Low-voltage intelligent switch cooperative control method, device and equipment and storage medium

Abstract

The invention discloses a low-voltage intelligent switch cooperative control method, device and equipment and a storage medium, and belongs to the field of switch control, and the method comprises the steps: obtaining task switch equipment of a to-be-controlled region and a current control task, and determining switch relation data and switch operation state data based on the task switch equipment; obtaining a control sub-graph set based on the switch relation data and the task switch equipment; based on the switch relation data and a switch cluster construction algorithm, obtaining a switch cluster set; obtaining a cooperative control score of the switch cluster set, and obtaining a cooperative control rule set corresponding to the switch cluster set based on the switch control mapping relation and the cooperative control score; obtaining an equipment control action set based on the current control task, the switch operation state data and an equipment control action obtaining algorithm; obtaining a joint instruction set based on the cooperative control rule set and the equipment control action set; the task switch equipment is controlled based on the joint instruction set, and the reliability of switch control in a low-voltage power distribution scene can be improved.

Description

Technical Field

[0001] This invention relates to the field of switch control, and more particularly to a low-voltage intelligent switch collaborative control method, apparatus, device, and storage medium. Background Technology

[0002] With the rapid development of new power systems and smart grids, a large number of intelligent switching devices with remote control and status awareness capabilities have been deployed in low-voltage distribution networks to achieve tasks such as refined management of terminal loads, power quality monitoring, and fault isolation and recovery.

[0003] In practical applications, these smart switches typically interact with upper-level systems and execute commands based on multiple communication protocols. However, due to the strong heterogeneity of communication protocols and the difficulty of system integration, the following problems are common in current power distribution automation systems: On the one hand, the large number of switching devices and their complex functional relationships make it difficult for existing methods to effectively mine switch relationship data to divide precise control units, often resulting in ambiguous control scope and inefficient resource scheduling; on the other hand, traditional control methods often rely on fixed logic to generate isolated device control actions, without combining switch operating status data to construct dynamic collaborative control rules, which easily leads to action conflicts or control disconnections; in addition, the lack of a mechanism to integrate collaborative rules and control actions to generate joint commands results in insufficient coordination and reliability of switch control, making it difficult to meet the needs of low-voltage power distribution systems for refined and efficient control. Summary of the Invention

[0004] This invention provides a low-voltage intelligent switch collaborative control method, device, equipment, and storage medium, which can solve the above-mentioned problems and improve the reliability of switch control in low-voltage power distribution scenarios.

[0005] This invention provides a low-voltage intelligent switch collaborative control method, comprising:

[0006] Obtain the task switching devices and current control tasks for the area to be controlled, and determine the switch relationship data and switch operation status data based on the task switching devices;

[0007] Based on the switch relationship data and the task switching devices, a set of control sub-graphs is obtained;

[0008] Based on the switch relationship data and the preset switch cluster construction algorithm, a set of switch clusters is obtained;

[0009] Obtain the collaborative control score of the switch cluster set, and obtain the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and the collaborative control score;

[0010] Based on the current control task, the switch operation status data, and the preset device control action acquisition algorithm, a set of device control actions is obtained;

[0011] Based on the set of collaborative control rules and the set of device control actions, a joint instruction set is obtained;

[0012] The task switching device is controlled based on the aforementioned joint instruction set.

[0013] In the above scheme, by acquiring the switch relationship data and operating status data of the task switching equipment in the area to be controlled, a comprehensive and accurate data foundation is provided for subsequent control decisions. Then, based on the switch relationship data, control sub-graph sets and switch cluster sets are constructed respectively, realizing fine-grained grouping and control unit division of the task switching equipment, effectively avoiding the problems of ambiguous scope and inefficient scheduling in traditional control. At the same time, a set of collaborative control rules is generated by combining the collaborative control score of the switch cluster set with the preset mapping relationship, ensuring that the rules can dynamically adapt to the correlation characteristics between switches and avoid the limitations of fixed logic. Subsequently, a set of equipment control actions is generated based on the switch operating status data, and a joint instruction set is obtained by integrating the collaborative control rules, which can effectively avoid control action conflicts and improve the coordination and rationality of switch control. Finally, the task switching equipment is controlled based on the joint instruction set to execute the current control task, thereby significantly improving the accuracy, efficiency and reliability of switch control in low-voltage power distribution scenarios and meeting the needs of intelligent collaborative control of switching equipment in various scenarios.

[0014] In another embodiment, the preset switch cluster construction algorithm includes a preset matrix construction algorithm and a preset control influence clustering algorithm. The step of obtaining a switch cluster set based on the switch relationship data and the preset switch cluster construction algorithm includes:

[0015] A control influence matrix is ​​constructed based on a preset matrix construction algorithm and the switch relationship data;

[0016] The set of switch clusters is obtained based on the preset control influence clustering algorithm and the control influence matrix.

