Substation intelligent inspection auxiliary system in electric power internet of things environment

Abstract

The invention discloses an intelligent substation inspection auxiliary system in an electric power internet of things environment, and relates to the technical field of intelligent inspection control. A development unit is used for acquiring infrastructures of a target substation and developing an intelligent inspection module by defining a task scene-inspection topology; the first-class inspection control unit is used for judging a first-class inspection task, switching a target inspection topology and executing topology-driven facility inspection control under a forward protocol; the second-class inspection control unit is used for transmitting self-inspection risk data to the edge computing gateway, triggering lightweight inspection analysis embedded in the gateway and executing topology-driven facility inspection control under a reverse protocol, and is used for solving the problem that in the prior art, it is difficult to achieve active sensing, quick response and risk closed-loop management, and the problem that in the prior art, it is difficult to carry out topology-driven facility inspection control is solved. The technical problem that the inspection efficiency and precision cannot be considered on the basis of self-adaptive flexible deployment is solved, active sensing, quick response and risk closed-loop management are achieved, and the inspection efficiency and precision are considered on the basis of self-adaptive flexible deployment.

Description

Technical Field

[0001] The present invention relates to the field of intelligent inspection control technology, and in particular to an intelligent inspection auxiliary system for substations in an electric power Internet of Things environment. Background Art

[0002] In existing power system inspection practices, substation inspections primarily rely on manual fixed-point inspections or semi-automated robotic inspections. While manual inspections offer a degree of flexibility and judgment, they are limited by human resources, operating environments, and task complexity. To improve efficiency, some systems have introduced image recognition, edge computing, or IoT sensing devices, but the following technical bottlenecks remain: First, existing systems often employ static task models, lacking the ability to dynamically adapt inspection tasks to device topologies, making it difficult to cope with complex scenarios and emergencies. Second, inspection paths and priorities cannot be adaptively adjusted. Third, current communication protocols are mostly based on traditional passive transmission architectures, making it difficult to support efficient control tasks.

[0003] In summary, there is currently a lack of a fully functional intelligent inspection system for substations, which makes it difficult to achieve active perception, rapid response and closed-loop risk management, and cannot ensure that inspection efficiency and accuracy are taken into account on the basis of adaptive and flexible deployment. Summary of the Invention

[0004] This application provides an intelligent inspection assistance system for substations in the power Internet of Things environment, which is used to solve the technical problems in the existing technology that it is difficult to achieve active perception, rapid response and risk closed-loop management, and cannot ensure both inspection efficiency and accuracy on the basis of adaptive and flexible deployment.

[0005] In view of the above problems, the present application provides a substation intelligent inspection assistance system in an electric power Internet of Things environment, the system including: a development unit for acquiring the infrastructure of the target substation, and developing an intelligent inspection module in the substation intelligent inspection assistance system by defining the task scenario-inspection topology; a first-class inspection control unit for, if an inspection task is uploaded to the intelligent inspection module, it is judged to be a first-class inspection task, and through task interpretation and scenario-based topology matching, it switches to the target inspection topology, assists the synaptic Internet of Things protocol, and executes topology-driven facility inspection control under the forward protocol; a second-class inspection control unit for, if there is risk data from self-inspection of the inspection infrastructure, transmits it to the edge computing gateway, reversely infers the risk event and generates a second-class inspection task, triggers the lightweight inspection module embedded in the gateway, determines the event inspection topology and executes topology-driven facility inspection control under the reverse protocol; wherein, the inspection topology is optimized and adjusted through self-healing networking.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The embodiment of the present application provides an intelligent inspection auxiliary system for substations in an electric power Internet of Things environment. The development unit is used to obtain the infrastructure of the target substation, and develop an intelligent inspection module in the intelligent inspection auxiliary system of the substation by defining the task scenario-inspection topology; the first-class inspection control unit is used to, if an inspection task is uploaded to the intelligent inspection module and is determined to be a first-class inspection task, switch to the target inspection topology through task interpretation and scenario-based topology matching, assist the synaptic IoT protocol, and execute topology-driven facility inspection control under the forward protocol; the second-class inspection control unit is used to, if there is risk data from self-inspection of the inspection infrastructure, transmit it to the edge computing gateway, reversely infer the risk event and generate a second-class inspection task, trigger the lightweight inspection module embedded in the gateway, determine the event inspection topology and execute topology-driven facility inspection control under the reverse protocol, which is used to solve the technical problems in the existing technology that it is difficult to achieve active perception, rapid response and risk closed-loop management, and cannot ensure that inspection efficiency and accuracy are balanced on the basis of adaptive and flexible deployment, so as to achieve active perception, rapid response and risk closed-loop management, and balance inspection efficiency and accuracy on the basis of adaptive and flexible deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A logical flow diagram of the substation intelligent inspection auxiliary system in the power Internet of Things environment is provided for this application;

[0009] Figure 2 This application provides a structural diagram of the substation intelligent inspection auxiliary system in the power Internet of Things environment.

[0010] Description of the reference numerals: development unit 11 , first-class patrol inspection control unit 12 , second-class patrol inspection control unit 13 . DETAILED DESCRIPTION

[0011] This application provides an intelligent inspection assistance system for substations in the power Internet of Things environment to solve the technical problems in the existing technology that it is difficult to achieve active perception, rapid response and risk closed-loop management, and cannot ensure both inspection efficiency and accuracy on the basis of adaptive and flexible deployment.

[0012] Example: Figure 1 、 Figure 2 As shown, the present application provides a substation intelligent inspection auxiliary system in the power Internet of Things environment, the system comprising:

[0013] The development unit 11 is used to obtain the infrastructure of the target substation, and develop an intelligent inspection module in the substation intelligent inspection auxiliary system by defining a task scenario-inspection topology.

