A method and device for maintenance and repair operation management of a power monitoring system

Abstract

This invention discloses a maintenance and operation management method and apparatus for a power monitoring system, relating to the field of operation management technology. The method includes: interacting with a power monitoring system's data acquisition cluster, aggregating and processing the data, and introducing maintenance scenarios to construct an operation and maintenance scenario data lake; constructing a task time-series network; establishing a response gradient relationship between the operation and maintenance scenario data lake and the task time-series network, and connecting the operation and maintenance scenario data lake and the task time-series network for execution; calculating the change characteristics of real-time monitoring data in the operation and maintenance scenario data lake based on the response gradient relationship, activating operation and maintenance scenario tasks in the task time-series network, and searching for task execution strategies according to the priority of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information. This invention solves the technical problem in the prior art that power operation and maintenance tasks cannot achieve real-time linkage response based on changes in monitoring data, achieving the technical effect of automatic triggering and efficient scheduling execution of maintenance and repair tasks.

Description

Technical Field

[0001] This invention relates to the field of operation management technology, specifically to a method and apparatus for maintenance, repair, and operation management of a power monitoring system. Background Technology

[0002] Power monitoring systems collect a large amount of monitoring data, including equipment operating status, environmental parameters, and event logs, during daily operation. However, this data often lacks effective linkage with subsequent maintenance and repair tasks, resulting in data being used only for record-keeping and failing to be promptly transformed into a basis for operation and maintenance decisions. Maintenance and repair tasks typically rely on manual experience or preset cycle triggers, making it difficult to respond promptly to changes in the characteristics of sudden faults or potential risks. Task scheduling lacks dynamic coupling with real-time monitoring data, causing problems such as response delays, unreasonable resource allocation, and unclear task execution priorities, seriously affecting system operation and maintenance efficiency and operational safety. Summary of the Invention

[0003] This application provides a method and apparatus for the maintenance, repair, operation and management of a power monitoring system, which addresses the technical problem in the prior art that power operation and maintenance tasks cannot achieve real-time linkage response based on changes in monitoring data.

[0004] In view of the above problems, this application provides a method and apparatus for maintenance, repair and operation management of a power monitoring system.

[0005] The first aspect of this application provides a method for maintenance, repair, operation, and management of a power monitoring system, the method comprising:

[0006] The interactive power monitoring system collects data clusters, aggregates and processes these data clusters, introduces maintenance and repair scenarios, and constructs an operation and maintenance scenario data lake. It then analyzes the task distribution, task attributes, and task timing requirements of power operation and maintenance tasks, constructing a task timing network. A response gradient relationship is established between the operation and maintenance scenario data lake and the task timing network, and this relationship is used to establish an execution connection between the two networks. Based on the changing characteristics of real-time monitoring data in the operation and maintenance scenario data lake, calculations are performed using the response gradient relationship to activate operation and maintenance scenario tasks in the task timing network. Finally, task execution strategies are searched according to the priority of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information.

[0007] A second aspect of this application provides a maintenance, repair, and operation management device for a power monitoring system, the device comprising:

[0008] An aggregation processing module is used to aggregate the acquisition data cluster of the interactive power monitoring system, perform aggregation processing on the acquisition data cluster, introduce maintenance and repair scenarios, and construct an operation and maintenance scenario data lake; an analysis module is used to analyze the task distribution, task attributes, and task time series requirements of the power operation and maintenance tasks, and construct a task time series network; an execution connection module is used to establish a response gradient relationship between the operation and maintenance scenario data lake and the task time series network, and perform execution connection on the operation and maintenance scenario data lake and the task time series network by using the response gradient relationship; a policy search module is used to perform operations with the response gradient relationship according to the change characteristics of the real-time monitoring data in the operation and maintenance scenario data lake, activate the operation and maintenance scenario tasks in the task time series network, and search for task operation strategies according to the priorities of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application aggregates the acquisition data cluster of the interactive power monitoring system, performs aggregation processing on the acquisition data cluster, introduces maintenance and repair scenarios, and constructs an operation and maintenance scenario data lake; analyzes the task distribution, task attributes, and task time series requirements of the power operation and maintenance tasks, and constructs a task time series network; establishes a response gradient relationship between the operation and maintenance scenario data lake and the task time series network, and performs execution connection on the operation and maintenance scenario data lake and the task time series network by using the response gradient relationship; performs operations with the response gradient relationship according to the change characteristics of the real-time monitoring data in the operation and maintenance scenario data lake, activates the operation and maintenance scenario tasks in the task time series network, and searches for task operation strategies according to the priorities of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information. This invention solves the technical problem that power operation and maintenance tasks in the prior art cannot achieve real-time linkage response according to the changes in monitoring data. By constructing an operation and maintenance scenario data lake and establishing a response gradient connection with the task time series network, the technical effect of realizing the automatic triggering and efficient scheduling execution of maintenance and repair tasks is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings required for the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0012] Figure 1 It is a schematic flowchart of a maintenance and repair operation management method for a power monitoring system provided in an embodiment of this application;

[0013] Figure 2This is a schematic diagram of the structure of a maintenance, repair, operation and management device for a power monitoring system provided in an embodiment of this application.

