AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method and system

Abstract

This application provides a method and system for station-city operation and maintenance and adaptive knowledge base management based on AR inspection. The method includes: acquiring real-time monitoring data of various devices in the station-city integration system; performing real-time preprocessing on the real-time monitoring data to obtain a real-time data stream; processing the real-time data stream using a pre-determined operation and maintenance performance quantification model to obtain the corresponding operation and maintenance target steps; performing intelligent resource scheduling based on the operation and maintenance target steps and a pre-determined adaptive knowledge base to obtain multiple maintenance plans; filtering the multiple maintenance plans to obtain the optimal maintenance plan; optimizing the operation and maintenance strategy based on the optimal maintenance plan to obtain the optimized operation and maintenance strategy; and displaying the optimized operation and maintenance strategy on an AR interactive interface, and performing station-city operation and maintenance according to the optimized strategy. This solution can improve the response speed and accuracy of station-city integration operation and maintenance, as well as optimize resource allocation and utilization efficiency.

Description

Technical Field

[0001] This application relates to the field of operation and maintenance management technology, specifically to a station and city operation and maintenance and adaptive knowledge base management method and system based on AR inspection. Background Technology

[0002] AR (Augmented Reality), as a new generation of human-computer interaction technology, achieves a deep integration of the digital and real worlds by overlaying virtual information onto the real environment in real time. In recent years, AR technology has been increasingly widely used in the industrial field, especially showing great potential in scenarios such as equipment inspection, maintenance guidance, and training.

[0003] The groundbreaking development of large language models has brought revolutionary changes to knowledge management and intelligent question answering. Large models, such as ChatGPT and Wenxin Yiyan, demonstrate powerful natural language understanding and generation capabilities, providing accurate and timely intelligent question answers based on massive knowledge bases. This provides the technological possibility for the transformation of traditional static knowledge bases into dynamic, interactive knowledge services.

[0004] Digitalization of operations and maintenance (O&M) management has become an important means for enterprises to improve operational efficiency and reduce costs. However, traditional O&M performance evaluation mainly relies on subjective judgment and outdated statistical data, lacking real-time accuracy and objectivity. Establishing a scientific, objective, and real-time O&M performance evaluation system, and achieving personalized capability improvement based on the evaluation results, has become a significant challenge facing current O&M management.

[0005] Station-city integration, as a new model of deep integration between modern urban rail transit and urban development, integrates various urban functions such as rail transit stations, commercial development, office space, and public services within a single spatial carrier, forming a complex of urban functions. This development model has been widely applied globally, becoming an important way to solve urban space constraints, improve land use efficiency, and promote sustainable development. Summary of the Invention

[0006] This application aims to provide a method and system for station and city operation and maintenance and adaptive knowledge base management based on AR inspection, which can improve the response speed and accuracy of station and city integrated operation and maintenance, as well as optimize resource allocation and utilization efficiency.

[0007] The technical solution of this application is implemented as follows:

[0008] In a first aspect, embodiments of this application provide a method for station and city operation and maintenance and adaptive knowledge base management based on AR inspection, the method comprising:

[0009] Acquire real-time monitoring data of various devices in the station-city integration system; and perform real-time preprocessing on the real-time monitoring data to obtain a real-time data stream;

[0010] The real-time data stream is processed using a pre-determined quantification model for operation and maintenance performance to obtain the corresponding operation and maintenance target steps. The operation and maintenance target steps include the operation and maintenance target equipment and the operation and maintenance data set related to the operation and maintenance target equipment.

[0011] Based on the aforementioned operational and maintenance objectives and a pre-determined adaptive knowledge base, intelligent resource scheduling is performed to obtain multiple maintenance plans; and the multiple maintenance plans are then filtered to obtain the optimal maintenance plan.

[0012] Based on the optimal maintenance scheme, the operation and maintenance strategy is optimized to obtain the optimized operation and maintenance strategy; the optimized operation and maintenance strategy is displayed on the AR interactive interface, and the station and city operation and maintenance are carried out according to the optimized operation and maintenance strategy.

[0013] In the above scheme, the real-time monitoring data includes environmental monitoring data, equipment operation data, pedestrian flow status data, and energy consumption data;

[0014] The real-time preprocessing of the real-time monitoring data to obtain a real-time data stream includes:

[0015] The environmental monitoring data, equipment operation data, pedestrian flow data, and energy consumption data are cleaned to remove invalid data, resulting in cleaned environmental monitoring data, cleaned equipment operation data, cleaned pedestrian flow data, and cleaned energy consumption data.

[0016] The environmental monitoring data, equipment operation data, pedestrian flow data, and energy consumption data after cleaning are converted into standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow data, and standardized energy consumption data, respectively.

[0017] Based on the standardized environmental monitoring data, the standardized equipment operation data, the standardized pedestrian flow data, and the standardized energy consumption data, anomaly detection is performed to obtain abnormal data; and the abnormal data is marked as abnormal.

[0018] The real-time data stream is determined based on the anomaly markers, the standardized environmental monitoring data, the standardized equipment operation data, the standardized pedestrian flow data, and the standardized energy consumption data.

[0019] In the above scheme, the step of processing the real-time data stream using a pre-determined operation and maintenance performance quantification model to obtain the corresponding operation and maintenance target steps of the real-time data stream includes:

[0020] The real-time data stream is evaluated using the aforementioned operation and maintenance performance quantification model to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream.

[0021] Based on the performance scores of each of the multiple operation and maintenance links, the operation and maintenance link with a performance score less than a preset performance threshold is selected as the operation and maintenance target link corresponding to the real-time data stream.

[0022] In the above scheme, the performance evaluation of the real-time data stream using the operation and maintenance performance quantification model to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream includes:

[0023] The real-time data stream is classified by topic to obtain data on multiple different message topics; wherein, the multiple different message topics correspond to the types of devices.

[0024] The multiple different message topics are partitioned to obtain subtopics for each of the multiple different message topics;

[0025] By using the operation and maintenance performance quantification model and the sub-topics of the multiple different message topics, the data of the multiple different message topics are evaluated simultaneously to obtain the performance scores of the multiple operation and maintenance links of the real-time data stream.

[0026] In the above scheme, the step of simultaneously evaluating the data of the multiple different message topics using the operation and maintenance performance quantification model and their respective subtopics to obtain the performance scores of multiple operation and maintenance links of the real-time data stream includes:

[0027] For each of the multiple different message topics, the data of each message topic is evaluated using the operation and maintenance performance quantification model, and the corresponding equipment management indicator value, energy consumption management indicator value, personnel flow management indicator value, and environmental management indicator value are obtained.