[0014] In an embodiment of the present application, the infrastructure of the target substation is acquired, including various substation equipment, access terminals, and their physical connection structures used for power transmission, control, measurement, and protection. In one feasible embodiment, the development unit 11 establishes a data connection with a substation asset management system or a real-time monitoring platform, accesses structured equipment archives and operating status information, and thereby obtains a complete substation facility inventory and layout topology. Preferably, the acquisition process supports data synchronization and dynamic update functions to ensure that the acquired equipment information remains consistent with actual on-site conditions.

[0015] After acquiring infrastructure data, development unit 11 defines a task scenario—an inspection topology. Specifically, a task scenario is a set of task execution conditions built around a specific inspection objective, such as equipment aging detection, high-temperature anomaly warning, or communication status verification. The inspection topology refers to the logical structure and control paths between inspection objects within a given scenario, reflecting the dependencies and action sequences between devices during the inspection process.

[0016] Optionally, the mapping relationship between the task scenario and the inspection topology is established through a rule engine. In a specific implementation process provided by this application, the development unit 11 divides the inspection units according to the substation infrastructure type, such as GIS interval, busbar connection, main transformer section, etc., extracts the associated equipment set of each unit, constructs the corresponding logical node, and generates a set of executable inspection topology structures.

[0017] Furthermore, in order to improve the intelligence and practicality of the task scenario-inspection topology construction, the development unit 11 further introduces multi-source data such as inspection history records, equipment health files and alarm data for fusion analysis.

[0018] For example, if historical fluctuations in the main transformer's oil temperature are detected alongside three-phase current imbalance, a task scenario for early warning of main transformer anomalies can be generated, along with a matching inspection topology with the main transformer, bushings, cooling fans, and monitoring units as core nodes. This topology reflects which inspection facilities should be activated in this specific scenario, and which facilities should be prioritized for tasks such as motion recognition, temperature sampling, or status verification.

[0019] Finally, based on the mapping of the task scenario to the inspection topology, an intelligent inspection module is constructed. In this application, this module serves as the core unit of the system execution layer, including functions such as inspection task scheduling, path control, data interaction and risk feedback, and supports dynamic topology switching and task evolution configuration during subsequent inspections.

[0020] Through the above method, a comprehensive deployment process from static facility data to dynamic task execution logic is realized, providing basic support for the operation of the substation intelligent inspection auxiliary system.

[0021] Furthermore, the development unit 11 includes:

[0022] A facility classification unit is used to determine, for the infrastructure, inspection infrastructure and substation infrastructure, wherein the inspection infrastructure is identified with an inspection domain based on the substation infrastructure, and the inspection infrastructure includes substation access facilities and dynamic inspection facilities; a scenario reconstruction unit is used to perform topology reconstruction based on the substation scenario for the inspection infrastructure, and define the task scenario-inspection topology.

[0023] In an embodiment of the present application, the facility classification unit is used to perform a structured classification operation on the infrastructure to achieve a clear division of functional roles of various types of equipment in the substation.

[0024] Specifically, the infrastructure includes power equipment (e.g., circuit breakers, disconnectors, voltage transformers, transformers, etc.), control and measurement devices, communication nodes, and their installation environment information. The facility classification unit first divides the infrastructure into two categories: substation infrastructure and inspection infrastructure, based on the functional attributes, layout location, and communication interface characteristics of the equipment.

[0025] Among them, substation infrastructure refers to the core equipment responsible for the transmission and control of primary and secondary system power, and its functions do not have mobility or active perception, such as busbars, main transformers, switch cabinets, etc.; while inspection infrastructure refers to ancillary facilities with status collection, information transmission or inspection execution capabilities, including but not limited to access terminals, sensing nodes, mobile inspection equipment (for example: inspection robots, drones), etc.

[0026] Furthermore, after completing the classification, the facility classification unit further identifies the inspection domain corresponding to the inspection infrastructure, that is, the monitoring range of the substation infrastructure covered by the facility.

[0027] In an optional embodiment, the inspection domain is dynamically defined based on the facility coverage radius, sensor parameter matching, and physical deployment path. For example, a wireless inspection facility connected to a GIS bay has an inspection domain that includes the circuit breakers, transformers, and switches within that bay.

[0028] Specifically, inspection infrastructure includes two categories: substation access facilities, such as fixed cameras, infrared thermal imaging modules, wiring terminals, etc., which are mainly used for local status monitoring and signal aggregation; dynamic inspection facilities, such as rail robots, wall-climbing detection devices, flying inspection units, etc., which have the ability to move between multiple inspection domains and perform tasks, and are suitable for large-scale or high-frequency task scenarios.

[0029] Based on this, the scenario reconstruction unit then performs a topology reconstruction operation on the inspection infrastructure based on the substation scenario. A substation scenario refers to a set of task environments formed under specific operating conditions or warning backgrounds, such as equipment insulation status inspection after a thunderstorm or busbar system re-inspection after a short-term trip.

[0030] Optionally, the scenario reconstruction unit, based on scenario-based characteristics, explores the structure and logic of the inspection infrastructure availability in different inspection scenarios, achieving a topological cascade reconstruction of the scenario-based inspection coverage relationship, and forming a logical inspection structure for the task scenario. The topology reconstruction process reflects the strength of the association between devices, the priority of the control path, and the execution process sequence, ensuring that the generated inspection topology reflects both the physical structure and the task logic.

[0031] Finally, the scenario reconstruction unit converts the topology reconstruction results into a task scenario-inspection topology mapping, which serves as the core basis for subsequent inspection control and module generation. This facility classification and scenario topology reconstruction process achieves a logical transition from a static structure to a dynamic task-driven approach, improving the scenario-based compatibility and configuration flexibility of the inspection-driven approach.

[0032] Furthermore, the scene reconstruction unit includes:

[0033] The clustering unit is used to retrieve the inspection records of the target substation, perform scenario-based clustering, and determine M record groups; the mining unit is used to mine the scenario inspection facilities-inspection orientation within the M record groups, with the preset ratio within the group as a constraint; the determination unit is used to determine the inspection topology based on the scenario inspection facilities-inspection orientation; the association unit is used to define the task scenario for the M record groups and perform the association between the task scenario and the inspection topology.