[0017] In the above scheme, the switch relationship data is quantified into a control influence matrix by introducing a preset matrix construction algorithm, and the matrix is ​​hierarchically clustered based on a preset control influence clustering algorithm. This can automatically identify and output a set of strongly coupled switch clusters. This effectively overcomes the limitations of traditional manual division of collaborative units, realizes objective and dynamic evaluation of the logical relationship between devices, and provides a precise data foundation for the subsequent generation of collaborative control rules.

[0018] In another embodiment, the preset device control action acquisition algorithm includes a running state weighted algorithm and a preset multi-objective reinforcement learning model. The step of obtaining a set of device control actions based on the current control task, the switch running state data, and the preset device control action acquisition algorithm includes:

[0019] Based on the weighted algorithm of the operating state and the switch operating state data, the switch state fusion result is obtained;

[0020] The preset multi-objective reinforcement learning model is updated based on the current control task, and a set of device control actions is obtained based on the updated preset multi-objective reinforcement learning model and the fusion result of the switch state.

[0021] In the above scheme, a weighted algorithm for operating state is used to fuse the heterogeneous switch operating state data of multiple protocols to generate a unified switch state fusion result, which solves the problem of heterogeneous modeling of multi-source information. Then, the weight preferences of the preset multi-objective reinforcement learning model are dynamically adjusted in combination with the current control task, and decisions are made based on the fused state to output a set of equipment control actions. This process realizes the intelligent coupling of state perception and task orientation, enabling the system to dynamically generate the optimal set of equipment control actions under multi-objective constraints.

[0022] In another embodiment, controlling the task switching device based on the joint instruction set includes:

[0023] Obtain the control source node of the area to be controlled, and determine the communication status data based on the task switching device;

[0024] The communication status data is weighted and fused to obtain the communication health score of each task switching device;

[0025] When the communication health score is lower than the preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. Then, a fault tolerance strategy is executed based on the control source node, the target switching device, and the updated control subgraph set to obtain a fault tolerance control instruction, and then the task switching device is controlled based on the fault tolerance control instruction.

[0026] In the above scheme, by acquiring the communication status data of the control source node and continuously monitoring the task switching equipment, the communication health score of each device is calculated, thereby realizing a quantitative evaluation of the control link performance. When the score is lower than the threshold, the system can automatically update the control subgraph, select the target switching equipment according to the joint instruction set, and combine the control source node to execute a fault-tolerant strategy based on graph search to generate fault-tolerant control instructions. This mechanism significantly enhances the reliability in communication anomalies, thereby improving the reliability of switch control in low-voltage power distribution scenarios and ensuring the reliable execution of the current control task.

[0027] In another embodiment, when the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including:

[0028] The task switching devices whose communication health score is lower than a preset health threshold are identified as faulty switching devices.

[0029] The faulty switching device is removed from the control subgraph set to obtain the updated control subgraph set.

[0030] In the above scheme, by logically deleting faulty switching devices with communication health scores below the threshold from the control subgraph set, the abnormal nodes are quickly isolated in the control topology. This operation prevents faulty switching devices from interfering with the execution of the current control task, thereby improving the reliability of switch control in low-voltage power distribution scenarios and ensuring the reliable execution of the current control task.

[0031] In another embodiment, when the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including:

[0032] The reconstructed control path is obtained based on the target switching device and the updated control subgraph set;

[0033] When the reconfiguration control path exists, the first device path cost set from the target switch device to the control source node is obtained, the target switch device path is determined based on the first device path cost set, and then the fault-tolerant control command is obtained based on the target switch device and the target switch device path.

[0034] In the above scheme, after detecting a communication anomaly, a reconstructed control path from the control source node is found for the target switching device based on the updated control subgraph set, and the optimal target switching device path is determined by evaluating the first device path cost set. This process enables the system to automatically and quickly select an alternative communication path when some nodes or links fail, thereby maintaining effective control over the original target switching device.

[0035] In another embodiment, when the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the control source node and the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the target switching device and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on the fault-tolerant control instructions, including:

[0036] When the reconstructed control path does not exist, an equivalent set of switching devices is obtained based on the target switching device and the updated control subgraph set.

[0037] Obtain the second device path cost set from the equivalent set of switching devices to the control source node;

[0038] The target equivalent switching device is selected from the equivalent switching device set based on the second device path cost set;

[0039] The target equivalent device path is determined based on the second device path cost set;

[0040] The fault-tolerant control command is obtained based on the target equivalent switching device and the target equivalent device path.

[0041] In the above scheme, when the control path of the original target switchgear cannot be reconstructed, the system finds a set of functionally equivalent switchgear from the control subgraph and calculates the second device path cost set from each equivalent switchgear to the control source node. Based on this, the optimal target equivalent switchgear and its target equivalent device path are selected. This mechanism realizes functional substitution control in fault conditions, ensuring that even if the original target switchgear is unreachable, the control task can still be completed through the equivalent switchgear, thus improving the reliability of executing the current control task.