[0028] The equipment management index value, the energy consumption management index value, the pedestrian flow management index value, and the environmental management index value are standardized to obtain the equipment management standard value, the energy consumption management standard value, the pedestrian flow management standard value, and the environmental management standard value.

[0029] Determine the weight values ​​for the equipment management standard value, the energy consumption management standard value, the pedestrian flow management standard value, and the environmental management standard value;

[0030] Based on the weight values, the equipment management standard value, the energy consumption management standard value, the pedestrian flow management standard value, and the environmental management standard value are weighted to obtain a comprehensive index value;

[0031] Based on the comprehensive index value, time decay processing and anomaly processing are performed to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream.

[0032] In the above scheme, the intelligent resource scheduling based on the operational and maintenance target and a pre-determined adaptive knowledge base yields multiple maintenance schemes; and the optimal maintenance scheme is obtained by filtering the multiple maintenance schemes, including:

[0033] Based on the adaptive knowledge base, knowledge is recommended for the operation and maintenance target process to obtain recommended information;

[0034] Based on the recommended information and the pre-acquired skill matrix information, intelligent resource scheduling is performed to obtain the multiple maintenance plans; wherein, the skill matrix information includes the professional ability and on-duty time of each staff member;

[0035] For each of the multiple maintenance schemes, the matching similarity between personnel, path and the target maintenance link is calculated to obtain the similarity value of each maintenance scheme;

[0036] Based on the similarity value of each of the multiple maintenance schemes, the maintenance scheme with the highest similarity value is selected as the optimal maintenance scheme.

[0037] The method in the above scheme further includes:

[0038] Based on the optimal maintenance scheme, the adaptive knowledge base is updated to obtain the updated adaptive knowledge base;

[0039] Based on the real-time monitoring data and historical data, fault prediction is performed to obtain prediction results;

[0040] Operation and maintenance optimization is performed based on the prediction results and the updated adaptive knowledge base.

[0041] Secondly, embodiments of this application provide a station-city operation and maintenance and adaptive knowledge base management system based on AR inspection. The AR inspection-based station-city operation and maintenance and adaptive knowledge base management system includes: an acquisition module, a data processing module, a resource scheduling module, and a display module, wherein...

[0042] The acquisition module is used to acquire real-time monitoring data of various devices in the station-city integration system; and to perform real-time preprocessing on the real-time monitoring data to obtain a real-time data stream;

[0043] The data processing module is used to process the real-time data stream using a pre-determined operation and maintenance performance quantification model to obtain the operation and maintenance target link corresponding to the real-time data stream; wherein, the operation and maintenance target link includes the operation and maintenance target device and the operation and maintenance data set related to the operation and maintenance target device;

[0044] The resource scheduling module is used to perform intelligent resource scheduling based on the operation and maintenance target and a pre-determined adaptive knowledge base to obtain multiple maintenance plans; and to filter the multiple maintenance plans to obtain the optimal maintenance plan.

[0045] The display module is used to optimize the operation and maintenance strategy based on the optimal maintenance scheme to obtain the optimized operation and maintenance strategy; and to display the optimized operation and maintenance strategy on the AR interactive interface, and to perform station and city operation and maintenance according to the optimized operation and maintenance strategy.

[0046] Thirdly, embodiments of this application provide a station-city operation and maintenance and adaptive knowledge base management device based on AR inspection, comprising: a processor and a memory; wherein,

[0047] The memory is used to store computer programs;

[0048] The processor is configured to call and run the computer program from the memory to perform the method as described in the first aspect.

[0049] Fourthly, embodiments of this application provide a computer-readable storage medium storing executable instructions for causing a processor to perform the method described in the first aspect.

[0050] This application provides a method and system for station-city operation and maintenance and adaptive knowledge base management based on AR inspection. The method includes: acquiring real-time monitoring data of various devices in the station-city integration system; performing real-time preprocessing on the real-time monitoring data to obtain a real-time data stream; processing the real-time data stream using a pre-determined operation and maintenance performance quantification model to obtain the operation and maintenance target link corresponding to the real-time data stream; wherein, the operation and maintenance target link includes the operation and maintenance target device and the operation and maintenance data set related to the operation and maintenance target device; performing intelligent resource scheduling based on the operation and maintenance target link and the pre-determined adaptive knowledge base to obtain multiple maintenance schemes; filtering the multiple maintenance schemes to obtain the optimal maintenance scheme; optimizing the operation and maintenance strategy based on the optimal maintenance scheme to obtain the optimized operation and maintenance strategy; displaying the optimized operation and maintenance strategy on an AR interactive interface, and performing station-city operation and maintenance according to the optimized operation and maintenance strategy. The above solution establishes a scientific quantitative evaluation system for the operation and maintenance level of the station-city integration system. By combining AR visualization technology, intelligent knowledge base and adaptive optimization mechanism, it achieves accurate identification, scientific allocation and dynamic optimization of operation and maintenance resources, improves the response speed and accuracy of station-city integration operation and maintenance, and thus significantly improves operation and maintenance efficiency. At the same time, it also optimizes resource allocation and utilization efficiency, thereby reducing operation and maintenance costs and enhancing system reliability. Attached Figure Description

[0051] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the technical solutions of this application. Obviously, the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0052] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0053] Figure 1 An optional flowchart illustrating a station and city operation and maintenance and adaptive knowledge base management method based on AR inspection provided in an embodiment of this application;

[0054] Figure 2 A system framework diagram of a station and city operation and maintenance and adaptive knowledge base management method based on AR inspection provided in the embodiments of this application;

[0055] Figure 3A schematic diagram of the structure of a station and city operation and maintenance and adaptive knowledge base management system based on AR inspection provided in this application embodiment;

[0056] Figure 4 This is a schematic diagram of the structure of a station and city operation and maintenance and adaptive knowledge base management device based on AR inspection, provided for an embodiment of this application. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the specific technical solutions of this application will be further described in detail below with reference to the accompanying drawings of the embodiments of this application. The following embodiments are used to illustrate this application, but are not intended to limit the scope of this application.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0059] In the following description, references to "some embodiments," "this embodiment," "this application embodiment," and examples, etc., describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subset of all possible embodiments and may be combined with each other without conflict.