[0034] In this embodiment of the present application, the clustering unit is used to retrieve inspection record data generated by the target substation during different operating cycles and perform scenario-based classification. The inspection record includes, but is not limited to, multi-dimensional data fields such as inspection task number, execution time, execution path, facility status data, alarm information, and task response results.

[0035] Preferably, in order to improve the accuracy of scenario-based analysis, the clustering unit preferentially extracts record entries with complete structures and good data timeliness, and performs unsupervised clustering or semi-supervised clustering operations based on the task feature vector.

[0036] Among them, in a feasible implementation of the scenario-based clustering, a feature space is constructed based on the similarity of inspection tasks in dimensions such as facility paths, task objectives, execution frequency, and associated alarm types. It is optional but not limited to using algorithms such as K-means, hierarchical clustering, or DBSCAN to divide and form M record groups. Each record group represents an evolution trajectory of scenario tasks with a common inspection logic, and M is a positive integer.

[0037] Subsequently, the mining unit is used to further mine the scene inspection facility-inspection guidance structure within the M record groups. This involves analyzing the execution path of the inspection tasks within each group and the relationship between the facility responses, and constructing a path guidance map from the task trigger point to the target facility. Specifically, this process uses parameters such as the frequency of inspection tasks within the group, the intensity of facility responses, and path stability as a basis. Using a preset ratio within the group data, for example, a high-frequency path ratio of at least 80%, this process identifies highly correlated facility sequences and extracts them as inspection guidance relationships.

[0038] For example, taking the busbar abnormal hot spot detection scenario as an example, if multiple tasks are concentrated in the busbar area and highly repetitively involve temperature sensing cameras, thermal imaging devices, GIS interval switches and busbar controllers, then the facility chain constitutes the scenario-guided path of the group.

[0039] Furthermore, the determination unit receives the scenario's inspection facilities and inspection directions as input and constructs an inspection topology with node-edge weight attributes based on the physical location relationships between facilities and the logical order of task execution. Optionally, each node in the topology represents an identifiable inspection infrastructure, and edges represent execution flows or data dependencies within a task. Edge weights can be generated based on metrics such as path frequency and importance scores. This topology serves as the minimum executable inspection structural unit within the scenario and is used for subsequent task driving and resource scheduling.

[0040] Furthermore, the association unit defines a corresponding task scenario label for each of the M record groups and maps and binds it to the determined inspection topology. The task scenario label is generated by comprehensively analyzing the mined alarm type, task goal, and historical response effect.

[0041] For example: high-voltage switchgear re-inspection scenario, cable temperature rise monitoring scenario, etc. In an optional embodiment provided by this application, the association between the task scenario and the inspection topology is stored in a structured index manner, supporting retrieval based on task features and quickly generating matching execution structures, which serves as one of the key bases for subsequent inspection control unit execution scheduling, path generation, and module adaptation.

[0042] In summary, by constructing a highly matching and highly adaptable task topology structure, data-driven intelligent support capabilities are provided for the intelligent inspection system.

[0043] Furthermore, the development unit includes:

[0044] The first deployment unit is used to deploy the first decision node by interpreting the inspection task, wherein the interpretation dimensions include at least the molecular scale, the component scale and the system scale; the second deployment unit is used to deploy the second decision node by matching the task scenario and switching the inspection topology; the third deployment unit is used to deploy the third decision node by making control decisions based on the topological inspection parameters; the module construction unit is used to construct an inspection decision chain based on the first decision node, the second decision node and the third decision node, and perform data-driven training until convergence, as the intelligent inspection module.

[0045] In an embodiment of the present application, the first deployment unit is configured to perform a multi-scale interpretation operation on an input inspection task and deploy a first decision node based on this operation. The inspection task interpretation proposed in this application refers to converting a task objective or structured task instruction described in natural language into a set of decision parameters that can be recognized by the system, thereby achieving a refined decomposition of the task logic.

[0046] In an optional embodiment provided herein, the interpretation process is based on a multi-dimensional framework, encompassing at least three levels: molecular, component, and system. The molecular scale represents the smallest device unit involved in the task, such as the inspection requirements for bolts, contacts, and interface points. The component scale analyzes the operational status of specific components, such as switch mechanisms, cooling fans, and busbar connections. The system scale identifies the coordinated behavior of the entire device system, such as the busbar system coordination status and power supply link connectivity.

[0047] Among them, the first deployment unit maps the above-mentioned multi-scale task elements into a structured vector form, and then generates a first decision node deployed in the control model, and the node function is task interpretation and scale mapping.

[0048] Furthermore, the second deployment unit is used to deploy a second decision node according to a mapping relationship between a task scenario matched by the task and a corresponding inspection topology after the task interpretation is completed.

[0049] Among them, task scenario matching refers to identifying the target scenario that is most similar to the current task characteristics from the task scenario library, that is, the collection of clustered recorded task scenarios mined in the previous steps, and extracting the topological structure bound to the scenario. The inspection topology switching refers to dynamically enabling the inspection path and execution node adapted to it according to the switching logic of different scenarios. The second deployment unit deploys the second decision node based on this logic. The node function is to complete the decision binding from the task scenario to the inspection path and support real-time adjustment of the topological structure during the task flow.

[0050] Furthermore, the third deployment unit is configured to deploy a third decision node based on key control parameters within the inspection topology. These control parameters include node evaluation weights within the inspection path, execution order priorities, device status criteria thresholds, and control action scheduling rules. This deployment process can be implemented by combining the operational capabilities of inspection facilities and communication feedback latency in real-world scenarios to generate a set of policies for real-time control. Based on these policies, the third decision node executes control actions such as path optimization, device scheduling, and status feedback triggering, thereby providing comprehensive decision support for policy execution at each node within the topology.

[0051] In summary, the deployment of the node function logic of each decision node has been completed. In order to further ensure its automated driving capability and decision convergence, logical association and training are performed on it.