[0014] Figure labeling: Aggregation processing module 11, parsing module 12, execution connection module 13, strategy search module 14. Detailed Implementation

[0015] This application provides a maintenance and operation management method and device for a power monitoring system. It addresses the technical problem in the prior art that power operation and maintenance tasks cannot achieve real-time linkage response based on changes in monitoring data. By constructing an operation and maintenance scenario data lake and establishing a response gradient connection with the task time sequence network, it achieves the technical effect of automatic triggering and efficient scheduling execution of maintenance and repair tasks.

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, apparatus, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or apparatus.

[0018] Example 1, as Figure 1 As shown, this application provides a maintenance, repair, operation, and management method for a power monitoring system, the method comprising:

[0019] Step S100: The data collection cluster of the interactive power monitoring system is aggregated and processed to introduce maintenance and repair scenarios and construct an operation and maintenance scenario data lake.

[0020] In this embodiment of the application, the data collection cluster related to equipment operation is first obtained through the interactive power monitoring system. The interactive power monitoring system refers to an intelligent monitoring platform with multi-point sensing, two-way communication and data interaction capabilities. The data collection cluster refers to a structured or semi-structured data set composed of data such as voltage, current, power, temperature and humidity, alarm events, and operation logs collected by various sensors, monitoring terminals and edge devices.

[0021] The collected data cluster is then aggregated, undergoing standardized preprocessing through unified format encoding, temporal and spatial alignment, and structural reconstruction. Next, a maintenance and repair scenario is introduced, utilizing semantic channels to parse and aggregate the semantic relationships between the collected data and pre-defined maintenance scenario templates. Finally, data with related semantics are uniformly stored in the maintenance scenario data lake.

[0022] Furthermore, the method provided in the application embodiment, which aggregates the collected data cluster, introduces maintenance and repair scenarios, and constructs an operation and maintenance scenario data lake, also includes:

[0023] The collected data cluster is preprocessed in a standardized manner; a semantic channel is established to perform semantic parsing on the collected data cluster; based on the data correlation between the collected data and the maintenance scenario, the collected data is semantically aggregated to construct the maintenance scenario data lake.

[0024] In this embodiment of the application, when performing standardized preprocessing on the collected data cluster, the collected data cluster is encoded in a unified format, and a standardized intermediate data object is constructed based on time alignment and spatial alignment mechanisms. The original data is then restructured to form a standardized data structure containing basic fields such as timestamp, device ID, parameters, and status.

[0025] Next, semantic parsing is performed on the collected data cluster by establishing a semantic channel. This process begins with building a scenario template library, which contains multiple pre-defined operation and maintenance scenario templates. Each template includes elements such as activation conditions, associated devices, geographical location, and task type to define a specific maintenance and repair scenario. Semantic association analysis is then performed using this scenario template library and standardized data to identify data feature combinations associated with specific scenario templates, constructing a semantic training set. Subsequently, a semantic communication training mechanism is used to model the training set, and a stable semantic channel is established through multiple rounds of training convergence.

[0026] Based on the aforementioned semantic channels, semantic aggregation of the collected data is performed according to the data correlation between the collected data and the maintenance and repair scenarios. Specifically, through the scenario association parameters output by the semantic channels, data features from multiple sources that point to the same operational meaning are aggregated into the corresponding operational scenario, realizing the organizational migration from raw data to scenario-aware data. For example, in the scenario of abnormal temperature rise in a power distribution cabinet, temperature data from different temperature measurement points, corresponding device IDs, and their spatial location tags are aggregated through the semantic channels into a set of input data used to trigger the scenario. Finally, all semantically aggregated data is stored according to the operational scenario dimension, constructing an operational scenario data lake with semantic expression capabilities and operational triggering basis.

[0027] Furthermore, in the method provided in the application embodiments, the standardization preprocessing of the collected data cluster further includes:

[0028] The collected data cluster is encoded in a unified format and aligned according to time and space. Based on the standardized intermediate data object, the original collected data is restructured according to the unified format encoding and time and space alignment relationship to form a basic field structure including timestamp, device ID, parameters, and status.

[0029] In this embodiment, a unified format encoding method is first used to standardize the format of the raw data in the collected data cluster. Based on preset field mapping rules, heterogeneous data formats uploaded from different devices or systems are uniformly converted into standard field structures, achieving consistency in data field names, unit expressions, and numerical types.

[0030] Subsequently, based on time alignment, the standard format data is synchronized using a unified time window. Data from different sampling frequencies are aligned to the standard time axis according to a unified time step, and data not at boundary time points are filled using sliding window compensation or interpolation to ensure that all data are comparable in the time dimension.

[0031] Next, spatial alignment is performed. Based on the device codes in the collected data, the device geographic information mapping table is called to bind the device number with the actual installation location. Data from different sources is labeled with a unified spatial identifier, such as the site, interval area, or electrical level. Through this step, the data is normalized in the spatial dimension.

[0032] Finally, the structure is reconstructed according to the standardized intermediate data object. The data that has been format-encoded, time-aligned, and spatially aligned is filled into the structure template of the standardized intermediate data object, and extracted and reconstructed into a data structure including basic fields such as timestamp, device number, parameter type, and status value. Through this structure reconstruction step, a standard data format that can be used for subsequent semantic analysis and scene recognition is finally formed.