[0060] If the application documents contain similar descriptions such as "first / second", the following explanation shall be added: In the following description, the terms "first / second / third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first / second / third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0061] This application provides a method for station and city operation and maintenance and adaptive knowledge base management based on AR inspection. Figure 1 This is an optional flowchart illustrating an AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method provided in an embodiment of this application. It will combine... Figure 1 The steps shown are explained.

[0062] S101. Obtain real-time monitoring data of various devices in the station-city integration system; and perform real-time preprocessing on the real-time monitoring data to obtain a real-time data stream.

[0063] In some embodiments of this application, the real-time monitoring data includes environmental monitoring data, equipment operation data, pedestrian flow data, and energy consumption data.

[0064] In some embodiments of this application, the AR-based station-city operation and maintenance and adaptive knowledge base management method is adapted to the station-city integrated operation and maintenance scenario.

[0065] In some embodiments of this application, the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method is adapted to the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management system.

[0066] In some embodiments of this application, an AR-based inspection-based operation and maintenance performance quantification and adaptive knowledge base system for station-city integration scenarios is described. This system adopts a layered architecture design, consisting of a device perception and data acquisition layer, a data processing and performance calculation layer, an adaptive knowledge base management layer, an AR inspection interactive display layer, and a resource optimization decision-making layer, from bottom to top. Data transmission between layers is achieved through standardized interfaces, forming a complete data flow loop and realizing full-process automation from data acquisition to intelligent decision-making. The core innovation of the system lies in establishing a dynamic feedback mechanism between operation and maintenance performance and the knowledge base. By tracking the effectiveness of knowledge usage in real time, the system automatically adjusts the knowledge recommendation strategy to achieve adaptive optimization of the knowledge base, thereby continuously improving operation and maintenance efficiency and quality.

[0067] For example, Figure 2 The framework of a station and city operation and maintenance and adaptive knowledge base management system based on AR inspection is shown, which includes the following four main parts: operation and maintenance data collection; operation and maintenance performance quantification model and knowledge base adaptive collaboration; resource allocation optimization and decision support; AR intelligent inspection and visualization interaction.

[0068] In some embodiments of this application, environmental monitoring data, equipment operation data, pedestrian flow data, and energy consumption data are cleaned to remove invalid data, resulting in cleaned environmental monitoring data, cleaned equipment operation data, cleaned pedestrian flow data, and cleaned energy consumption data. The cleaned environmental monitoring data, cleaned equipment operation data, cleaned pedestrian flow data, and cleaned energy consumption data are then converted to standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow data, and standardized energy consumption data. Based on the standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow data, and standardized energy consumption data, anomaly detection is performed to obtain abnormal data. The abnormal data is then marked as abnormal. Based on the anomaly marks, the standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow data, and standardized energy consumption data, a real-time data stream is determined.

[0069] S102. Through a pre-determined quantification model of operation and maintenance performance, the real-time data stream is processed to obtain the operation and maintenance target links corresponding to the real-time data stream; wherein, the operation and maintenance target links include the operation and maintenance target equipment and the operation and maintenance data set related to the operation and maintenance target equipment.

[0070] In some embodiments of this application, the target device for operation and maintenance is one of various types of devices.

[0071] In some embodiments of this application, a performance evaluation model for operations and maintenance (O&M) is used to assess the performance of real-time data streams and obtain performance scores for each of the multiple O&M stages of the real-time data stream. Based on the performance scores of each of the multiple O&M stages, the O&M stages with performance scores less than a preset performance threshold are selected as the target O&M stages for the real-time data stream.

[0072] It should be noted that the preset performance threshold is 60, and the performance score ranges from 0 to 100. Generally, one device corresponds to one maintenance step. By evaluating the real-time data stream, the performance scores of multiple maintenance steps are obtained, which are the performance scores for multiple devices.

[0073] S103. Based on the operation and maintenance target links and the pre-determined adaptive knowledge base, perform intelligent resource scheduling to obtain multiple maintenance plans; and select the optimal maintenance plan from the multiple maintenance plans.

[0074] In some embodiments of this application, knowledge recommendation is performed on the target maintenance stage based on an adaptive knowledge base to obtain recommended information; based on the recommended information and pre-acquired skill matrix information, intelligent resource scheduling is performed to obtain multiple maintenance plans; wherein, the skill matrix information includes the professional ability and on-duty time of each staff member; for each of the multiple maintenance plans, the matching similarity between personnel, path and target maintenance stage is calculated to obtain the similarity value of each maintenance plan; based on the similarity value of each of the multiple maintenance plans, the maintenance plan with the highest similarity value is selected as the optimal maintenance plan.

[0075] S104. Based on the optimal maintenance plan, optimize the operation and maintenance strategy to obtain the optimized operation and maintenance strategy; display the optimized operation and maintenance strategy on the AR interactive interface, and carry out station and city operation and maintenance according to the optimized operation and maintenance strategy.

[0076] In some embodiments of this application, the original operation and maintenance strategy is optimized through the optimal maintenance scheme to obtain the optimized operation and maintenance strategy; the optimized operation and maintenance strategy is displayed on the AR interactive interface, and the station and city operation and maintenance is carried out according to the optimized operation and maintenance strategy.

[0077] For example, a real-time interactive and guidance interface is developed based on the Unity3D game engine. The interface integrates three interaction methods: touch, voice, and gesture, and displays relevant information overlaid on a real device. Remote assistance functionality is integrated, supporting video calls, screen sharing, and AR annotation sharing, allowing experts to remotely guide on-site operations.

[0078] An integrated operations and maintenance (O&M) performance analysis engine automatically pushes relevant maintenance knowledge and solutions when system O&M performance declines. The recommendation priority is adjusted based on the contribution of knowledge to improving O&M performance. Further deterioration in O&M performance triggers the supplementation and updating of relevant technical knowledge.

[0079] Intelligent task scheduling comprehensively considers factors such as task priority, personnel skill matching, and path planning when allocating tasks. A skill matrix is ​​established to record each person's professional abilities, and optimal matching is achieved through vector similarity calculation. The Dijkstra shortest path algorithm is integrated for path optimization, taking into account practical factors such as traffic conditions and elevator waiting times. Dynamic scheduling is supported, adjusting task allocation in real time based on actual execution. Through intelligent scheduling, the workload of manual scheduling can be reduced, avoiding resource waste and conflicts.