[0052] Specifically, the module construction unit logically connects the first decision node, the second decision node and the third decision node in series to construct a complete inspection decision chain. The decision chain is structurally embodied as a multi-layer structure, in which different nodes correspond to different task understandings, path matching and control logic. The module construction unit further adopts a data-driven training mechanism to perform end-to-end optimization training on the decision chain. The training data comes from historical inspection task execution data, equipment response logs, inspection feedback results, etc., and is iteratively updated based on supervised learning or reinforcement learning methods. After the final training convergence, the resulting decision chain constitutes an intelligent inspection module with adaptive and generalizable capabilities, which is used to support efficient response and precise control of multiple tasks.

[0053] Furthermore, the intelligent inspection module establishes a communication loop with the edge computing gateway and the inspection infrastructure; wherein, inspection scheduling management is performed by deploying a synaptic-like IoT protocol, wherein the synaptic-like IoT protocol is an event-driven IoT interaction protocol, including a forward protocol and a reverse protocol.

[0054] In an embodiment of the present application, a stable, low-latency communication loop is established between the intelligent inspection module and the edge computing gateway and the inspection infrastructure. Optionally, the communication loop supports two-way state synchronization and command distribution functions.

[0055] Specifically, the intelligent inspection module, as the system's core decision-maker, acquires uploaded task information in real time and forwards generated control instructions to specific execution terminals via the edge computing gateway, achieving closed-loop control of the system. The communication loop supports multi-protocol access and task priority scheduling, automatically selecting the optimal transmission path based on the task's time sensitivity, data volume, and security level.

[0056] Preferably, in the above communication structure, to improve the adaptability and response efficiency of inspection tasks, a synaptic-like IoT protocol is introduced. This protocol simulates the signal transmission mechanism of biological neural synapses and adopts an event-driven interaction method for task scheduling and data response. Among other features, this type of protocol has dynamic routing, state awakening, and collaborative feedback. It can automatically start the relevant task chain when the state of the inspection facility suddenly changes or when an external event triggers it, avoiding the resource waste and response delay caused by traditional periodic polling.

[0057] Furthermore, the synaptic IoT protocol is specifically divided into a forward protocol and a reverse protocol. The forward protocol is suitable for active inspection tasks, where the intelligent inspection module initiates the task, issues instructions to the inspection infrastructure, and performs status collection and target confirmation according to a pre-set path. The forward protocol emphasizes a forward transmission mechanism of task-driven and path-controlled transmission, and is commonly used in scenarios such as routine inspections and periodic maintenance.

[0058] Similarly, the reverse protocol is suitable for passive response tasks. This means that when a piece of inspection infrastructure detects an anomaly—for example, abnormal temperature, arcing signals, or loose cables—it reports this to the edge computing gateway, triggering the intelligent inspection module to generate a corresponding emergency task and quickly deploy an execution path. The reverse protocol emphasizes event perception and a response mechanism based on retroactive deployment, making it suitable for emergency scenarios such as fault tracing and anomaly verification.

[0059] In summary, a closed-loop linkage from status triggering to task response is achieved, which effectively improves the dynamic adaptability and intelligence level of substation inspection, and provides a technical foundation for realizing high-reliability inspection in the power Internet of Things environment.

[0060] Furthermore, the inspection tasks include type 1 inspection tasks and type 2 inspection tasks; wherein, the type 1 inspection tasks are active inspection tasks of the intelligent inspection module-inspection infrastructure, and the type 2 inspection tasks are passive inspection tasks of the inspection infrastructure-edge computing gateway-inspection infrastructure; wherein, the passive inspection tasks are guided by self-inspection risk triggering events.

[0061] Among them, the first type of inspection task adopts a forward protocol based on the synaptic-like Internet of Things protocol, and the second type of inspection task adopts a reverse protocol based on the synaptic-like Internet of Things protocol.

[0062] In the embodiment of the present application, the inspection tasks can be divided into type 1 inspection tasks and type 2 inspection tasks according to their triggering methods and task paths. The type 1 inspection tasks are inspection tasks initiated by the intelligent inspection module, and their execution path is usually intelligent inspection module → inspection infrastructure.

[0063] Specifically, based on pre-set inspection strategies or periodic scheduling rules, combined with the task scenario and inspection topology, task instructions are automatically generated and distributed to the corresponding inspection infrastructure, such as fixed sensors, rail inspection robots, or infrared imaging devices. This task type is primarily used for routine inspections, periodic inspections, or strategy-based coverage tasks, and features clear task planning, controllable paths, and pre-emptive execution.

[0064] In contrast, the second type of inspection task is a passive inspection task initiated by the inspection infrastructure itself when it senses an abnormal state. Its complete path is inspection infrastructure → edge computing gateway → inspection infrastructure.

[0065] Specifically, some sensing nodes deployed in the system, such as temperature sensors, partial discharge detectors, and image recognition terminals, have real-time self-checking capabilities. Upon detecting that a device's status exceeds a preset safety threshold, such as excessive cable temperature rise, loose connections, or increased partial discharge signals, they immediately upload abnormal data to the edge computing gateway. Upon receiving this data, the edge computing gateway uses its embedded event recognition model to perform anomaly inference and task generation, triggering the scenario topology and inspection path corresponding to the event. It then issues a new round of targeted verification tasks to the executing devices, completing risk identification and emergency response operations.

[0066] In a feasible embodiment, the event recognition model is a brief deduction functional component of sample training. For example, the state vector is used as the input sample and the substation risk event is used as the output sample. The retrieved historical data records are integrated and supervised training is performed until convergence, that is, the preset accuracy is met, and the constructed event recognition model is determined to be completed.

[0067] Therefore, the second-class inspection tasks are guided by self-inspection risk triggering events, and have the characteristics of rapid response, flexible paths, and dynamic task generation. They are widely used in scenarios such as temporary inspections, fault tracing, and abnormal re-inspection.