[0042] Another embodiment of the present invention provides a low-voltage intelligent switch collaborative control device, comprising:

[0043] The acquisition module is used to acquire the task switching devices and current control tasks of the area to be controlled, and to determine the switch relationship data and switch operation status data based on the task switching devices;

[0044] The control subgraph module is used to obtain a set of control subgraphs based on the switch relationship data and the task switching devices;

[0045] The switch cluster module is used to obtain a set of switch clusters based on the switch relationship data and a preset switch cluster construction algorithm;

[0046] The collaborative control rule module is used to obtain the collaborative control score of the switch cluster set, and to obtain the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and the collaborative control score;

[0047] The device control action module is used to obtain a set of device control actions based on the current control task, the switch operation status data, and a preset device control action acquisition algorithm;

[0048] The joint instruction module is used to obtain a joint instruction set based on the set of collaborative control rules and the set of device control actions;

[0049] The control module is used to control the task switching device based on the joint instruction set.

[0050] Another embodiment of the present invention provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of the low-voltage intelligent switch collaborative control method of the present invention.

[0051] Another embodiment of the present invention provides a computer-readable storage medium item, including: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to perform the steps of a low-voltage intelligent switch collaborative control method of the present invention. Attached Figure Description

[0052] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0053] Figure 1 This is a flowchart illustrating a low-voltage intelligent switch collaborative control method provided in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the structure of a low-voltage intelligent switch collaborative control device provided in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0057] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0058] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0059] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0060] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0061] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0062] See Figure 1To address the aforementioned problems and improve the reliability of switch control in low-voltage power distribution scenarios, an embodiment of the present invention provides a low-voltage intelligent switch collaborative control method, comprising:

[0063] Step S1: Obtain the task switching devices and current control tasks for the area to be controlled, and determine the switch relationship data and switch operation status data based on the task switching devices;

[0064] Step S2: Based on the switch relationship data and the task switching devices, obtain the control sub-diagram set;

[0065] Step S3: Based on the switch relationship data and the preset switch cluster construction algorithm, obtain a set of switch clusters;

[0066] Step S4: Obtain the collaborative control score of the switch cluster set, and obtain the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and collaborative control score;

[0067] Step S5: Based on the current control task, switch operation status data, and preset equipment control action acquisition algorithm, obtain the set of equipment control actions;

[0068] Step S6: Based on the set of collaborative control rules and the set of device control actions, obtain the joint instruction set;

[0069] Step S7: Control the task switching device based on the joint instruction set.

[0070] In the above scheme, by acquiring the switch relationship data and operating status data of the task switching equipment in the area to be controlled, a comprehensive and accurate data foundation is provided for subsequent control decisions. Then, based on the switch relationship data, control sub-graph sets and switch cluster sets are constructed respectively, realizing fine-grained grouping and control unit division of the task switching equipment, effectively avoiding the problems of ambiguous scope and inefficient scheduling in traditional control. At the same time, a set of collaborative control rules is generated by combining the collaborative control score of the switch cluster set with the preset mapping relationship, ensuring that the rules can dynamically adapt to the correlation characteristics between switches and avoid the limitations of fixed logic. Subsequently, a set of equipment control actions is generated based on the switch operating status data, and a joint instruction set is obtained by integrating the collaborative control rules, which can effectively avoid control action conflicts and improve the coordination and rationality of switch control. Finally, the task switching equipment is controlled based on the joint instruction set to execute the current control task, thereby significantly improving the accuracy, efficiency and reliability of switch control in low-voltage power distribution scenarios and meeting the needs of intelligent collaborative control of switching equipment in various scenarios.

[0071] Furthermore, regarding step S1: obtaining the task switching devices of the area to be controlled, and determining switch relationship data and switch operating status data based on the task switching devices; and step S2: obtaining a control sub-graph set based on the switch relationship data and the task switching devices, specifically:

[0072] All low-voltage smart switches in the area to be controlled are regarded as task switching devices. The task switching devices are connected by edges based on the physical wiring of the power grid, communication network and other relationships to form an overall graph. Then, according to the switch relationship data, including spatial location, physical circuit or functional affiliation, the task switching devices are divided into multiple control subgraphs to obtain a set of control subgraphs.

[0073] In another embodiment, the preset switch cluster construction algorithm includes a preset matrix construction algorithm and a preset control influence clustering algorithm. The step of obtaining a switch cluster set based on the switch relationship data and the preset switch cluster construction algorithm includes:

[0074] A control influence matrix is ​​constructed based on a preset matrix construction algorithm and the switch relationship data;

[0075] The set of switch clusters is obtained based on the preset control influence clustering algorithm and the control influence matrix.

[0076] It should be noted that for each control subgraph in the set of control subgraphs, a control influence matrix is ​​constructed and hierarchical clustering is performed. Based on the clustering results, a set of strongly coupled switch clusters is output, and each switch cluster is defined as a cooperative control unit. Each cooperative control unit, as an independent control entity, participates in control strategy generation and scheduling decisions. It is understandable that obtaining the set of switch clusters first requires constructing a control influence matrix, with each control subgraph corresponding to an M... nxn Where n is the number of switches in the control sub-diagram, and M ij The matrix construction algorithm is used to represent the degree of control influence between task switching devices i and j.