[0080] Based on historical and real-time monitoring data from the equipment, an LSTM (Long Short-Term Memory) network is integrated to predict the probability of failures. Operational strategies are automatically optimized according to actual operating conditions. Through predictive maintenance, faults can be addressed before they occur, reducing the impact of unexpected failures on operations.

[0081] Understandably, by connecting multiple sensors to the STM32F407 microcontroller, real-time perception of the operational status of key equipment in the station-city integration system is achieved, effectively improving the comprehensiveness and accuracy of data acquisition. The deployed edge computing nodes perform data cleaning and anomaly detection at the data source, significantly reducing reliance on cloud resources and improving the real-time performance and fault tolerance of data processing. By establishing a scientific quantitative evaluation system for the station-city integration system's operation and maintenance level, combined with AR visualization technology, intelligent knowledge bases, and adaptive optimization mechanisms, accurate identification, scientific allocation, and dynamic optimization of operation and maintenance resources are achieved, improving the response speed and accuracy of station-city integration operation and maintenance, thus significantly improving operation and maintenance efficiency. Simultaneously, resource allocation and utilization efficiency are optimized, thereby reducing operation and maintenance costs and enhancing system reliability.

[0082] In some embodiments of this application, the real-time preprocessing of real-time monitoring data in S101 to obtain a real-time data stream can be implemented through S201-S204, as follows:

[0083] S201. Perform data cleaning on environmental monitoring data, equipment operation data, pedestrian flow data, and energy consumption data respectively to remove invalid data and obtain cleaned environmental monitoring data, cleaned equipment operation data, cleaned pedestrian flow data, and cleaned energy consumption data.

[0084] S202. Convert the environmental monitoring data, equipment operation data, pedestrian flow data, and energy consumption data after cleaning into standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow data, and standardized energy consumption data.

[0085] S203. Based on standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow data, and standardized energy consumption data, perform anomaly detection to obtain abnormal data; and mark the abnormal data as abnormal.

[0086] S204. Based on anomaly markers, standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow status data, and standardized energy consumption data, determine the real-time data stream.

[0087] For example, by using multi-sensor fusion to collect real-time monitoring data from various devices in a city-station fusion system (STM32F407 microcontroller, DHT22 temperature and humidity sensor, ACS712 current sensor, etc.), accurate and timely equipment operation data can be provided to the upper-level system, providing a data foundation for performance calculation and fault diagnosis. Through multi-sensor fusion technology, the operating status of equipment can be comprehensively reflected, improving the accuracy of anomaly detection.

[0088] Edge computing nodes are deployed in nearby locations on-site to perform real-time preprocessing of the collected raw data, including data cleaning, format conversion, and anomaly detection. This reduces the data processing load on the cloud, improves system response speed, and ensures data reliability even under unstable network conditions.

[0089] It provides standardized data interfaces (designed using RESTful APIs) to solve the problem of data silos between different devices and systems, providing a unified data access method for upper-layer applications and simplifying the complexity of system integration.

[0090] Understandably, performing data cleaning and anomaly detection at the data source significantly reduces reliance on cloud resources and improves the real-time performance and fault tolerance of data processing. Building upon this, the system employs a streaming data processing architecture built with Kafka and Spark Streaming, enabling stable access and distributed processing of high-frequency data, thus significantly improving the timeliness of anomaly identification and feedback. On-site maintenance personnel receive real-time equipment status, fault information, and operational suggestions through an AR interface, allowing for immediate response to anomalies and enhancing the system's speed and accuracy in handling unexpected issues.

[0091] In some embodiments of this application, S102 can be implemented by S301 and S302, as follows:

[0092] S301. Through the operation and maintenance performance quantification model, the performance of real-time data stream is evaluated to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream.

[0093] In some embodiments of this application, the real-time data stream is classified by topic to obtain data on multiple different message topics; wherein, the multiple different message topics correspond to the types of various devices; the multiple different message topics are partitioned to obtain sub-topics of each of the multiple different message topics; and the data on the multiple different message topics are evaluated simultaneously by the operation and maintenance performance quantification model and the sub-topics of each of the multiple different message topics to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream.

[0094] In some embodiments of this application, for each message topic among multiple different message topics, a multi-indicator evaluation is performed on the data of each message topic using an operation and maintenance performance quantification model to obtain the corresponding equipment management indicator value, energy consumption management indicator value, personnel flow management indicator value, and environmental management indicator value for each message topic's data. The equipment management indicator value, energy consumption management indicator value, personnel flow management indicator value, and environmental management indicator value are then standardized to obtain equipment management standard value, energy consumption management standard value, personnel flow management standard value, and environmental management standard value. The weight values ​​for each of these standard values ​​are then determined. Based on these weight values, the equipment management standard value, energy consumption management standard value, personnel flow management standard value, and environmental management standard value are weighted to obtain a comprehensive indicator value. Based on the comprehensive indicator value, time decay processing and anomaly processing are performed to obtain the performance scores for each of the multiple operation and maintenance links in the real-time data stream.

[0095] S302. Based on the performance scores of each of the multiple operation and maintenance links, select the operation and maintenance link with a performance score less than the preset performance threshold as the operation and maintenance target link corresponding to the real-time data stream.

[0096] For example, a real-time stream processing engine (using Apache Kafka message queues and the Apache Spark Streaming computing framework) receives real-time data streams from edge computing nodes, performs data parsing, verification, computation, and storage. A topic partitioning mechanism enables distributed data processing; different message topics are created based on device type, and each topic has multiple partitions to improve parallel processing capabilities, enabling real-time processing and analysis of large-scale data and providing timely and accurate data support for performance calculations. The stream processing architecture can handle high-frequency data writes, ensuring the system's real-time performance and scalability.

[0097] A hierarchical indicator system is adopted for performance evaluation. The evaluation system includes four primary indicators (equipment management, energy consumption management, personnel management, and environmental management) and multiple secondary indicators. A comprehensive performance score is obtained through weighted calculation, which transforms the traditionally difficult-to-quantify operation and maintenance effects into comparable numerical indicators, providing an objective basis for operation and maintenance management decisions.

[0098] The module uses a PostgreSQL relational database to store performance calculation rules and historical data, and employs a scheduled task manager to periodically execute performance calculation tasks. The calculation process includes steps such as indicator standardization, weight allocation, time decay, and anomaly handling, ultimately outputting a performance score from 0 to 100. From the performance scores of multiple operation and maintenance stages, the operation and maintenance stages with performance scores less than 60 are selected as the target operation and maintenance stages corresponding to the real-time data stream.