[0068] In the implementation of this application, the aforementioned inspection tasks utilize a forward protocol based on the Synapse IoT protocol. This forward protocol establishes a forward command chain from the intelligent inspection module to the inspection execution end, emphasizing task scheduling priorities, inspection path sequencing, and data flow control between facilities. This protocol incorporates mechanisms such as task broadcasting, path wakeup, and sequence verification, making it suitable for clearly structured proactive inspection processes.

[0069] Accordingly, the second type of inspection tasks utilize a reverse protocol based on the Synaptic IoT protocol. This protocol simulates the reverse feedback mechanism of neurons and supports event-driven data inference, path selection, and task reallocation. This reverse protocol enables rapid resource reverse scheduling, emergency path construction, and closed-loop response to abnormal events, significantly improving responsiveness to sudden failures or potential risks.

[0070] In summary, the inspection task realizes an efficient, dynamic, and adaptive task linkage mechanism under the intelligent inspection module and edge computing architecture by distinguishing between active and passive path types and matching forward and reverse communication protocols.

[0071] A type I inspection control unit 12 is used to upload an inspection task to the intelligent inspection module, determine it as a type I inspection task, interpret the task and match it with the scenario-based topology, switch to the target inspection topology, assist the synaptic IoT protocol, and execute topology-driven facility inspection control under the forward protocol.

[0072] In an embodiment of the present application, the first-class inspection control unit 12 is used to identify whether the task meets the judgment conditions of the first-class inspection task after receiving the inspection task upload instruction, and execute the corresponding inspection logic if it is confirmed to be a first-class inspection task.

[0073] Among them, the aforementioned type of inspection tasks are tasks actively generated by the system based on a preset cycle, policy drive or scheduling plan, and the task instructions usually come from the internal planning area of ​​the central control platform or the intelligent inspection module.

[0074] Specifically, when an inspection task is uploaded to the intelligent inspection module, the Class I inspection control unit 12 first invokes task identification logic to analyze the task type, trigger source, task tag, and execution priority. This logic determines whether the task was actively generated by the intelligent inspection module, or whether it was subjectively uploaded by a terminal device and does not contain an external abnormal event trigger field. If the above determination logic is met, the task is confirmed as a Class I inspection task.

[0075] After confirming the task type, the first-class inspection control unit 12 invokes the first decision-making process to perform a structured interpretation of the task content. Specifically, the task interpretation process is based on a multi-dimensional rule set and includes: inspection target identification (such as target device type and area number), inspection parameter extraction (such as sampling frequency, accuracy requirements, and task duration), and abnormality tolerance setting. The interpretation results form a task attribute vector, which is passed as an input parameter to the second decision node.

[0076] Furthermore, after receiving the interpreted task attribute vector, the second decision node performs a matching search operation in a preset task scenario-inspection topology mapping library to find a target inspection topology that most closely matches the task's characteristics. The target inspection topology is a directed graph structure that reflects the task execution path, consisting of inspection facilities represented by nodes and execution logic represented by edges. Once the match is complete, the system switches to the target inspection topology, allowing the inspection task to be executed according to a specific path order, prioritizing high-weighted facilities.

[0077] Then, the third decision node is triggered to convert the determined target inspection topology into readability parameter control information based on specific inspection facilities.

[0078] Subsequently, a type-I inspection control unit 12 invokes the forward protocol within the synaptic IoT protocol to establish a one-way control link from the intelligent inspection module to the inspection infrastructure. The forward protocol supports three communication modes: command issuance, status request, and feedback reception. It features lightweight, low latency, and high command consistency.

[0079] Specifically, a type of inspection control unit dispatches corresponding facilities in topological order between inspection path nodes to perform data collection, action execution or visual recognition operations, and makes intermediate adjustments based on feedback information during the execution process to ensure the path is closed.

[0080] In summary, a type of inspection control unit 12 realizes the accurate identification and efficient execution of active inspection tasks through the processing flow of task identification → interpretation → topology matching → forward protocol control, and builds an intelligent inspection control link with tasks as the core, topology as the structure, and protocol as the communication.

[0081] The second-class inspection control unit 13 is used to transmit risk data to the edge computing gateway if there is self-inspection of the inspection infrastructure, reversely infer the risk event and generate a second-class inspection task, trigger the lightweight inspection module embedded in the gateway, determine the event inspection topology and execute the topology-driven facility inspection control under the reverse protocol, wherein the inspection topology is optimized and adjusted through self-healing networking.

[0082] In an embodiment of the present application, the Class II inspection control unit 13 is used to identify and process the self-inspection risk data generated by the inspection infrastructure during the operation of the substation, and generate Class II inspection tasks that are highly correlated with the events based on this, so as to achieve rapid response and closed-loop handling of sudden abnormalities.

[0083] For example, when a patrol infrastructure, such as a temperature sensor, partial discharge probe, visual monitoring node, etc., or a self-inspection unit embedded in the substation infrastructure detects abnormal data during normal operation, such as a sudden temperature rise, current fluctuation, or suspicious deformation in the image, the abnormal information is marked as risk data and uploaded to the edge computing gateway in real time through the communication loop.

[0084] Subsequently, after the edge computing gateway receives the risk data, the second-class inspection control unit 13 first calls the embedded event recognition model to perform structured analysis on the data, identify the abnormal trigger source, impact range and related equipment characteristics, and reversely generate possible risk events based on historical scenario data, equipment health model or alarm mode.

[0085] For example, if the temperature in a certain area continues to rise and is accompanied by frequent fluctuations in the signal of a certain type of connector, the abnormal contact resistance event of the medium-voltage cable can be inferred based on this, and this event can be used as input to generate a targeted risk inspection task.

[0086] Subsequently, the Class II inspection control unit 13 invokes the lightweight inspection module within the gateway, triggering the task dispatch mechanism. This lightweight inspection module is an edge version of the intelligent inspection module, compressed and trained using the attention transfer mechanism. It features low resource usage, rapid deployment, and rapid response capabilities. Upon detecting a risk event, this module rapidly determines the corresponding event inspection topology based on its facility relationships and spatial distribution. This topology, centered around the event device, expands outward to potentially affected related facilities, forming a highly centralized regional task chain.