[0077] M ij =λ1·f 共事件 (i,j)+λ2·f 通信频度 (i,j)+λ3·f 故障联动历史 (i,j);

[0078] Among them, f 共事件 f represents the number of times that the switching devices for tasks i and j have been simultaneously controlled or participated in the same strategy. 通信频度 f is the frequency of the interaction commands between the switching devices for tasks i and j. 故障联动历史 For the switching devices of tasks i and j, are there any records of linkage fault recovery? λ k These are empirical weighting coefficients.

[0079] Based on the degree of control influence between any two task switching devices in the control subgraph, a control influence matrix is ​​obtained. Then, a preset control influence clustering algorithm is used to cluster the control influence matrix. The preset control influence clustering algorithm adopts an existing hierarchical clustering algorithm, such as hierarchical clustering, and outputs several strongly coupled switch clusters, resulting in a set of switch clusters. For example, control subgraph A is divided into 3 switch clusters: A1, A2, and A3.

[0080] Furthermore, regarding the empirical weighting coefficient λ k This invention proposes two weighting mechanisms: a static empirical weighting mechanism, which presets empirical weighting coefficients based on system type; and a dynamic empirical weighting coefficient optimization process, which updates the empirical weighting coefficients based on the existing entropy weighting method and the existing fuzzy hierarchical analysis method, to address the constantly changing collaborative relationships during equipment operation. This dynamic empirical weighting coefficient optimization process can be automatically updated periodically (e.g., every 24 hours or every thousand control events) to maintain the real-time sensitivity of the control influence matrix to the system state.

[0081] Furthermore, regarding step S4: obtaining the collaborative control score of the switch cluster set, and obtaining the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and the collaborative control score, specifically:

[0082] Constructing fuzzy rules R k The expression is as follows:

[0083]

[0084] in, Let y be the fuzzy linguistic term for the nth condition in the kth rule, and B be the output variable of the fuzzy inference. k This is the output fuzzy linguistic term for the k-th rule; this expression represents a rule consisting of an IF condition part and a THEN conclusion part. In the condition part: multiple sub-conditions are connected by "AND" (∧). In the conclusion part: the output y equals a certain fuzzy linguistic value B. k .

[0085] Perform fuzzy inference (i,j)∈D (where D is any cluster of switches) on task switching devices i and j, and output the corresponding cooperative control score S. ij :

[0086]

[0087] Where, α k For the preset rule confidence level, w k Output the rule's level value, where K is the total number of rules in the rule base, and w k With Bk There are preset mapping relationships between them. For example, if THEN collaboration level = high, then the corresponding rule output level value = 1.0. For control variables or state variables x n Corresponding fuzzy language terms The degree of membership, i.e. It is to fuzzy language items Convert it into a specific numerical value. and There are pre-defined mapping relationships between them.

[0088] The coordinated control score S for each task switching device pair (i,j) ij Based on the set hierarchical classification, collaborative control rules are generated. This means that each task switching device is automatically matched with a corresponding control rule template according to a preset switch control mapping relationship, forming a formal set of collaborative control rules corresponding to the switch cluster set. During system operation, the rule confidence α can be dynamically adjusted according to actual needs through reinforcement learning or fuzzy weight correction algorithms. kSpecifically: After calculating the collaborative control score between each pair of task switching devices, the collaborative control score is classified into levels according to a preset multi-level threshold range. For example, the collaborative control score range is divided into three levels: "weak correlation (0~0.3)", "medium collaboration (0.3~0.7)" and "strong coupling (0.7~1.0)". For the collaborative control score of different levels, the rule patterns in the corresponding control rule template library are automatically matched according to the preset switch control mapping relationship. Specifically, when the collaborative control score of a pair of task switching devices is at the "strong coupling" level, a "linkage interlocking" rule template is matched to generate mandatory linkage or mutual exclusion control rules such as "If switch A is open, then switch B must be opened synchronously" or "A and B are backups for each other, and only one is allowed to be in the closed state." When the collaborative control score is at the "medium coordination" level, a "sequential execution" or "delayed coordination" rule template is matched to generate coordination rules with time dependence or condition constraints, such as "When the main circuit switch is opened, the branch line switches should be opened sequentially within 5 seconds" or "Before load transfer, it is necessary to confirm that the communication status of the tie switch is normal." For a pair of task switching devices with a collaborative control score at the "weak correlation" level, an "independent control" or "alarm reminder" template is matched. No mandatory linkage command is generated; only a related risk reminder is issued during operation. The rule template library has built-in logical structures, action types, and constraints corresponding to each coordination level and supports expansion and optimization based on field operation experience. The system compares the collaborative control scores of task device pairs within their respective grade ranges, automatically calls the corresponding templates based on preset switch control mapping relationships, and fills in the actual device names, action types, and parameter thresholds. This generates a complete and semantically clear set of formal collaborative control rules, achieving an automated conversion from quantitative evaluation to executable rules, ensuring the transparency, interpretability, and engineering practicality of the rule generation process.