[0099] Understandably, by using an operation and maintenance performance quantification model and an operation and maintenance level quantification evaluation system to evaluate the performance of real-time data streams, performance scores are obtained for each of the multiple operation and maintenance links of the real-time data stream. Based on the performance scores of each of the multiple operation and maintenance links, the operation and maintenance links with performance scores less than a preset performance threshold are selected as the operation and maintenance target links corresponding to the real-time data stream. Targeted operation and maintenance optimization can then be carried out based on the operation and maintenance target links.

[0100] In some embodiments of this application, S103 can be implemented by S401-S404, as follows:

[0101] S401. Based on an adaptive knowledge base, knowledge recommendations are made for the target operation and maintenance processes to obtain recommended information.

[0102] S402. Based on the recommended information and the pre-acquired skill matrix information, intelligent resource scheduling is performed to obtain multiple maintenance plans; among which, the skill matrix information includes the professional ability and on-duty time of each staff member.

[0103] S403. For each of the multiple maintenance plans, calculate the matching similarity between personnel, paths and operation and maintenance target links to obtain the similarity value of each maintenance plan.

[0104] S404. Based on the similarity value of each maintenance scheme among multiple maintenance schemes, the maintenance scheme with the highest similarity value is selected as the optimal maintenance scheme.

[0105] For example, the knowledge effectiveness tracking engine records the knowledge usage behavior of inspection personnel in real time by embedding monitoring code in the AR interactive interface. Whenever a user views or applies a piece of knowledge, the system automatically creates an operation tracking record, including user ID, knowledge ID, device ID, operation time, and operation result. A performance comparison mechanism before and after operation is established, calculating the actual contribution of knowledge by analyzing performance changes before and after knowledge use. Statistical analysis methods are used to verify the significance of the knowledge effect. The role of the knowledge effectiveness tracking engine is to establish a causal relationship between knowledge use and performance improvement, providing quantitative evidence for adaptive optimization of the knowledge base, identifying truly effective knowledge content, and eliminating ineffective or outdated knowledge.

[0106] A knowledge graph is constructed (using Neo4j graph database as the storage platform), and a BERT pre-trained model is used for entity recognition. Dependency parsing is used to identify relationships between entities. Incremental updates to the knowledge graph are supported; newly added maintenance records and expert knowledge are automatically integrated into the graph structure. Dispersed operational knowledge is organized into a structured relational network, enabling the discovery of potential connections between knowledge points and improving the accuracy of knowledge recommendations.

[0107] The intelligent recommendation algorithm engine employs a hybrid recommendation strategy combining collaborative filtering and content filtering. A deep learning recommendation model is built using the TensorFlow Recommenders framework, and Elasticsearch is integrated for knowledge content similarity retrieval. A user profiling system is established to collect data such as users' basic information, skill levels, operational preferences, and historical behavior. Context-aware recommendation is supported, adjusting the recommendation strategy based on environmental factors such as current time, location, device status, and task type. Recommendation results comprehensively consider multiple factors including relevance, knowledge quality, personal preference, and novelty.

[0108] This application proposes a multi-level AR-assisted system architecture for station-city integrated operation and maintenance scenarios, which covers a five-layer system: "equipment perception and data acquisition layer → data processing and performance calculation layer → adaptive knowledge base management layer → AR inspection and interactive display layer → resource optimization decision layer", realizing closed-loop linkage of data acquisition, performance evaluation, knowledge optimization, visualization interaction and task scheduling.

[0109] A dynamic feedback mechanism of "knowledge usage—performance change—contribution evaluation—recommendation strategy optimization" was constructed. Through performance comparison and significance analysis before and after operations, quantitative value assessment of knowledge and adaptive updates of recommendation strategies were achieved, improving the intelligence level of knowledge management. A multimodal AR interface integrating voice, gestures, and touch was built using Unity3D, enabling information visualization and overlay on physical devices. Combined with remote expert annotation and communication functions, this enhanced the efficiency of on-site inspections and problem handling, reducing reliance on personnel experience.

[0110] In some embodiments of this application, the AR-based station and city operation and adaptive knowledge base management method further includes:

[0111] Based on the optimal maintenance scheme, the adaptive knowledge base is updated to obtain the updated adaptive knowledge base.

[0112] Based on real-time monitoring data and historical data, fault prediction is performed to obtain prediction results;

[0113] Operation and maintenance optimizations are performed based on the prediction results and the updated adaptive knowledge base.

[0114] Understandably, by updating the adaptive knowledge base through the optimal maintenance scheme, the adaptive knowledge base can be improved, thereby achieving adaptive optimization of the knowledge base and continuously improving operational efficiency and quality.

[0115] Example 1: Subway Station Equipment Inspection Scenario

[0116] Maintenance personnel wearing AR glasses enter the subway platform for routine inspections. The system acquires real-time operational data from equipment such as escalators, ventilation and air conditioning systems, and lighting systems within the station through a multi-sensor fusion acquisition layer. Edge computing nodes clean and detect anomalies in the collected raw data such as temperature, current, and vibration, and upload standardized data to the data processing layer via a RESTful API.

[0117] The data processing layer's stream processing engine receives data streams based on Apache Kafka, and Spark Streaming distributes the data to different topic partitions for parallel computation according to device type. The performance calculation module calculates the real-time performance score for the current region based on a preset hierarchical indicator system (device management, energy consumption management, and environmental management). The knowledge effect tracking engine records every piece of knowledge accessed by operations and maintenance personnel in the AR interface and its usage time.

[0118] The Neo4j graph database in the adaptive knowledge base management layer stores the relationships between devices, faults, and solutions. When maintenance personnel scan a device using AR glasses, the BERT model identifies the device entity and extracts the device's current state features through dependency parsing. The intelligent recommendation engine, based on the TensorFlow Recommenders framework, combines user profiles (skill level, historical operation records) and the current environmental context (time, location, device type) to push relevant maintenance knowledge through collaborative filtering algorithms.

[0119] The AR interaction layer, based on Unity3D, overlays maintenance steps, safety precautions, and other information onto the real device. Maintenance personnel interact with the system via voice, gestures, or touch. After the operation is completed, the system automatically records data such as operation duration and step completeness. The knowledge effect tracking engine compares the changes in device performance indicators before and after the operation, calculates the contribution of the knowledge, and updates the recommendation weight.