[0033] Furthermore, in the method provided in the application embodiments, establishing a semantic channel to perform semantic parsing on the collected data cluster further includes:

[0034] A scenario template library is constructed, including multiple preset operation and maintenance scenario templates, which have activation conditions, associated devices, geographical locations, and task types. Based on the scenario template library, semantic association analysis is performed between the collected data cluster and the multiple preset operation and maintenance scenario templates to construct a semantic training set. Semantic communication training convergence is performed based on the semantic training set to obtain the semantic channel, which is used to process the collected data for scenario semantic association and output scenario association parameters.

[0035] In this embodiment, a scenario template library is first constructed. During this process, multiple types of operation and maintenance scenario templates are preset, combining historical operation and maintenance data, equipment operation models, and fault rules. Each template includes activation conditions, associated equipment, geographical location, and task type. For example, a template might be defined as "temperature exceeding 85 degrees Celsius for 5 minutes" as the activation condition, "110 kV circuit breaker" as the associated equipment, "substation bay 1" as the geographical location, and "early warning driven task" as the task type.

[0036] After the template construction is completed, the standardized collected data cluster is semantically correlated with various operation and maintenance scenario templates. This analysis uses field matching and condition judgment methods to identify whether there are records in the data that meet the activation conditions of a certain template, and to determine whether the corresponding device and spatial location are consistent with the template requirements. If they are, the data item is marked as a training sample for the corresponding scenario, thus constructing a semantic training set.

[0037] Subsequently, semantic communication training is conducted based on the semantic training set. Through multiple rounds of training and optimization, the structural semantic relationship between the input data and template features is modeled, allowing the semantic channel to gradually converge and obtain a stable structural mapping relationship. Once this semantic channel is formed, it has the ability to input collected data and automatically identify its corresponding operation and maintenance scenario.

[0038] Finally, the semantic channel is used to perform scene semantic correlation processing on the collected data, and scene correlation parameters are output. Scene correlation parameters refer to a set of structured information generated based on the semantic matching results between the input data and each scene template based on the semantic channel. These parameters include the identifier of the target scene template, the task type to which the data belongs, the parameter fields that trigger the activation conditions and their corresponding values, and the device number and location code of the data source.

[0039] Step S200: Analyze the task distribution, task attributes, and task timing requirements of power operation and maintenance tasks, and construct a task timing network.

[0040] In this embodiment, when analyzing the task distribution, task attributes, and task timing requirements of power operation and maintenance tasks, the tasks are first categorized based on their physical spatial regions, completing the first-level task category division and clarifying the corresponding substation, line, or equipment unit for each task category. Then, the operation and maintenance activation conditions are analyzed based on task attributes, completing the second-level task category classification. Task attributes include periodic tasks, early warning-driven tasks, fault handling tasks, and environmental perception tasks. Next, the time-based startup dependencies between tasks are analyzed, identifying features such as prerequisite tasks, concurrent tasks, or sequential tasks, completing the construction of the third-level task categories. Finally, by integrating the information from the first-level, second-level, and third-level task categories, task graph nodes are constructed, and directed connections are established based on the timing dependency logic between tasks, completing the task timing network.

[0041] Furthermore, in the method provided in the application embodiments, the step of parsing the task distribution, task attributes, and task timing requirements of power operation and maintenance tasks to construct a task timing network further includes:

[0042] Power operation and maintenance tasks are classified into primary categories based on their spatial distribution to obtain primary task categories; secondary categories are classified based on task attributes to obtain secondary task categories based on operation and maintenance activation conditions; tertiary categories are classified based on the time-series requirements of power operation and maintenance tasks to obtain tertiary task categories; and task graph nodes are constructed according to the primary, secondary, and tertiary task categories, and the temporal dependencies between task nodes are determined to obtain the task temporal network.

[0043] Furthermore, the method provided in the application embodiments also includes:

[0044] The task attributes include: periodic tasks, early warning-driven tasks, fault handling tasks, and environmental perception tasks.

[0045] In this embodiment, a spatial partitioning mapping method is first used to analyze task distribution. In this process, by retrieving data from the power equipment deployment database and the GIS geographic information system, the equipment number and spatial coordinates corresponding to each power operation and maintenance task are extracted. Spatial partitioning is then performed by combining the substation area code, feeder number, and main wiring unit identifier, resulting in a primary classification of the task distribution space and obtaining primary task categories. This process divides tasks into spatially defined categories such as substation main equipment tasks, feeder inspection tasks, and switch operation tasks.

[0046] Subsequently, a task attribute tag extraction method was used to analyze the operation and maintenance activation conditions. This process identifies typical triggering mechanisms for each task by analyzing scheduling instruction records, equipment alarm linkage strategies, and operation and maintenance logs. Combined with preset task attribute classification rules, secondary classification is performed to obtain secondary task categories. Task attributes include periodic tasks, early warning-driven tasks, fault handling tasks, and environmental awareness tasks. For example, monthly transformer inspections belong to periodic tasks, and fan startup triggered by abnormal temperature belongs to early warning-driven tasks.

