Operation and maintenance technology service remote guidance interaction method and system combined with agent assistance

Abstract

The invention provides an operation and maintenance technology service remote guidance interaction method and system combined with agent assistance, and relates to the field of operation and maintenance technology services. Firstly, a remote guidance interaction request which is initiated by operation and maintenance personnel and comprises operation and maintenance object information, operation and maintenance problem description and an expected guidance direction is received; and performing operation and maintenance scene feature recognition through the virtual artificial intelligence agent, extracting key features, then generating an operation and maintenance guidance scheme containing a scene adaptive operation guidance sequence and a real-time semantic interaction node, and pushing the operation and maintenance guidance scheme to an operation and maintenance personnel terminal. Operation feedback description is received in real time, a guidance scheme is adjusted through a feedback semantic analysis module, an updated scheme is obtained and pushed, meanwhile, the evolution track of the guidance scheme and the operation feedback description in the interaction process are stored, and efficient, accurate and dynamic remote guidance interaction is achieved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and more specifically, to a remote guidance interaction method and system for operation and maintenance technical services combined with intelligent agent assistance. Background Art

[0002] In traditional operations and maintenance technical services, when operators encounter complex technical issues requiring remote guidance, they typically rely on human experts for communication. However, human experts are limited and may not be able to respond promptly to operators' requests due to factors such as location and time. Furthermore, during manual guidance, information transmission and understanding can be distorted, resulting in inefficient guidance.

[0003] While some existing remote guidance systems can provide certain guidance functions, they often lack the ability to accurately identify O&M scenarios. Based on the specific information of the O&M object, a detailed description of the O&M problem, and the desired guidance direction, they are unable to automatically extract key features, such as the type of operating parameters associated with the O&M object, the scope of the fault impact corresponding to the O&M problem, and the level of operational complexity that matches the desired guidance direction. This results in the generated guidance plans lacking specificity and adaptability, failing to meet the needs of actual O&M scenarios. Furthermore, while O&M personnel are performing operations, existing systems struggle to dynamically adjust guidance plans based on real-time feedback, making efficient and accurate remote guidance interactions impossible. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a remote guidance interaction method and system for operation and maintenance technical services combined with intelligent agent assistance.

[0005] According to a first aspect of the present application, a method for remote guidance interaction of operation and maintenance technical services combined with agent assistance is provided, the method comprising: Receive a remote guidance interaction request for operation and maintenance technical services initiated by an operation and maintenance personnel, wherein the remote guidance interaction request includes operation and maintenance object information, operation and maintenance problem description, and desired guidance direction; The virtual artificial intelligence agent identifies the operation and maintenance scenario characteristics of the remote guidance interaction request, extracts the operation parameter type associated with the operation and maintenance object, the fault impact range corresponding to the operation and maintenance problem, and the operation complexity level matching the expected guidance direction; Based on the operation and maintenance scenario feature recognition results, a dynamic guidance strategy generation module of the virtual artificial intelligence agent is called to generate an operation and maintenance guidance plan including a scenario-adaptive operation guidance sequence and real-time semantic interaction nodes; Pushing the operation and maintenance guidance plan to the operation and maintenance personnel terminal, receiving in real time the operation feedback description returned by the operation and maintenance personnel after performing the operation based on the operation and maintenance guidance plan, and inputting the operation feedback description into the feedback semantic parsing module of the virtual artificial intelligence agent; The real-time semantic interaction nodes in the operation and maintenance guidance plan are associated with the feedback semantic parsing module, the step content and presentation order of the scenario-adaptive operation guidance sequence are adjusted, and an updated operation and maintenance guidance plan is obtained. The updated operation and maintenance guidance plan is pushed to the operation and maintenance personnel terminal and the guidance plan evolution trajectory and operation feedback description during this interaction process are stored.

[0006] According to the second aspect of the present application, a remote guidance interaction system for operation and maintenance technical services combined with the assistance of an intelligent agent is provided. The remote guidance interaction system for operation and maintenance technical services combined with the assistance of an intelligent agent includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the remote guidance interaction system for operation and maintenance technical services combined with the assistance of an intelligent agent implements the aforementioned remote guidance interaction method for operation and maintenance technical services combined with the assistance of an intelligent agent.

[0007] According to the third aspect of the present application, a computer-readable storage medium is provided, in which computer-executable instructions are stored. When the computer-executable instructions are executed, the aforementioned remote guidance interaction method for operation and maintenance technical services combined with intelligent agent assistance is implemented.

[0008] According to any of the above aspects, the technical effects of this application are: By receiving remote guidance interaction requests initiated by operation and maintenance personnel and using virtual artificial intelligence agents to conduct comprehensive operation and maintenance scenario feature recognition on the requests, key information can be accurately extracted, such as the operating parameter types associated with the operation and maintenance objects, the fault impact range corresponding to the operation and maintenance problems, and the operation complexity level that matches the expected guidance direction. Based on the scene feature recognition results, an operation and maintenance guidance plan containing a scene-adaptive operation guidance sequence and real-time semantic interaction nodes is generated, which can effectively guide operation and maintenance personnel to solve practical problems. During the operation and maintenance personnel's operation process, the operation feedback description is received in real time and the real-time semantic interaction nodes are associated through the feedback semantic parsing module. The step content and presentation order of the scene-adaptive operation guidance sequence are dynamically adjusted to obtain an updated operation and maintenance guidance plan, realizing real-time optimization of the guidance plan and improving the efficiency and accuracy of remote guidance. At the same time, storing the evolution trajectory of the guidance plan and the operation feedback description during this interaction process helps to improve the overall operation and maintenance technical service level and realize efficient, accurate, and dynamic operation and maintenance technical service remote guidance interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without any creative work.