[0120] The resource optimization decision-making layer plans the optimal inspection path based on the performance scores of each region and the current task priority using the Dijkstra algorithm, and assigns tasks to suitable personnel based on skill matrix vector matching. The LSTM prediction model analyzes historical data to identify potential failure risks and generates preventative maintenance tasks in advance.

[0121] Example 2: Operation and Maintenance Scenario of Commercial Area in Integrated Transportation Hub

[0122] The integrated station-city complex includes various business formats such as shopping malls, office areas, and public service areas. The system connects to DHT22 temperature and humidity sensors and ACS712 current sensors via an STM32F407 microcontroller to collect operating parameters of the air conditioning systems, elevator equipment, and fire protection systems in each area. When an edge node detects abnormal air conditioning energy consumption in a certain area, it immediately triggers an alarm and marks the status of that device.

[0123] After receiving abnormal data, the data processing layer recalculates the energy management index score for that region and updates the overall performance. The Neo4j graph in the knowledge base management layer stores historical fault modes and expert repair experience for this type of air conditioner. The recommendation engine retrieves three relevant pieces of knowledge: routine inspection procedures, filter cleaning methods, and refrigerant testing steps.

[0124] Based on the skill level and past operational feedback of the current on-duty personnel, the system prioritizes recommending filter cleaning methods they have successfully applied. Maintenance personnel view visual guidance through an AR interface, and the system tracks their operational steps and time in real time. After the operation is completed, the system automatically compares energy consumption data and performance score changes before and after the treatment to verify the significance of the knowledge's effectiveness. If performance improvement is not significant after repeated use of the knowledge, the system automatically lowers its recommendation priority and triggers expert review.

[0125] The resource scheduling layer detected a continuous decline in the region's performance and predicted a high probability of equipment failure within the next 72 hours using an LSTM model. It then automatically generated a deep maintenance task and assigned it to qualified technicians. The task scheduling module comprehensively considered the personnel's current location, skill matching, and the priority of other pending tasks to calculate the optimal task assignment plan.

[0126] The beneficial effects of this application are as follows:

[0127] Achieving deep integration of station-city integrated operation and maintenance effectiveness quantification and knowledge management: A closed-loop mechanism of "performance indicators—knowledge usage—effectiveness evaluation—strategy updates" is constructed, centered on hierarchical indicators across four dimensions: equipment management, energy consumption management, personnel flow management, and environmental management. This ensures that knowledge value is measurable, comparable, and traceable. Through performance comparison and significance testing before and after operations, highly efficient knowledge is selected, while inefficient or outdated knowledge is eliminated. Knowledge lifecycle management is driven by an indicator system. Performance scores, knowledge contribution, usage frequency, and on-site feedback serve as the decision-making basis for knowledge upload, de-upload, and version iteration, forming a continuously self-optimizing knowledge ecosystem.

[0128] Improving the speed and accuracy of station-city integrated operation and maintenance: By connecting multiple sensors to the STM32F407 microcontroller, real-time perception of the operating status of key equipment in the station-city integrated system is achieved, effectively improving the comprehensiveness and accuracy of data acquisition. The deployed edge computing nodes perform data cleaning and anomaly detection at the data source end, greatly reducing dependence on cloud resources and improving the real-time performance and fault tolerance of data processing. Based on this, the system adopts a streaming data processing architecture built with Kafka and Spark Streaming, enabling stable access and distributed processing of high-frequency data, significantly improving the timeliness of anomaly identification and feedback. On-site maintenance personnel receive real-time equipment status, fault information, and operational suggestions through an AR interface, enabling immediate response to abnormal situations and improving the speed and accuracy of the operation and maintenance system's response to unexpected problems.

[0129] Optimizing resource allocation and utilization efficiency: In terms of resource scheduling, multi-dimensional factors such as task priority assessment, personnel skill matching, and path planning are introduced. By constructing a skill matrix and combining it with a vector similarity algorithm, precise matching of personnel and tasks is achieved, effectively reducing the mismatch between personnel and positions and resource waste in traditional scheduling. Simultaneously, Dijkstra's shortest path algorithm is used to optimize task paths, considering factors such as traffic, spatial location, and elevator waiting times, to achieve optimal task route configuration under limited resources. In practical applications, the dynamic scheduling mechanism can adjust the allocation strategy in real time according to the task execution status, ensuring that tasks are completed in an orderly manner according to priority, improving the operational coverage and manpower utilization efficiency per unit time.

[0130] Based on the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method of the above embodiments, this application also provides an AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management system, such as... Figure 3 As shown, Figure 3This is a schematic diagram of the structure of a station and city operation and maintenance and adaptive knowledge base management system based on AR inspection, provided in an embodiment of this application. The station and city operation and maintenance and adaptive knowledge base management system 3 based on AR inspection includes: an acquisition module 301, a data processing module 302, a resource scheduling module 303, and a display module 304, wherein...

[0131] The acquisition module 301 is used to acquire real-time monitoring data of various devices in the station-city integration system; and to perform real-time preprocessing on the real-time monitoring data to obtain a real-time data stream;

[0132] The data processing module 302 is used to process the real-time data stream using a pre-determined operation and maintenance performance quantification model to obtain the operation and maintenance target link corresponding to the real-time data stream; wherein, the operation and maintenance target link includes the operation and maintenance target equipment and the operation and maintenance data set related to the operation and maintenance target equipment;

[0133] The resource scheduling module 303 is used to perform intelligent resource scheduling based on the operation and maintenance target links and a pre-determined adaptive knowledge base to obtain multiple maintenance schemes; and to filter the multiple maintenance schemes to obtain the optimal maintenance scheme.

[0134] The display module 304 is used to optimize the operation and maintenance strategy based on the optimal maintenance scheme to obtain the optimized operation and maintenance strategy; and to display the optimized operation and maintenance strategy on the AR interactive interface, and to perform station and city operation and maintenance according to the optimized operation and maintenance strategy.

[0135] In some embodiments of this application, the real-time monitoring data includes environmental monitoring data, equipment operation data, pedestrian flow status data, and energy consumption data;

[0136] The data processing module 302 is further configured to perform data cleaning on the environmental monitoring data, the equipment operation data, the pedestrian flow status data, and the energy consumption data respectively, removing invalid data to obtain cleaned environmental monitoring data, cleaned equipment operation data, cleaned pedestrian flow status data, and cleaned energy consumption data; perform data format conversion on the cleaned environmental monitoring data, the cleaned equipment operation data, the cleaned pedestrian flow status data, and the cleaned energy consumption data respectively to obtain standardized environmental monitoring data, standardized equipment operation data, standardized pedestrian flow status data, and standardized energy consumption data; perform anomaly detection based on the standardized environmental monitoring data, the standardized equipment operation data, the standardized pedestrian flow status data, and the standardized energy consumption data to obtain abnormal data; and mark the abnormal data as anomalies; and determine the real-time data stream based on the anomaly marks, the standardized environmental monitoring data, the standardized equipment operation data, the standardized pedestrian flow status data, and the standardized energy consumption data.