[0047] Next, a task dependency modeling approach is used to analyze the temporal requirements of tasks. By extracting operation and maintenance procedures, scheduling operation processes, and historical task execution sequences, the startup sequence and synchronization dependency logic between tasks are identified, and operation and maintenance tasks are classified into three levels to obtain three-level task categories. Specifically, a task dependency graph is constructed to mark the pre- and post-task relationships between tasks. For example, the grounding switch closing task starts after the circuit breaker opening task, forming a serial dependency relationship; the environmental monitoring task and the auxiliary control equipment inspection task are executed in parallel within the same time window, forming a parallel dependency structure.

[0048] After completing the three-level task classification described above, a graph structure construction method is used to generate task graph nodes. Each task graph node is bound to its corresponding first-level, second-level, and third-level task categories, and directed edges are established based on task temporal dependencies to determine the temporal connections between task nodes. Through node construction and connection operations, a task temporal network is formed. The task temporal network possesses spatial location identifiers, activation attribute identifiers, and temporal dependency structures, fully expressing the distribution relationships, response logic, and execution order of power operation and maintenance tasks.

[0049] Step S300: Establish the response gradient relationship between the operation and maintenance scenario data lake and the task time series network, and use the response gradient relationship to perform an execution connection between the operation and maintenance scenario data lake and the task time series network.

[0050] In this embodiment of the application, in order to establish the response gradient relationship between the operation and maintenance scenario data lake and the task time series network, the key monitoring parameters and their change characteristics in the operation and maintenance scenario data lake are first extracted, and the response gradient relationship of the task time series network is constructed in combination with the task trigger response requirements of each task node to represent the activation urgency and priority of each task node.

[0051] Subsequently, the operation and maintenance scenario data lake and the task time series network are connected by the response gradient relationship. That is, the activation priority of task nodes is set based on the response gradient relationship, the response relationship between the collected data and the task nodes is established, and the relationship is injected into the task scheduling engine to realize the execution connection between the operation and maintenance scenario data lake and the task time series network.

[0052] Furthermore, in the method provided in the application embodiments, establishing the response gradient relationship between the operation and maintenance scenario data lake and the task time-series network further includes:

[0053] Extract key monitoring parameters and their changing characteristics from the data lake of the operation and maintenance scenario; based on the key monitoring parameters and their changing characteristics, and the task triggering response requirements of each task node, determine the response gradient relationship of the task time series network, which is used to represent the urgency and priority of task activation.

[0054] In this embodiment, key monitoring parameters and their changing characteristics are first extracted from the operation and maintenance scenario data lake. The operation and maintenance scenario data lake contains structured monitoring data related to equipment operating status and environmental conditions under multiple operation and maintenance scenarios. For different operation and maintenance scenarios, indicative monitoring parameters are selected as key monitoring parameters, such as temperature, current, insulation resistance, humidity, and oil level. For each key monitoring parameter, its historical data sequence is retrieved over a continuous time period. Through processing such as time series differencing, extreme point identification, and three-point smoothing fitting, the changing characteristics of the parameter, such as the rate of change, abrupt change amplitude, and direction of change, are extracted to reflect the dynamic behavior of the parameter. For example, when dealing with high-voltage equipment temperature rise scenarios, the temperature rise rate is calculated based on a 10-minute sliding window, and abnormal trends are identified by the interval changes of continuous sudden increases.

[0055] After extracting key monitoring parameters and their changing characteristics, the extraction results are matched one by one with the task trigger response requirements of each task graph node in the task time-series network. Each task graph node contains structured definitions such as task type, activation conditions, response parameter fields, and associated equipment. Its task trigger response requirements are usually set in the form of threshold judgments, such as "current fluctuation amplitude greater than 10 amperes", "temperature rise rate greater than 5 degrees Celsius / minute", or "insulation resistance decreases more than twice consecutively". Through field parsing and condition comparison operations, the parameters corresponding to the changing characteristics are compared with the trigger conditions of the task graph node to identify the task graph node that currently meets the trigger requirements and mark it as being in an active state.

[0056] After identifying task graph nodes that meet the task trigger response requirements, a response gradient relationship of the task time-series network is constructed based on the extracted key monitoring parameters and their changing characteristics. This gradient represents the task activation urgency and priority of each task graph node in its current operating state. The urgency of task activation is indicated by the deviation between the current changing characteristics and the trigger threshold. The deviation is calculated by measuring the difference between the actual value of the key parameter and the threshold set in the task graph node. Simultaneously, the direction and rate of change are considered to determine whether the parameter is rapidly approaching or continuously exceeding the threshold, thus determining the activation urgency of the task graph node. For example, if the threshold is set to "temperature rise rate ≥ 4℃ / minute," and the current rate is 7℃ / minute with a continuously increasing trend, then the node has a high activation urgency level. The priority of task activation is determined by the task attributes of the task graph node and the operating level of the associated equipment. Based on the category labels in the task attributes, such as periodic tasks, early warning-driven tasks, fault handling tasks, and environmental perception tasks, and combined with the equipment's voltage level, system location, and safety responsibility level, priority mapping rules are constructed. For example, fault handling tasks on main substation equipment have a higher priority than periodic inspection tasks at the end of feeders. Ultimately, a response gradient structure is established based on the two dimensions of urgency and priority, and this structure is attached to the task graph nodes as a response gradient relationship, forming the response gradient relationship of the task temporal network.

