; Activate the least recently used standby node based on the resource shortage signal and add it to the task acceptance list.

8. A multi-agent collaborative dynamic security control system for power systems according to claim 7, characterized in that: It also includes an execution monitoring module, which is used to synchronously execute the task migration plan and continuously monitor the load changes of the remaining agents, and dynamically adjust the task allocation ratio. Specifically, the following steps are included: According to the task migration matrix, each task is distributed to the corresponding agent to start execution, and the initial allocation timestamp is recorded; During execution, the current load status and communication quality information of each agent are periodically collected; Calculate the deviation between the current load status and the pre-allocated baseline. If the deviation exceeds a set threshold, an adjustment mechanism is activated. Based on the deviation, the tasks exceeding the expected amount are reassigned to other agents with lower workloads, and the task mapping is updated.

Images