[0089] In another embodiment, the preset device control action acquisition algorithm includes a running state weighted algorithm and a preset multi-objective reinforcement learning model. The step of obtaining a set of device control actions based on the current control task, the switch running state data, and the preset device control action acquisition algorithm includes:

[0090] Based on the weighted algorithm of the operating state and the switch operating state data, the switch state fusion result is obtained;

[0091] The preset multi-objective reinforcement learning model is updated based on the current control task, and a set of device control actions is obtained based on the updated preset multi-objective reinforcement learning model and the fusion result of the switch state.

[0092] It should be noted that, based on the attention weight α ij Value vector V for all switch operating status dataj We perform weighted summation to obtain the fusion result of the switch states. The weighted algorithm for the running states is as follows:

[0093]

[0094] in, The fusion result of the i-th switch state is obtained by fusing the i-th predefined target device or control point, where n is the total number of task switch devices currently participating in the fusion, and V j This is a vector of the switching operation status data of the j-th task switching device. The predefined target device or control point can be any task switching device, typically the core task switching device that makes control decisions and outputs control actions based on this fused state vector.

[0095] Furthermore, the attention weight α in the running state weighted algorithm ij By constructing a Query-Key-Value mechanism, for the i-th target device or control point, the attention weight α corresponding to the switching operation status data of each task switching device is calculated. ij :

[0096]

[0097] Among them, Q i K is the query vector for the i-th preset control point or target device. j This is a vector of the switching operation status data of the j-th task switching device.

[0098] Furthermore, a pre-defined existing multi-objective reinforcement learning model (i.e., the policy network in existing reinforcement learning) is constructed. At each time step, the input switch state fusion result is fed into the pre-defined multi-objective reinforcement learning model, and the output action probability distribution or action value estimate is used to select the optimal set of control actions according to the action policy. This involves constructing a pre-defined multi-objective reinforcement learning model (i.e., the policy network in existing technology). θ (·), receive the fusion result of the switch state Output the probability distribution or action value estimate of each device control action, where the preset multi-objective reinforcement learning model is:

[0099]

[0100] Here, 'a' represents the control operation of the task switching device. The preset multi-objective reinforcement learning model supports the output of control actions for multiple devices, such as task switching device 1 remaining on; task switching device 2 briefly turning off; and task switching device 3 delaying its action while waiting for feedback.

[0101] Based on the action execution result, the next state and multi-objective reward signal are returned. Using the sampled state, action, reward, and next state data, the parameters of the preset multi-objective reinforcement learning model are iteratively updated according to the multi-objective reinforcement learning algorithm to optimize policy performance until the preset multi-objective reinforcement learning model completes training. Based on the trained preset multi-objective reinforcement learning model, a set of device control actions is obtained. It is understood that in this process, updating the preset multi-objective reinforcement learning model based on the current control task refers to determining the decision preferences or objective weights of the multi-objective reinforcement learning model based on the current control task, i.e., the relative importance weights of each objective.

[0102] Furthermore, for step S6, based on the set of collaborative control rules and the set of device control actions, a joint instruction set is obtained, specifically:

[0103] Based on the established set of collaborative control rules and the set of equipment control actions, a joint instruction set containing multiple task switching equipment operations is generated: A pre-established set of collaborative control rules is obtained, including the linkage conditions, priorities, and action triggering logic between task switching equipment; a set of equipment control actions dynamically generated by a pre-defined multi-objective reinforcement learning model at the current time step is obtained; for each equipment control action in the set of equipment control actions, the corresponding task switching equipment ID, action type, and action parameters are identified, and classified according to the nature of the action, distinguishing between independent actions of a single task switching equipment and actions involving collaborative actions of multiple task switching equipment; for control actions involving collaborative actions of multiple task switching equipment, matching is performed according to the set of collaborative control rules to determine whether the joint triggering conditions are met; actions that meet the collaborative conditions in the set of collaborative control rules are merged and coordinated to form a joint operation instruction; actions that do not meet the collaborative rules in the set of collaborative control rules remain as independent instructions of a single task switching equipment. Specifically: Based on the above-mentioned construction of a collaborative control rule set, the collaborative control rule set includes: the linkage relationship between task switching devices (e.g., task switching device a and task switching device b must be linked); linkage conditions (e.g., specific time, state, scenario, etc.); control priority (e.g., task switching device a takes precedence over task switching device b); action constraints (e.g., "delay", "mutual exclusion", "sequence"); simultaneously, based on the current time step, a set of device control actions is output by a preset multi-objective reinforcement learning model (policy network); the set of device control actions is parsed item by item to extract key information: whether the device control action has constraint parameters, such as whether it is delayed; whether multiple device control actions point to the same task switching device; whether it is a mutual exclusion operation (e.g., two actions attempt to simultaneously turn on and off the same switch); execution on independent task devices does not require coordination; there may be collaborative logic between actions, which needs to be matched with the rule base to determine whether they can be merged. For task switching device pairs in the collaborative control rule set, rule matching is performed: checking for the existence of valid control rule entries; determining whether the triggering conditions in the rules are met (e.g., time period, communication health, environmental status); determining whether there are control conflicts; if the current system status meets the rule conditions, the action group can be merged to form a collaborative control unit. Actions that meet the collaborative conditions in the collaborative control rule set are integrated to generate a structured joint instruction set: including a list of task switching device IDs, control sequences, and parameter dependencies (e.g., delay time); adding scheduling tags, such as group number, execution context, and dependencies; and rationally sorting the instruction order, handling dependency constraints and concurrency requirements. It is understandable that the task switching devices in the joint instruction set at this point include not only predefined target devices or control points, but also other task switching devices connected to them and required to execute the current control task.