[0057] Furthermore, in the method provided in the application embodiment, the execution connection between the operation and maintenance scenario data lake and the task time-series network is established using the response gradient relationship, and further includes:

[0058] Based on the response gradient relationship, the activation priority of task nodes is set; the response relationship between the collected data and the activation priority of task nodes is established, injected into the task scheduling engine, and the execution connection relationship between the operation and maintenance scenario data lake and the task time-series network is determined.

[0059] In this embodiment, the activation priority of each task graph node is set according to the response gradient relationship. The response gradient relationship defines two dimensions for the task activation urgency and the task activation priority corresponding to the task graph node. By dividing the task activation urgency into intervals (e.g., a temperature rise rate higher than a set threshold of 50% is marked as "high urgency", higher than 20% as "medium urgency", and lower than 10% as "low urgency"), and combining the task attributes corresponding to the task graph node (including periodic tasks, early warning-driven tasks, fault handling tasks, and environmental perception tasks) and the voltage level of the equipment associated with the task (e.g., 220 kV equipment, 110 kV equipment, 10 kV equipment), a priority value is set using a priority numbering method and attached to the task graph node as the task node activation priority. For example, if a "fault handling task" is identified as being associated with a 220 kV main transformer, and the monitored parameter "oil temperature" continuously exceeds a set threshold with a rate of change in the high urgency interval, then the priority value of this task graph node is set to 1, representing the highest priority.

[0060] After setting the activation priority of task nodes, a response relationship between the collected data and the activation priority of task nodes is established by real-time monitoring of key monitoring parameters and their changing characteristics in the data lake of the operation and maintenance scenario. This response relationship is implemented based on conditional judgment logic. The parameter fields in the data lake are traversed and compared with the task trigger response requirements set in the task graph nodes. For example, for a task graph node with a trigger condition set as "temperature rise rate greater than 5℃ / minute, lasting for 3 cycles," if the current detection result in the data lake is "6.2℃ / minute, lasting for 3 cycles," then the task graph node is determined to have met the activation condition, its priority value is read as 1, and this judgment result is written into the response mapping table as a structured mapping relationship "temperature parameter—task graph node A—priority value 1." By establishing a complete mapping relationship table for all nodes that meet the conditions, a set of response relationships between the collected data and the activation priority of task nodes is formed.

[0061] Subsequently, this set of response relationships is injected into the task scheduling engine. The task scheduling engine employs a priority queue scheduling strategy, enqueuing task graph nodes that have met the activation conditions according to their priority values ​​from smallest to largest, and performing dynamic scheduling. The scheduling engine also invokes the node topology structure in the task sequence network, using depth-first traversal or topology sorting algorithms to determine the dependencies between task graph nodes, ensuring that subsequent nodes are only allowed to be scheduled if the preceding node's status is "completed." For example, if the "circuit breaker status confirmation task" is a preceding task to the "grounding switch operation task," the execution status of the former is checked first; if it is not completed, the latter is temporarily stored, even if its priority value is higher.

[0062] Finally, through the execution of task nodes, the setting of activation priorities, the establishment of the response relationship between collected data and task node priorities, and the priority scheduling control of the task scheduling engine, the execution connection relationship between the operation and maintenance scenario data lake and the task time series network is constructed and completed.

[0063] Step S400: According to the change characteristics of the real-time monitoring data in the operation and maintenance scenario data lake, perform an operation with the response gradient relationship, activate the operation and maintenance scenario tasks in the task time series network, and search for task operation strategies according to the priorities of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information.

[0064] In the embodiment of the present application, first, the change characteristics of the real-time monitoring data are extracted from the operation and maintenance scenario data lake, including information such as the change rate, fluctuation amplitude, and trend direction of key monitoring parameters. The change characteristics are logically matched with the constructed response gradient relationship. By comparing the current change characteristics item by item with the task trigger response requirements recorded in the task graph nodes, the task graph nodes that meet the conditions are identified, and the corresponding operation and maintenance scenario tasks are activated. For example, when the rising rate of the real-time temperature monitoring value exceeds the temperature rise trigger threshold set by the task graph node, immediately mark this node as the activated state. After the activation of the operation and maintenance scenario tasks is completed, all the activated tasks are sorted according to the task node activation priorities attached to each task graph node, and a scheduling candidate set is established.

[0065] On this basis, combined with the maintenance and repair tasks currently being executed, extract the task execution status, repair task attributes, task impact, and task time series dependencies of the task graph nodes to establish the current execution task chain. At the same time, obtain the newly activated operation and maintenance scenario tasks and parse their corresponding task requirement chains. Perform structural matching and coupling of the execution task chain and the task requirement chain, identify the schedulable paths in the task graph, and sort and filter the paths according to the task node activation priorities.