[0087] After the task topology is constructed, the Class II inspection control unit 13 schedules tasks according to the reverse protocol in the Synapse-like IoT protocol. Unlike the forward method, which issues commands from the central server, the reverse protocol is event-driven, distributing commands from the edge to the inspection side. It emphasizes response speed, path flexibility, and fault tolerance. Following the sequence specified in the event topology, the relevant inspection facilities are scheduled to perform operations such as image capture, partial discharge detection, thermal imaging, or action review. The results are then transmitted back in real time to support risk identification and expansion assessment.

[0088] In a preferred embodiment of this application, during Class I and Class II inspections, if node communication failures, equipment failures, or insufficient energy consumption occur along the inspection path, a self-healing networking mechanism is automatically invoked to optimize and adjust the current inspection topology in real time. This self-healing mechanism leverages redundant connections and state sharing between facility nodes to automatically re-establish the task path or dispatch backup equipment to ensure a complete, closed-loop inspection.

[0089] For example, if the original track robot A path fails, the nearby robot B with redundant access capabilities can be dispatched to continue to perform the remaining tasks to ensure that the tasks are not interrupted.

[0090] In summary, the second-class inspection control unit 13 realizes the full-process control chain from risk data triggering, event identification, lightweight scheduling, to topology adaptive control and task closed-loop execution. It has event-driven, edge autonomy and path recovery capabilities, providing the substation intelligent inspection system with efficient, flexible and real-time exception handling capabilities.

[0091] Furthermore, the system further comprises:

[0092] The mechanism introduction unit is used to introduce a self-healing networking mechanism for the inspection infrastructure; the mechanism triggering unit is used to trigger the self-healing networking mechanism based on the inspection facility status based on the inspection topology and perform inspection optimization and adjustment.

[0093] In an embodiment of the present application, the mechanism introduction unit is used to embed the self-healing networking mechanism into the network structure of the inspection infrastructure during the inspection deployment phase. The self-healing networking mechanism refers to establishing a dynamic communication and task scheduling system with redundant connections, state sharing, and path self-adjustment capabilities between inspection device nodes. In an optional embodiment, this mechanism is borrowed from a distributed fault-tolerant system, allowing for the automatic selection of alternative paths or nodes in the event of local communication interruptions, device failures, or abnormal task execution to ensure the continuity and stability of inspection tasks.

[0094] In an optional embodiment, the mechanism introduction unit implements the deployment of the mechanism specifically in the following manner: first, in the initial configuration of the system, an adjacency table is established for each inspection infrastructure node, and its redundant communication links and alternative functional nodes are marked; second, status broadcast and heartbeat detection methods are introduced at the node level to ensure that the operating status, task execution progress and availability of each inspection infrastructure can be perceived in real time; finally, a path variability rule base is established between the edge computing gateway and the intelligent inspection module to support dynamic topology reconstruction operations.

[0095] Furthermore, the mechanism triggering unit is used to dynamically evaluate task execution consistency during inspection task execution based on the operating status of each facility in the inspection topology, and trigger the self-healing networking mechanism when an execution anomaly or structural degradation is detected. The triggering judgment is based on a combination of multiple conditions, such as node communication timeout, status update frequency below a threshold, execution feedback anomaly, or path prediction mismatch.

[0096] For example, when a rail robot in the inspection path fails to upload task feedback data within the scheduled time, or the uploaded data content indicates that the current path task cannot be continued, the mechanism trigger unit will determine that there is a risk of rupture in the inspection topology and immediately activate the self-healing mechanism.

[0097] Subsequently, after the self-healing mechanism is triggered, the current topology will be re-evaluated with reference to the preset redundant connection relationship and task priority rules, and the inspection path will be optimized and adjusted. This adjustment includes but is not limited to the following methods: switching to a backup node to perform the task (such as using a second drone to inspect the same area), rebuilding the task path (such as avoiding failed nodes and detouring to the target device), and reallocating tasks (such as transferring some tasks to other idle inspection units). After the optimization and adjustment is completed, the updated topology will be written into the current task process as the new execution path, ensuring the smooth closed-loop execution of the task in a dynamic network structure with a certain degree of fault tolerance.

[0098] In summary, during the inspection process, the system has the capabilities of online fault tolerance, autonomous recovery, and path self-reconstruction, which effectively improves the execution robustness and system stability of inspection tasks in the complex power Internet of Things environment.

[0099] Furthermore, the second type inspection control unit includes:

[0100] A lightweight training unit is used to perform lightweight training on the intelligent inspection module based on local event-oriented attention migration to determine the lightweight inspection module; a module deployment unit is used to deploy the lightweight inspection module to the edge computing gateway.

[0101] In an embodiment of the present application, the lightweight training unit is used to perform structural compression and task transfer on the intelligent inspection module to meet the constraints on resource occupancy, execution delay and response speed in the edge computing environment.

[0102] Specifically, this unit uses a local event-driven attention transfer mechanism as its core training strategy. Local event-driven attention refers to high-frequency risk events collected during task execution or recurring task scenarios within specific inspection areas. For example, frequent temperature rise alarms in a GIS bay or frequent failures of the main transformer cooling fan are examples. After identifying these local events, they are used as key areas of the model to enhance the edge module's discernment capabilities in critical scenarios.

[0103] Among them, the attention transfer mechanism is to transfer the attention weight map in the original intelligent inspection module to the target lightweight model during the model lightweighting process, retaining its feature recognition ability for highly correlated areas and eliminating redundant structures and low-contribution parameters.

[0104] In a feasible training process, the lightweight training unit constructs a dual network structure: the main model retains the complete parameter structure, and the lightweight model is streamlined through parameter pruning, channel sparsification, knowledge distillation, etc. The two perform attention map matching and output consistency constraint training under shared input, so that the lightweight model can significantly reduce the model size and computational complexity while maintaining the core performance of the main model.