[0104] In another embodiment, controlling the task switching device based on the joint instruction set includes:

[0105] Obtain the control source node of the area to be controlled, and determine the communication status data based on the task switching device;

[0106] The communication status data is weighted and fused to obtain the communication health score of each task switching device;

[0107] When the communication health score is lower than the preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. Then, a fault tolerance strategy is executed based on the control source node, the target switching device, and the updated control subgraph set to obtain a fault tolerance control instruction, and then the task switching device is controlled based on the fault tolerance control instruction.

[0108] It should be noted that after obtaining the joint instruction set, the task switching equipment can be directly controlled. However, to further improve the reliability of the task switching equipment control, the control source node of the area to be controlled is obtained, and the communication status data of the task switching equipment is acquired. The communication status data includes status reporting frequency, instruction response capability, and communication link stability. The communication status data is normalized, and the normalized communication status data is then weighted and fused to obtain the communication health score H. i :

[0109]

[0110] Where w1, w2, and w3 are preset weighting coefficients for the scoring of each communication status data dimension. This represents the normalized state reporting frequency. For normalized instruction response capability, L i To ensure the stability of the normalized communication link, during long-term operation, the communication status data of each task switching device is recorded. The correlation coefficient of each communication status data point is periodically calculated, and the weighting coefficients are dynamically adjusted based on their contribution. For example, if historical data indicates that "unstable communication link" is the main source of failure, w3 will automatically increase; if "response delay" has a greater impact on overall control performance, w2 will increase. This achieves dynamic redistribution of the weighting coefficients for communication health, improving system adaptability and accuracy. When the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching devices, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the target switching device and the updated control subgraph set to obtain fault-tolerant control instructions, which are then used to control the task switching devices.

[0111] It should be further noted that the system updates the H of each task switching device at a fixed time window (e.g., every minute). i It also maintains a recent one-hour time-series data for each task switching device in memory for trend identification and health analysis. A preset health threshold H is set. 阈值 (e.g., 0.6). When a communication health score H of a certain task switching device is detected. i When the health score falls below the preset health threshold, a fault-tolerant recovery mechanism is triggered. Upon triggering fault tolerance, the system first updates the control subgraph set based on the task switching devices: task switching devices with communication health scores below the preset health threshold are marked as having communication anomalies and logically removed from the control subgraph set, forming a control subgraph set to be reconstructed. Simultaneously, based on the control task context defined by the joint instruction set, a target switching device (i.e., the specific task switching device that the current control task needs to operate) is explicitly selected from the task switching devices. Subsequently, based on the target switching device and the updated control subgraph set, a fault-tolerant strategy is executed to generate a fault-tolerant control instruction, which is then used to control the task switching device to complete or replace the original control objective.

[0112] In another embodiment, when the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including:

[0113] The task switching devices whose communication health score is lower than a preset health threshold are identified as faulty switching devices.

[0114] The faulty switching device is removed from the control subgraph set to obtain the updated control subgraph set.

[0115] It should be noted that "deletion" refers to logical removal, which means marking the node corresponding to the faulty switch and its connected edges as "unavailable" from the control subgraph set used for path planning, thereby forming a new control subgraph set to be reconstructed. This is not physically deleting the device, but rather excluding it from subsequent path searches to ensure that control commands are not transmitted via unreliable nodes or links.

[0116] In another embodiment, when the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including:

[0117] The reconstructed control path is obtained based on the target switching device and the updated control subgraph set;

[0118] When the reconfiguration control path exists, the first device path cost set from the target switch device to the control source node is obtained, the target switch device path is determined based on the first device path cost set, and then the fault-tolerant control command is obtained based on the target switch device and the target switch device path.