[0066] Subsequently, search for task operation strategies among multiple feasible paths, and preferentially select the paths with high completion degree, low resource occupancy, and few task dependency conflicts in the current execution state to form the optimal operation strategy. Finally, output structured maintenance and repair management execution information, including scheduling task numbers, task execution sequences, task graph structure paths, key monitoring parameter support information, and scheduling control flags, to guide the subsequent scheduling system or operation and maintenance personnel to perform tasks, ensuring that the system realizes orderly and priority-clear linkage operation control according to the task graph nodes driven by real-time data.

[0067] Furthermore, in the method provided by the embodiment of the application, obtaining the maintenance and repair management execution information further includes:

[0068] The current maintenance and repair tasks are analyzed in terms of task execution status, maintenance task attributes, task impact, and task timing to establish an execution task chain; newly added activated maintenance and repair scenario tasks are obtained, and the task requirement chain of the activated scenario tasks is parsed; the execution task chain and the task requirement chain are coupled together, and a task link strategy is searched with the goal of maximizing task completion evaluation and minimizing task scheduling loss to obtain the maintenance and repair management execution information.

[0069] In this embodiment, firstly, for the currently executing maintenance and repair tasks, the task execution status, maintenance task attributes, task impact, and task sequence information of each task are obtained, and an execution task chain is constructed based on this. The task execution status includes the task start time, elapsed time, and current completion rate; the maintenance task attributes include required resources (personnel, equipment), task type (planned maintenance or fault handling), and executing unit; the task impact is graded from 1 to 5 levels according to the criticality of the system corresponding to the task, where the grading is determined by matching according to a preset task criticality matching table; the task sequence information includes the order of dependencies between the task and other tasks (e.g., replacing a component must be performed after disassembly and inspection). By uniformly modeling this information, an execution task chain representing the order of task execution, resource occupancy relationships, and task importance is established.

[0070] Then, for newly activated maintenance tasks, the triggering reasons (such as sensor alarms, periodic plans) are identified, and their time requirements, resource needs, and task objectives are extracted to establish a task requirement chain. The task requirement chain also includes information such as the task's planned start and end times, required equipment and personnel, task type, task impact level, and its preceding dependent tasks. For example, a maintenance task activated by a "pump station vibration anomaly alarm" requires equipment inspection and preliminary handling to be completed within 4 hours. The resources involved include a vibration detector and two maintenance personnel, with an impact level of 4. Its task requirement chain will record the above information for this task and mark its potential dependencies on other possible subsequent tasks (such as bearing replacement).

[0071] Next, we will perform a coupled analysis of the task chain and the task requirement chain. Specifically, this involves comparing the overlap in time and resources between the tasks: firstly, checking if there are tasks requiring the same type of personnel or equipment within the same timeframe; secondly, checking whether a new task depends on the completion of the current task before it can be executed. For example, if the current task A requires the use of a high-voltage tester for 2 hours, and a new task B also requires the same equipment for the next hour, then there is a resource conflict between the two. Similarly, if a new task C can only enter the testing phase after the completion of the current task D, then there is a time dependency between the two.

[0072] After detecting conflicts or dependencies, perform task link strategy search with the goal of maximizing task completion evaluation and minimizing task scheduling losses. The task completion evaluation can be quantitatively calculated by the formula: task completion evaluation value = current completion rate × task impact level. The task scheduling loss value = delay duration × task impact level × penalty coefficient, where the delay duration comes from the time difference between the plan and the actual execution; the penalty coefficient is the converted value of the business loss per unit delay (for example, the delay cost per hour is 10 points). If task F needs to be postponed for 2 hours, the impact level is 4, and the penalty coefficient is 10, then the loss value is 2×4×10 = 80 points.

[0073] For task pairs with conflicts, generate different scheduling schemes, such as maintaining the current task priority, switching the scheduling resources to support new tasks, or rearranging the start and end order of all tasks. For each scheme, calculate the total completion evaluation value and total scheduling loss value of all tasks respectively, and then select the strategy that maximizes the total task completion evaluation and minimizes the total scheduling loss as the final task link strategy. Finally, based on this final task link strategy, output the adjusted task start and end times, resource allocation plan, and task priority arrangement as maintenance and repair management execution information to support the subsequent scheduling execution system or scheduler to make the final instruction.

[0074] In the embodiments of the present application, in summary, the embodiments of the present application at least have the following technical effects:

[0075] The data collection cluster of the interactive power monitoring system of the present application aggregates the data collection cluster, introduces the maintenance and repair scenario, and constructs the operation and maintenance scenario data lake; analyzes the task distribution, task attributes, and task timing requirements of the power operation and maintenance tasks, and constructs the task timing network; establishes the response gradient relationship between the operation and maintenance scenario data lake and the task timing network, and uses the response gradient relationship to perform an execution connection between the operation and maintenance scenario data lake and the task timing network; according to the change characteristics of the real-time monitoring data in the operation and maintenance scenario data lake, perform an operation with the response gradient relationship, activate the operation and maintenance scenario tasks in the task timing network, and search for the task operation strategy according to the priority of the operation and maintenance scenario tasks to obtain the maintenance and repair management execution information. The present invention solves the technical problem that the power operation and maintenance tasks in the prior art cannot achieve real-time linkage response according to the change of monitoring data. By constructing the operation and maintenance scenario data lake and establishing the response gradient connection with the task timing network, the technical effect of realizing the automatic triggering and efficient scheduling execution of the maintenance and repair tasks is achieved.