[0105] After training, the identified lightweight inspection module is solidified as a deployable component, featuring fast loading, low-power execution, optimized event response, and basic intelligent inspection capabilities such as task scheduling, path generation, and feedback processing for specific scenarios. This lightweight inspection module is suitable for resource-constrained deployment environments such as edge computing gateways, and can independently complete risk event response and local task handling without frequent reliance on master control.

[0106] Furthermore, the module deployment unit is responsible for deploying the generated lightweight inspection module to the edge computing gateway node. The deployment process includes three stages: resource initialization, module registration, and communication interface binding.

[0107] Specifically, first, the module deployment unit initializes the operating environment according to the hardware specifications of the edge gateway, including configuring the memory usage limit, computing resource scheduling weight, network access rights, etc.; secondly, the lightweight module is registered to the task scheduling platform of the edge system to ensure that it has the authority and execution capabilities to receive abnormal data, trigger tasks, and feedback results; finally, by binding the communication interface of the synaptic IoT protocol, it is ensured that the lightweight module can interact with the inspection infrastructure in a two-way manner to achieve autonomous response under event-driven conditions.

[0108] In summary, the function migration and capability transfer from central intelligence to edge response nodes have been achieved, and an edge intelligent inspection system with regional autonomy, rapid response, and resource sensitivity has been built, effectively improving the overall real-time and reliability of the system.

[0109] Furthermore, after executing the facility inspection control, the system further includes:

[0110] The data return unit is used to obtain task inspection data and return it to the edge computing gateway; the conflict processing unit is used to perform data conflict analysis and conflict resolution decisions on the task inspection data based on inspection homology to determine valid inspection data, wherein the conflict resolution decision method is data analysis resolution or directional re-inspection resolution; the risk management unit is used to return the valid inspection data for substation risk positioning and alarm management.

[0111] In an embodiment of the present application, the data return unit is used to collect and upload the task inspection data generated by the inspection infrastructure after the inspection task is completed, so as to ensure that the actual execution status and result data of the inspection process can be mastered in real time. The task inspection data includes but is not limited to equipment status values ​​(such as voltage, current, temperature), image information, behavior logs, abnormal event records, and task completion marks. The data return unit packages the above data in a structured format and transmits it to the gateway side through a high-speed communication channel established with the edge computing gateway to ensure that the data is not lost due to communication interruption or delay. In a preferred embodiment, breakpoint resumption and data verification mechanisms are supported to improve data integrity and security.

[0112] Subsequently, the conflict processing unit is used to perform consistency checks and conflict resolution on the returned task inspection data. Its core logic is based on the inspection homology principle, that is, to analyze whether there are multiple inspection facilities that generate conflicting data in the same task cycle, the same inspection object, or the same task node. Specifically, the types of conflicts include but are not limited to: numerical differences in the same device status (such as the uploaded temperature of device A is 65°C and that of device B is 80°C), inconsistent image recognition results, conflicts in task completion status flags, etc. To this end, the conflict processing unit constructs a data index table within the task cycle, and uses the timestamp, device number, and task identifier as the joint primary key to compare and analyze the data source.

[0113] Furthermore, in terms of conflict handling strategies, two conflict resolution decision-making methods are supported: the first is data analysis resolution, which uses statistical methods such as multi-source fusion, weighted averaging, and trend extrapolation to analyze and process conflicting data to generate unified valid data items; the second is targeted re-inspection resolution. When the conflict level or conflict range exceeds a preset threshold, a rescheduling mechanism is automatically triggered, instructing designated inspection facilities or backup channels to quickly re-inspect the conflicting nodes, prioritizing the latest collected results as the final valid data. This processing flow ensures data consistency and accuracy under the multi-path, multi-terminal redundant inspection architecture.

[0114] Furthermore, the risk management unit is used to analyze and process the valid inspection data filtered by the conflict processing unit, and complete the risk positioning and alarm management operations for the substation equipment or area. Specifically, by calling a multimodal risk assessment model based on fault feature pattern recognition, state trend prediction and scenario deduction, the valid data is correlated and interpreted to determine whether it constitutes an early warning signal of equipment abnormality, operation deviation or potential failure. If a risk event is identified, the alarm response process is triggered by the preset alarm rules, including alarm level classification (such as yellow warning, red fault), notification of responsible persons, generation of maintenance suggestions, etc., and the results are fed back to the intelligent inspection module and the substation monitoring center to form a closed-loop control system.

[0115] Optionally, the multimodal risk assessment model is a functional unit that takes detection data as input and outputs real-time faults and predicted faults, and can be constructed using a neural network architecture and sample training.

[0116] In summary, a complete post-processing process chain has been established from inspection data collection, conflict resolution to risk warning response, realizing high-reliability inspection result management and risk linkage disposal driven by data, and significantly improving the operational stability of the intelligent inspection system and the predictability of substation operation and maintenance management.

[0117] The intelligent inspection auxiliary system for substations in the power Internet of Things environment provided by this application has the following technical effects:

[0118] 1. Task Scenario-Inspection Topology Definition: This system distinguishes between inspection and substation infrastructure, clusters and mines scenario characteristics based on historical inspection records, and associates task scenarios with inspection topologies. This system implements scenario-based pre-planning of inspection routes, improving inspection targeting, reducing redundant operations, and increasing inspection efficiency. It deploys a decision chain encompassing task interpretation, topology switching, and parameter control, and implements multi-scale task analysis and response through data-driven training. This system supports intelligent analysis and execution of complex inspection tasks, ensuring the accuracy and automation of the inspection process.

[0119] 2. Bidirectional Protocol-Driven Inspection: Utilizing a synaptic-like IoT protocol, Class I tasks utilize a forward protocol for active inspection, while Class II tasks utilize a reverse protocol for passive inspection, triggered by risk data. This dual inspection mechanism combines proactive prevention with reactive response to ensure timely risk detection and resolution, enhancing system reliability. Self-Healing Network Optimization: A self-healing network mechanism is triggered based on the status of inspection facilities, dynamically adjusting the inspection topology. This allows for real-time adaptation to equipment anomalies or environmental changes, ensuring the stability of the inspection link and reducing the risk of missed inspections.