[0119] It should be noted that the control source node s k This refers to the source that issues control commands to the task switching equipment, which can be determined based on the current control task scheduling information. The existence of a reconfigurable control path means that the control path can be reconfigured; that is, even if the faulty switching equipment is deleted from the updated control subgraph set, the target switching equipment can still be controlled through other control paths. When searching for a path, the edges in the control subgraph set are assigned weights w based on the communication health score. ij (For example, the weights can be negatively correlated with link health), for each target switching device d k From the control source node s k To reach the target switchgear, perform the following operations on the updated control subgraph set G′:

[0120]

[0121] Among them, C path e is the total cost of the path. ij Let `path` be the set of edges in the control subgraph, and `path` be the set of control paths under consideration. When the reconstructed control path exists (i.e., at least one feasible path is found), the system obtains the costs of all possible paths, forming a first set of device path costs for the target switching device. Based on this first set of device path costs, a classic shortest path algorithm (such as Dijkstra's algorithm) is used to search for edges from `s` in the updated control subgraph set `G′`. k To the target switchgear d k The path corresponding to the minimum device path cost determines the final target switchgear path.

[0122] It should be further explained that if multiple minimum-cost paths with the same or similar costs exist, the system will introduce a priority-weighted scoring function for judgment: First, the system sorts the paths according to the preset control priorities of the task switching devices, prioritizing the allocation of the best path to the task switching devices with higher priorities; if the priorities are the same, the system further selects the path with the higher communication health score and fewer hops. After determining the target switching device path, the system generates a new control command sequence based on the target switching device and the path, which is the fault-tolerant control command. The above is the fault-tolerant strategy when a reconstructed control path exists.

[0123] In another embodiment, when the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including:

[0124] When the reconstructed control path does not exist, an equivalent set of switching devices is obtained based on the target switching device and the updated control subgraph set.

[0125] Obtain the second device path cost set from the equivalent set of switching devices to the control source node;

[0126] The target equivalent switching device is selected from the equivalent switching device set based on the second device path cost set;

[0127] The target equivalent device path is determined based on the second device path cost set;

[0128] The fault-tolerant control command is obtained based on the target equivalent switching device and the target equivalent device path.

[0129] It should be noted that "the control path does not exist" means that the control path cannot be reconstructed. This means that after the updated control subgraph set removes the faulty switchgear, the target switchgear cannot be controlled through other control paths, or the deleted faulty switchgear is the target switchgear. In this case, it is necessary to obtain a set of equivalent switchgear equivalent to the target switchgear. "Equivalent" refers to functional substitutability, such as controlling the same load area, being in a similar topological location, or having the same operational capabilities. The equivalent switchgear set is pre-set based on the power grid topology analysis, the functional attribute labeling of the task switchgear, and historical operation records. When obtaining the equivalent switchgear set corresponding to the target switchgear, it is important to note that the equivalent switchgear set should be obtained based on the updated control subgraph set to avoid selecting a deleted task switchgear. Subsequently, for each equivalent switchgear in the equivalent switchgear set, the above path search process is repeated: that is, the path from the control source node to the equivalent switchgear is searched in the updated control subgraph set, and the path cost is calculated to obtain the costs of all paths corresponding to all equivalent switchgears, forming a second set of device path costs. Based on the second set of device path costs, the device with the lowest path cost is selected from the set of equivalent switching devices as the target equivalent switching device, and its corresponding path is determined as the target equivalent device path. Finally, based on the target equivalent switching device and the target equivalent device path, an alternative control instruction is generated, which is the fault-tolerant control instruction. This allows the original current control task to be completed through the task switching device with equivalent control function (i.e., the target equivalent switching device). This is the fault-tolerant strategy when the reconstructed control path does not exist.

[0130] like Figure 2 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided;

[0131] An embodiment of the present invention provides a low-voltage intelligent switch collaborative control device, comprising:

[0132] The acquisition module is used to acquire the task switching devices and current control tasks of the area to be controlled, and to determine the switch relationship data and switch operation status data based on the task switching devices;

[0133] The control subgraph module is used to obtain a set of control subgraphs based on the switch relationship data and the task switching devices;

[0134] The switch cluster module is used to obtain a set of switch clusters based on the switch relationship data and a preset switch cluster construction algorithm;

[0135] The collaborative control rule module is used to obtain the collaborative control score of the switch cluster set, and to obtain the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and the collaborative control score;

[0136] The device control action module is used to obtain a set of device control actions based on the current control task, the switch operation status data, and a preset device control action acquisition algorithm;

[0137] The joint instruction module is used to obtain a joint instruction set based on the set of collaborative control rules and the set of device control actions;

[0138] The control module is used to control the task switching device based on the joint instruction set.

[0139] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the low-voltage intelligent switch collaborative control method provided by any of the above-described method embodiments of the present invention.

[0140] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0141] Based on the above-described embodiment of a low-voltage intelligent switch collaborative control method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a low-voltage intelligent switch collaborative control method according to any embodiment of the present invention.

[0142] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0143] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0144] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0145] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute a low-voltage intelligent switch collaborative control method as described in any of the above-described method embodiments of the present invention.