[0076] Embodiment 2, based on the same inventive concept as the maintenance and repair operation management method of a power monitoring system in the foregoing embodiment, as Figure 2As shown, this application provides a maintenance, repair, and operation management device for a power monitoring system. The device and method embodiments in this application are based on the same inventive concept. The device includes:

[0077] The aggregation processing module 11 is used to aggregate the collected data cluster of the interactive power monitoring system, introduce maintenance and repair scenarios, and construct an operation and maintenance scenario data lake; the parsing module 12 is used to parse the task distribution, task attributes, and task timing requirements of power operation and maintenance tasks, and construct a task timing network; the execution connection module 13 is used to establish the response gradient relationship between the operation and maintenance scenario data lake and the task timing network, and use the response gradient relationship to perform execution connection between the operation and maintenance scenario data lake and the task timing network; the strategy search module 14 is used to perform calculations based on the change characteristics of real-time monitoring data in the operation and maintenance scenario data lake and the response gradient relationship, activate the operation and maintenance scenario tasks in the task timing network, and search for task operation strategies according to the priority of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information.

[0078] Furthermore, the device is also used to perform the following functions:

[0079] The collected data cluster is preprocessed in a standardized manner; a semantic channel is established to perform semantic parsing on the collected data cluster; based on the data correlation between the collected data and the maintenance scenario, the collected data is semantically aggregated to construct the maintenance scenario data lake.

[0080] Furthermore, the device is also used to perform the following functions:

[0081] The collected data cluster is encoded in a unified format and aligned according to time and space. Based on the standardized intermediate data object, the original collected data is restructured according to the unified format encoding and time and space alignment relationship to form a basic field structure including timestamp, device ID, parameters, and status.

[0082] Furthermore, the device is also used to perform the following functions:

[0083] A scenario template library is constructed, including multiple preset operation and maintenance scenario templates, which have activation conditions, associated devices, geographical locations, and task types. Based on the scenario template library, semantic association analysis is performed between the collected data cluster and the multiple preset operation and maintenance scenario templates to construct a semantic training set. Semantic communication training convergence is performed based on the semantic training set to obtain the semantic channel, which is used to process the collected data for scenario semantic association and output scenario association parameters.

[0084] Furthermore, the device is also used to perform the following functions:

[0085] Power operation and maintenance tasks are classified into primary categories based on their spatial distribution to obtain primary task categories; secondary categories are classified based on task attributes to obtain secondary task categories based on operation and maintenance activation conditions; tertiary categories are classified based on the time-series requirements of power operation and maintenance tasks to obtain tertiary task categories; and task graph nodes are constructed according to the primary, secondary, and tertiary task categories, and the temporal dependencies between task nodes are determined to obtain the task temporal network.

[0086] Furthermore, the device is also used to perform the following functions:

[0087] The task attributes include: periodic tasks, early warning-driven tasks, fault handling tasks, and environmental perception tasks.

[0088] Furthermore, the device is also used to perform the following functions:

[0089] Extract key monitoring parameters and their changing characteristics from the data lake of the operation and maintenance scenario; based on the key monitoring parameters and their changing characteristics, and the task triggering response requirements of each task node, determine the response gradient relationship of the task time series network, which is used to represent the urgency and priority of task activation.

[0090] Furthermore, the device is also used to perform the following functions:

[0091] Based on the response gradient relationship, the activation priority of task nodes is set; the response relationship between the collected data and the activation priority of task nodes is established, injected into the task scheduling engine, and the execution connection relationship between the operation and maintenance scenario data lake and the task time-series network is determined.

[0092] Furthermore, the device is also used to perform the following functions:

[0093] The current maintenance and repair tasks are analyzed in terms of task execution status, maintenance task attributes, task impact, and task timing to establish an execution task chain; newly added activated maintenance and repair scenario tasks are obtained, and the task requirement chain of the activated scenario tasks is parsed; the execution task chain and the task requirement chain are coupled together, and a task link strategy is searched with the goal of maximizing task completion evaluation and minimizing task scheduling loss to obtain the maintenance and repair management execution information.

[0094] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0095] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0096] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method of maintenance and repair operation management of a power monitoring system, characterized by, include: The data collection cluster of the interactive power monitoring system is aggregated and processed, and a maintenance and repair scenario is introduced to construct an operation and maintenance scenario data lake. The task distribution, task attributes, and task timing requirements of power operation and maintenance tasks are analyzed, and a task timing network is constructed. Establish the response gradient relationship between the operation and maintenance scenario data lake and the task time series network, and use the response gradient relationship to execute the connection between the operation and maintenance scenario data lake and the task time series network; Based on the changing characteristics of real-time monitoring data in the operation and maintenance scenario data lake, calculations are performed on the relationship with the response gradient to activate the operation and maintenance scenario tasks in the task time sequence network, and task operation strategies are searched according to the priority of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information. Establishing the response gradient relationship between the operation and maintenance scenario data lake and the task time-series network includes: Extract the key monitoring parameters and their changing characteristics from the data lake in the operation and maintenance scenario; Based on the key monitoring parameters and their changing characteristics, and the task triggering response requirements of each task node, the response gradient relationship of the task time-series network is determined to represent the urgency and priority of task activation. Utilizing the response gradient relationship to perform an execution connection between the operation and maintenance scenario data lake and the task time series network includes: Based on the response gradient relationship, set the activation priority of the task nodes; Establish a response relationship between collected data and task node activation priority, inject it into the task scheduling engine, and determine the execution connection relationship between the operation and maintenance scenario data lake and the task time-series network.