[0120] 3. Edge computing and data processing: The edge computing gateway deploys a lightweight inspection module to analyze and resolve conflicts in returned data, locate risks, and generate alarms. This enables localized and rapid data processing, reduces transmission delays, accurately identifies risk sources, and enhances substation safety management capabilities.

[0121] Through the above detailed description of the substation intelligent inspection auxiliary system in the power Internet of Things environment in this specification, those skilled in the art can clearly understand the substation intelligent inspection auxiliary system in the power Internet of Things environment in this embodiment. For the device disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the system part description.

[0122] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The intelligent inspection auxiliary system for substations in the power Internet of Things environment is characterized by: The system comprises: The development unit is used to obtain the infrastructure of the target substation, define the task scenario-inspection topology, and develop the intelligent inspection module within the substation intelligent inspection auxiliary system; A first-class inspection control unit is used to, if an inspection task is uploaded to the intelligent inspection module and is determined to be a first-class inspection task, switch to the target inspection topology through task interpretation and scenario-based topology matching, assist the synaptic IoT protocol, and execute topology-driven facility inspection control under the forward protocol; The second-class inspection control unit is used to transmit risk data from the inspection infrastructure self-inspection to the edge computing gateway, reversely infer the risk event and generate a second-class inspection task, trigger the lightweight inspection module embedded in the gateway, determine the event inspection topology and execute the topology-driven facility inspection control under the reverse protocol; Among them, inspection topology optimization and adjustment are carried out through self-healing networking.

2. The substation intelligent inspection auxiliary system in the power Internet of Things environment according to claim 1 is characterized in that: The development unit includes: a facility classification unit, configured to determine, for the infrastructure, an inspection infrastructure and a substation infrastructure, wherein the inspection infrastructure is identified by an inspection domain based on the substation infrastructure, and the inspection infrastructure includes substation access facilities and dynamic inspection facilities; The scenario reconstruction unit is used to perform topology reconstruction based on the substation scenario for the inspection infrastructure and define the task scenario-inspection topology.

3. The substation intelligent inspection auxiliary system in the power Internet of Things environment according to claim 2 is characterized in that: The scene reconstruction unit includes: The clustering unit is used to retrieve the inspection records of the target substation, perform scenario-based clustering, and determine M record groups; A mining unit is configured to mine, for the M record groups, scene inspection facilities and inspection directions within the groups based on a preset ratio within the groups as a constraint; A determination unit, used to determine the inspection topology based on the scenario inspection facilities-inspection orientation; The association unit is used to define a task scenario for the M record groups and associate the task scenario with the inspection topology.

4. The substation intelligent inspection auxiliary system in the power Internet of Things environment according to claim 3 is characterized in that: The development unit includes: A first deployment unit is configured to deploy a first decision node based on inspection task interpretation, wherein the interpretation dimensions include at least a molecular scale, a component scale, and a system scale; The second deployment unit is used to deploy the second decision node based on task scenario matching and inspection topology switching; A third deployment unit is configured to deploy a third decision node based on a control decision of a topology inspection parameter; A module construction unit is used to construct an inspection decision chain based on the first decision node, the second decision node and the third decision node, and perform data-driven training until convergence, as the intelligent inspection module.

5. The substation intelligent inspection auxiliary system in the power Internet of Things environment according to claim 1 is characterized in that: The intelligent inspection module establishes a communication loop with the edge computing gateway and the inspection infrastructure; Among them, inspection scheduling management is carried out by deploying a synaptic-like Internet of Things protocol, wherein the synaptic-like Internet of Things protocol is an event-driven Internet of Things interaction protocol, which includes a forward protocol and a reverse protocol.

6. The substation intelligent inspection auxiliary system in the electric power Internet of Things environment according to claim 5 is characterized in that: Inspection tasks include first-class inspection tasks and second-class inspection tasks; Among them, the first type of inspection task is the active inspection task of the intelligent inspection module-inspection infrastructure, and the second type of inspection task is the passive inspection task of the inspection infrastructure-edge computing gateway-inspection infrastructure; Among them, there are passive inspection tasks guided by self-inspection risk trigger events.

7. The substation intelligent inspection auxiliary system in the electric power Internet of Things environment according to claim 6 is characterized in that: The first type of inspection task adopts a forward protocol based on a synaptic-like Internet of Things protocol, and the second type of inspection task adopts a reverse protocol based on a synaptic-like Internet of Things protocol.

8. The substation intelligent inspection auxiliary system in the electric power Internet of Things environment according to claim 1 is characterized in that: The system further comprises: Mechanism introduction unit, used to introduce a self-healing networking mechanism for inspection infrastructure; The mechanism triggering unit is used to trigger the self-healing networking mechanism based on the inspection facility status based on the inspection topology and perform inspection optimization adjustment.

9. The substation intelligent inspection auxiliary system in the electric power Internet of Things environment according to claim 1 is characterized in that: The second type inspection control unit includes: A lightweight training unit, configured to perform lightweight training on the intelligent inspection module based on attention transfer guided by local events, and determine a lightweight inspection module; A module deployment unit is used to deploy the lightweight inspection module to the edge computing gateway.

10. The substation intelligent inspection auxiliary system in the electric power Internet of Things environment according to claim 1, characterized in that: After executing the facility inspection control, the system further includes: The data return unit is used to obtain task inspection data and return it to the edge computing gateway; A conflict processing unit is used to perform data conflict analysis and conflict resolution decision on the task inspection data based on inspection homology to determine valid inspection data, wherein the conflict resolution decision method is data analysis resolution or directional re-inspection resolution; The risk management unit is used to transmit the valid inspection data back to perform substation risk positioning and alarm management.

Images