[0146] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0147] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A low-voltage intelligent switch collaborative control method, characterized in that, include: Obtain the task switching devices and current control tasks for the area to be controlled, and determine the switch relationship data and switch operation status data based on the task switching devices; Based on the switch relationship data and the task switching devices, a set of control sub-graphs is obtained; Based on the switch relationship data and the preset switch cluster construction algorithm, a set of switch clusters is obtained; Obtain the collaborative control score of the switch cluster set, and obtain the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and the collaborative control score; Based on the current control task, the switch operation status data, and the preset device control action acquisition algorithm, a set of device control actions is obtained; Based on the set of collaborative control rules and the set of device control actions, a joint instruction set is obtained; The task switching device is controlled based on the aforementioned joint instruction set.

2. The low-voltage intelligent switch collaborative control method according to claim 1, characterized in that, The preset switch cluster construction algorithm includes a preset matrix construction algorithm and a preset control influence clustering algorithm. Based on the switch relationship data and the preset switch cluster construction algorithm, a set of switch clusters is obtained, including: A control influence matrix is ​​constructed based on a preset matrix construction algorithm and the switch relationship data; The set of switch clusters is obtained based on the preset control influence clustering algorithm and the control influence matrix.

3. The low-voltage intelligent switch collaborative control method according to claim 1, characterized in that, The preset device control action acquisition algorithm includes a running state weighted algorithm and a preset multi-objective reinforcement learning model. Based on the current control task, the switch running state data, and the preset device control action acquisition algorithm, a set of device control actions is obtained, including: Based on the weighted algorithm of the operating state and the switch operating state data, the switch state fusion result is obtained; The preset multi-objective reinforcement learning model is updated based on the current control task, and a set of device control actions is obtained based on the updated preset multi-objective reinforcement learning model and the fusion result of the switch state.

4. The low-voltage intelligent switch collaborative control method according to claim 1, characterized in that, The control of the task switching device based on the joint instruction set includes: Obtain the control source node of the area to be controlled, and determine the communication status data based on the task switching device; The communication status data is weighted and fused to obtain the communication health score of each task switching device; When the communication health score is lower than the preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. Then, a fault tolerance strategy is executed based on the control source node, the target switching device, and the updated control subgraph set to obtain a fault tolerance control instruction, and then the task switching device is controlled based on the fault tolerance control instruction.

5. The low-voltage intelligent switch collaborative control method according to claim 4, characterized in that, When the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including: The task switching devices whose communication health score is lower than a preset health threshold are identified as faulty switching devices. The faulty switching device is removed from the control subgraph set to obtain the updated control subgraph set.

6. The low-voltage intelligent switch collaborative control method according to claim 4, characterized in that, When the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including: The reconstructed control path is obtained based on the target switching device and the updated control subgraph set; When the reconfiguration control path exists, the first device path cost set from the target switch device to the control source node is obtained, the target switch device path is determined based on the first device path cost set, and then the fault-tolerant control command is obtained based on the target switch device and the target switch device path.

7. The low-voltage intelligent switch collaborative control method according to claim 6, characterized in that, When the communication health score is lower than a preset health threshold, the control subgraph set is updated based on the task switching device, and a target switching device is selected from the task switching devices based on the joint instruction set. A fault-tolerant strategy is then executed based on the control source node, the target switching device, and the updated control subgraph set to obtain fault-tolerant control instructions. Finally, the task switching device is controlled based on these fault-tolerant control instructions, including: When the reconstructed control path does not exist, an equivalent set of switching devices is obtained based on the target switching device and the updated control subgraph set; Obtain the second device path cost set from the equivalent set of switching devices to the control source node; The target equivalent switching device is selected from the equivalent switching device set based on the second device path cost set; The target equivalent device path is determined based on the second device path cost set; The fault-tolerant control command is obtained based on the target equivalent switching device and the target equivalent device path.

8. A low-voltage intelligent switch collaborative control device, characterized in that, include: The acquisition module is used to acquire the task switching devices and current control tasks of the area to be controlled, and to determine the switch relationship data and switch operation status data based on the task switching devices; The control subgraph module is used to obtain a set of control subgraphs based on the switch relationship data and the task switching devices; The switch cluster module is used to obtain a set of switch clusters based on the switch relationship data and a preset switch cluster construction algorithm; The collaborative control rule module is used to obtain the collaborative control score of the switch cluster set, and to obtain the collaborative control rule set corresponding to the switch cluster set based on the preset switch control mapping relationship and the collaborative control score; The device control action module is used to obtain a set of device control actions based on the current control task, the switch operation status data, and a preset device control action acquisition algorithm; The joint instruction module is used to obtain a joint instruction set based on the set of collaborative control rules and the set of device control actions; The control module is used to control the task switching device based on the joint instruction set.

9. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements a low-voltage intelligent switch collaborative control method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform a low-voltage intelligent switch cooperative control method as described in any one of claims 1-7.

Images