[0137] In some embodiments of this application, the data processing module 302 is further configured to perform performance evaluation on the real-time data stream through the operation and maintenance performance quantification model to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream; based on the performance scores of each of the multiple operation and maintenance links, select the operation and maintenance link with the performance score less than a preset performance threshold as the operation and maintenance target link corresponding to the real-time data stream.

[0138] In some embodiments of this application, the data processing module 302 is further configured to classify the real-time data stream by topic to obtain data on multiple different message topics; wherein, the multiple different message topics correspond to the types of various devices; partition the multiple different message topics to obtain sub-topics of each of the multiple different message topics; and simultaneously evaluate the data on the multiple different message topics by using the operation and maintenance performance quantification model and the sub-topics of each of the multiple different message topics to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream.

[0139] In some embodiments of this application, the data processing module 302 is further configured to: evaluate the data of each message topic among the plurality of different message topics using the operation and maintenance performance quantification model, thereby obtaining equipment management indicator values, energy consumption management indicator values, personnel flow management indicator values, and environmental management indicator values ​​corresponding to the data of each message topic; perform indicator standardization processing on the equipment management indicator values, energy consumption management indicator values, personnel flow management indicator values, and environmental management indicator values ​​to obtain equipment management standard values, energy consumption management standard values, personnel flow management standard values, and environmental management standard values; determine the weight values ​​of each of the equipment management standard values, energy consumption management standard values, personnel flow management standard values, and environmental management standard values; perform weighted processing on the equipment management standard values, energy consumption management standard values, personnel flow management standard values, and environmental management standard values ​​based on the weight values ​​to obtain a comprehensive indicator value; and perform time decay processing and anomaly processing based on the comprehensive indicator value to obtain the performance scores of each of the multiple operation and maintenance links of the real-time data stream.

[0140] In some embodiments of this application, the resource scheduling module 303 is further configured to: recommend knowledge to the target maintenance stage based on the adaptive knowledge base to obtain recommended information; perform intelligent resource scheduling based on the recommended information and pre-acquired skill matrix information to obtain the multiple maintenance schemes; wherein, the skill matrix information includes the professional capabilities and on-duty time of each staff member; calculate the matching similarity between personnel, path and the target maintenance stage for each of the multiple maintenance schemes to obtain a similarity value for each maintenance scheme; and select the maintenance scheme with the highest similarity value as the optimal maintenance scheme based on the similarity value of each of the multiple maintenance schemes.

[0141] In some embodiments of this application, the resource scheduling module 303 is further configured to update the adaptive knowledge base based on the optimal maintenance scheme to obtain an updated adaptive knowledge base; perform fault prediction based on the real-time monitoring data and historical data to obtain a prediction result; and perform operation and maintenance optimization based on the prediction result and the updated adaptive knowledge base.

[0142] Based on the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method of the above embodiments, this application also provides an AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management device, such as... Figure 4 As shown, Figure 4This is a schematic diagram of the structure of an AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management device provided in an embodiment of this application. The AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management device 4 includes a processor 401 and a memory 402. The memory 402 is used to store computer programs; the processor 401 is used to call and run the computer programs from the memory to execute the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method as described in the above embodiment.

[0143] In the embodiments of this application, the processor 401 described above can be at least one of the following: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field-Programmable Gate Array (FPGA), Central Processing Unit (CPU), Controller, Microcontroller, and Microprocessor. It is understood that for different devices, the electronic device used to implement the above processor function can also be other types, and the embodiments of this application do not specifically limit it.

[0144] This application provides a computer-readable storage medium storing a computer program for implementing, when executed by a processor, the station and city operation and maintenance and adaptive knowledge base management method based on AR inspection as described in any of the above embodiments.

[0145] For example, the program instructions corresponding to the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method in this embodiment can be stored on storage media such as optical discs, hard disks, and USB flash drives. When the program instructions corresponding to the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method in the storage media are read or executed by an electronic device, the AR-based inspection-based station and city operation and maintenance and adaptive knowledge base management method described in any of the above embodiments can be realized.

[0146] Furthermore, in the embodiments of this application, the functional modules can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional module.

[0147] If the integrated unit is implemented as a software functional module and is not sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this embodiment, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the method of this embodiment. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0148] It should be understood that the phrases "one embodiment," "an embodiment," or "some embodiments" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment," "in one embodiment," or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the embodiments in this application are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments; their similarities or commonalities can be referred to mutually, and for the sake of brevity, these will not be repeated here.

[0149] The modules described above as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules. They may be located in one place or distributed across multiple network units. Some or all of the modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0150] In addition, each functional module in the various embodiments of this application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated modules can be implemented in hardware or in the form of hardware plus software functional units.

[0151] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0152] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0153] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0154] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0155] The above description is merely an embodiment of this application, but the protection scope of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A station and city operation and maintenance and adaptive knowledge base management method based on AR inspection, characterized in that, The method comprises: acquiring real-time monitoring data of various devices in a station-city integration system; and performing real-time preprocessing on the real-time monitoring data to obtain a real-time data stream; performing data processing on the real-time data stream through a pre-determined operation and maintenance performance quantification model to obtain an operation and maintenance target link corresponding to the real-time data stream; wherein the operation and maintenance target link comprises an operation and maintenance target device and an operation and maintenance data set related to the operation and maintenance target device; based on the operation and maintenance target link and a pre-determined adaptive knowledge base, performing intelligent resource scheduling to obtain a plurality of maintenance schemes; and selecting the maintenance schemes to obtain an optimal maintenance scheme; based on the optimal maintenance scheme, performing operation and maintenance strategy optimization to obtain an optimized operation and maintenance strategy; and displaying the optimized operation and maintenance strategy on an AR interactive interface and performing station-city operation and maintenance according to the optimized operation and maintenance strategy; wherein the data processing on the real-time data stream through the pre-determined operation and maintenance performance quantification model to obtain the operation and maintenance target link corresponding to the real-time data stream comprises: performing subject classification on the real-time data stream to obtain data of a plurality of different message subjects; wherein the plurality of different message subjects correspond to types of the various devices; partitioning the plurality of different message subjects to obtain respective sub-topics of the plurality of different message subjects; performing performance evaluation on the data of the plurality of different message subjects through the operation and maintenance performance quantification model and the respective sub-topics of the plurality of different message subjects to obtain performance scores of a plurality of operation and maintenance links of the real-time data stream; based on the performance scores of the plurality of operation and maintenance links, selecting an operation and maintenance link with a performance score less than a pre-set performance threshold as the operation and maintenance target link corresponding to the real-time data stream.