2. The maintenance, repair, operation, and management method for a power monitoring system according to claim 1, characterized in that, The collected data cluster is aggregated and processed, and a maintenance and repair scenario is introduced to construct an operation and maintenance scenario data lake, including: The collected data cluster is subjected to standardized preprocessing; A semantic channel is established to perform semantic parsing on the collected data cluster. Based on the data correlation between the collected data and the maintenance and repair scenario, the semantic aggregation of the collected data is performed to construct the operation and maintenance scenario data lake.

3. The maintenance, repair, operation, and management method for a power monitoring system according to claim 2, characterized in that, The collected data cluster undergoes standardized preprocessing, including: The collected data clusters are encoded in a unified format and aligned according to time and space. Based on the standardized intermediate data object, the original collected data is restructured according to the unified format encoding and time and space alignment relationship to form a basic field structure including timestamp, device ID, parameters, and status.

4. The maintenance, repair, operation, and management method for a power monitoring system according to claim 2, characterized in that, Establishing a semantic channel to perform semantic parsing on the collected data cluster includes: Build a scenario template library, including multiple preset operation and maintenance scenario templates, with activation conditions, associated devices, geographical location, and task type; Based on the scenario template library, semantic association analysis is performed between the collected data cluster and preset multiple types of operation and maintenance scenario templates to construct a semantic training set; The semantic communication training converges based on the semantic training set to obtain the semantic channel, which is used to process the collected data for scene semantic correlation and output scene correlation parameters.

5. The maintenance, repair, operation, and management method for a power monitoring system according to claim 1, characterized in that, The process of analyzing task distribution, task attributes, and task timing requirements for power operation and maintenance tasks, and constructing a task timing network, includes: Power operation and maintenance tasks are classified into primary categories based on their spatial distribution. Based on task attributes, secondary classification of operation and maintenance activation conditions is performed to obtain secondary task categories; Based on the time-based startup dependencies of power operation and maintenance tasks, the task timing requirements are classified into three levels to obtain three-level task categories. Based on the first-level task category, second-level task category, and third-level task category, a task graph node is constructed and the temporal dependencies between task nodes are determined to obtain the task temporal network.

6. The maintenance, repair, operation, and management method for a power monitoring system according to claim 5, characterized in that, The task attributes include: periodic tasks, early warning-driven tasks, fault handling tasks, and environmental perception tasks.

7. The maintenance, repair, operation, and management method for a power monitoring system according to claim 1, characterized in that, Obtaining maintenance and repair management execution information also includes: Analyze the current maintenance and repair tasks by analyzing their execution status, maintenance task attributes, task impact, and task sequence, and establish an execution task chain. Retrieve newly added activation and maintenance scenario tasks, and parse the task requirement chain of the activation scenario tasks; By coupling the execution task chain with the task requirement chain, a task link strategy search is performed with the goal of maximizing task completion evaluation and minimizing task scheduling loss, thereby obtaining the maintenance and repair management execution information.

8. A maintenance, repair, and operation management device for a power monitoring system, characterized in that, The device is used to execute the maintenance, repair, operation, and management method for a power monitoring system as described in any one of claims 1-7, and the device comprises: The aggregation processing module is used to aggregate the data cluster of the interactive power monitoring system, introduce maintenance and repair scenarios, and build an operation and maintenance scenario data lake. The parsing module is used to parse the task distribution, task attributes, and task timing requirements of power operation and maintenance tasks, and to construct a task timing network. An execution connection module is used to establish the response gradient relationship between the operation and maintenance scenario data lake and the task time series network, and to use the response gradient relationship to perform an execution connection between the operation and maintenance scenario data lake and the task time series network. The strategy search module is used to perform calculations based on the change characteristics of real-time monitoring data in the operation and maintenance scenario data lake and the response gradient relationship, activate the operation and maintenance scenario tasks in the task time sequence network, and search for task operation strategies according to the priority of the operation and maintenance scenario tasks to obtain maintenance and repair management execution information. Extract key monitoring parameters and their changing characteristics from the data lake of the operation and maintenance scenario; based on the key monitoring parameters and their changing characteristics, and the task triggering response requirements of each task node, determine the response gradient relationship of the task time series network to represent the urgency and priority of task activation; Based on the response gradient relationship, the activation priority of task nodes is set; the response relationship between the collected data and the activation priority of task nodes is established, injected into the task scheduling engine, and the execution connection relationship between the operation and maintenance scenario data lake and the task time-series network is determined.

Images