2. The method of claim 1, wherein, The real-time monitoring data comprises environmental monitoring data, device operation data, people flow state data and energy consumption data; The real-time preprocessing on the real-time monitoring data to obtain the real-time data stream comprises: performing data cleaning on the environmental monitoring data, the device operation data, the people flow state data and the energy consumption data respectively to remove invalid data and obtain cleaned environmental monitoring data, cleaned device operation data, cleaned people flow state data and cleaned energy consumption data; performing data format conversion on the cleaned environmental monitoring data, the cleaned device operation data, the cleaned people flow state data and the cleaned energy consumption data respectively to obtain standardized environmental monitoring data, standardized device operation data, standardized people flow state data and standardized energy consumption data; performing anomaly detection based on the standardized environmental monitoring data, the standardized device operation data, the standardized people flow state data and the standardized energy consumption data to obtain anomaly data; and performing anomaly marking on the anomaly data; determining the real-time data stream based on the anomaly marking, the standardized environmental monitoring data, the standardized device operation data, the standardized people flow state data and the standardized energy consumption data.

3. The method of claim 1, wherein, The performance evaluation of the data of the plurality of different message topics is simultaneously performed by the operation and maintenance performance quantification model and the sub-topics of the plurality of different message topics respectively, and a plurality of performance scores of the operation and maintenance links of the real-time data stream are obtained, including: For each message topic in the plurality of different message topics, the data of each message topic in the plurality of different message topics is evaluated by the operation and maintenance performance quantification model, and equipment management index values, energy consumption management index values, people flow management index values and environment management index values corresponding to the data of each message topic are obtained; The equipment management index values, the energy consumption management index values, the people flow management index values and the environment management index values are subjected to index standardization processing, and equipment management standard values, energy consumption management standard values, people flow management standard values and environment management standard values are obtained; The weight values of the equipment management standard values, the energy consumption management standard values, the people flow management standard values and the environment management standard values are determined; Based on the weight values, the equipment management standard values, the energy consumption management standard values, the people flow management standard values and the environment management standard values are subjected to weighted processing, and a comprehensive index value is obtained; Based on the comprehensive index value, time decay processing and abnormal processing are performed, and a plurality of performance scores of the operation and maintenance links of the real-time data stream are obtained.

4. The method of claim 1, wherein, The intelligent resource scheduling is performed based on the operation and maintenance target link and the pre-determined adaptive knowledge base, and a plurality of maintenance schemes are obtained; and the plurality of maintenance schemes are screened to obtain an optimal maintenance scheme, including: Based on the adaptive knowledge base, knowledge recommendation is performed on the operation and maintenance target link, and recommendation information is obtained; Based on the recommendation information and pre-acquired skill matrix information, intelligent resource scheduling is performed, and the plurality of maintenance schemes are obtained; wherein the skill matrix information includes professional ability and on-duty time of each worker; For each maintenance scheme in the plurality of maintenance schemes, the matching similarity of personnel, path and the operation and maintenance target link is calculated, and a similarity value of each maintenance scheme is obtained; Based on the similarity value of each maintenance scheme in the plurality of maintenance schemes, the maintenance scheme with the highest similarity value is selected as the optimal maintenance scheme.

5. The method of claim 1, wherein, The method further includes: Based on the optimal maintenance scheme, knowledge update is performed on the adaptive knowledge base, and an updated adaptive knowledge base is obtained; Based on the real-time monitoring data and historical data, fault prediction is performed, and a prediction result is obtained; According to the prediction result and the updated adaptive knowledge base, operation and maintenance optimization is performed.

6. An AR-based inspection station city operation and maintenance and adaptive knowledge base management system, characterized in that, The station and city operation and maintenance and adaptive knowledge base management system based on AR inspection includes an acquisition module, a data processing module, a resource scheduling module and a display module, wherein The acquisition module is configured to acquire real-time monitoring data of various devices in a station and city fusion system, and perform real-time preprocessing on the real-time monitoring data to obtain a real-time data stream; The data processing module is configured to perform data processing on the real-time data stream by using a predetermined operation and maintenance performance quantification model to obtain an operation and maintenance target link corresponding to the real-time data stream; wherein the operation and maintenance target link comprises an operation and maintenance target device and an operation and maintenance data set related to the operation and maintenance target device. The resource scheduling module is configured to perform intelligent resource scheduling based on the operation and maintenance target link and a predetermined adaptive knowledge base to obtain a plurality of maintenance schemes; and to select the optimal maintenance scheme from the plurality of maintenance schemes. The display module is configured to perform operation and maintenance strategy optimization based on the optimal maintenance scheme to obtain an optimized operation and maintenance strategy; and to display the optimized operation and maintenance strategy on an AR interaction interface and perform station and city operation and maintenance according to the optimized operation and maintenance strategy. The data processing module is further configured to perform subject classification on the real-time data stream to obtain data of a plurality of different message subjects; wherein the plurality of different message subjects correspond to types of the various types of devices; to perform partitioning on the plurality of different message subjects to obtain respective sub-topics of the plurality of different message subjects; to perform performance evaluation on the data of the plurality of different message subjects simultaneously by using the operation and maintenance performance quantification model and the respective sub-topics of the plurality of different message subjects to obtain performance scores of a plurality of operation and maintenance links of the real-time data stream; and to select an operation and maintenance link with a performance score less than a predetermined performance threshold as the operation and maintenance target link corresponding to the real-time data stream based on the performance scores of the plurality of operation and maintenance links.

7. An AR inspection-based station-city operation and maintenance and adaptive knowledge base management device, characterized by, Comprise: a processor and a memory, the memory is configured to store a computer program; the processor is configured to call and run the computer program from the memory to execute the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, executable instructions are stored for causing the processor to implement the method of any one of claims 1 to 5 when executed.

Images