[0010] Figure 1 A flow chart of a remote guidance interaction method for operation and maintenance technical services combined with agent assistance provided in an embodiment of the present application is shown; Figure 2 A schematic diagram of the component structure of the remote guidance interaction system for operation and maintenance technical services combined with the assistance of an intelligent agent, provided in an embodiment of the present application, is shown for realizing the above-mentioned remote guidance interaction method for operation and maintenance technical services combined with the assistance of an intelligent agent. DETAILED DESCRIPTION

[0011] The following describes the embodiments of the present application in conjunction with the accompanying drawings. It should be understood that the embodiments described below in conjunction with the accompanying drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions of the embodiments of the present application.

[0012] Those skilled in the art will appreciate that, unless expressly stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present application refer to that the corresponding features can be implemented as the features, information, data, steps, operations, elements and / or components presented, but do not exclude implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the present technical field. It should be understood that when an element is said to be "connected" or "coupled" to another element, the element can be directly connected or coupled to the other element, or it can refer to that the element and the other element establish a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling, and the term "and / or" used herein indicates at least one of the items defined by the term, for example, "A and / or B" can be implemented as "A", or as "B", or as "A and B".

[0013] In order to make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail with reference to the accompanying drawings. The following description of several exemplary embodiments will illustrate the technical solutions of the embodiments of the present application and the technical effects produced by the technical solutions of the present application. It should be noted that the following embodiments can refer to, draw on or combine with each other, and the same terms, similar features and similar implementation steps in different embodiments will not be repeated.

[0014] Figure 1 The flowchart of the method and system for remote guidance interaction of operation and maintenance technical services combined with agent assistance provided by an embodiment of the present application is shown. It should be understood that in other embodiments, the order of some steps in the method for remote guidance interaction of operation and maintenance technical services combined with agent assistance of this embodiment can be shared according to actual needs, or some steps can be omitted or maintained. The detailed steps of the method for remote guidance interaction of operation and maintenance technical services combined with agent assistance include: Step S110: receiving a remote guidance interaction request for operation and maintenance technical services initiated by an operation and maintenance personnel, wherein the remote guidance interaction request includes operation and maintenance object information, operation and maintenance problem description, and desired guidance direction.

[0015] This example focuses on the operation and maintenance of smart air conditioners in chain stores. During on-site maintenance of a smart air conditioner at a store, an on-site engineer initiates a remote guidance interaction request through a mobile device operation and maintenance application. The operation and maintenance target information, including the air conditioner brand and model, installation date, store location, installation region, and unique hardware code, is automatically retrieved by the application from the store's device archive.

[0016] The engineer enters a text description of the maintenance issue, including the time of issue occurrence (e.g., at startup, during operation), specific symptoms (e.g., cooling drop, abnormal code, operating noise), preliminary checks performed (e.g., filter cleanliness, power connection), and environmental changes (e.g., voltage fluctuation, sudden increase in ambient temperature). The desired guidance should clearly define the type of technical support, such as fault code interpretation, component replacement specifications, noise troubleshooting steps, or parameter adjustment solutions. The request is encrypted and transmitted to the technical center's virtual artificial intelligence agent server.

[0017] Step S120: The virtual artificial intelligence agent is used to identify the operation and maintenance scenario features of the remote guidance interaction request, and extract the operation parameter type associated with the operation and maintenance object, the fault impact range corresponding to the operation and maintenance problem, and the operation complexity level matching the expected guidance direction.

[0018] After receiving the request, the virtual AI agent initiates the feature recognition process, first converting the unstructured text description into analyzable feature data. It then calls the device feature database based on the O&M target information, matches the model's technical manual to identify key operating parameter types, and then selects core parameters based on the problem symptoms. It analyzes the problem's associated components and store needs, assessing the impact of the fault on functional modules and operations, such as the risk of a cooling issue affecting store comfort and customer experience. The complexity level is determined based on the operation complexity assessment model, taking into account the operation type and technical difficulty.

[0019] Step S121: Input the remote guidance interaction request into the scene feature extraction unit of the virtual artificial intelligence agent, and split out the operation and maintenance object's domain, operation and maintenance object's core component type, and problem triggering conditions and problem-related components in the operation and maintenance problem description.

[0020] The scenario feature extraction unit performs multi-level decomposition of the request. It analyzes the O&M object information to identify the commercial air conditioning O&M domain, further subdividing it into the intelligent variable-frequency air conditioning subdomain. A dedicated feature recognition model is then invoked. Core component types are extracted based on the device model and problem symptoms. For example, refrigeration problems are associated with the compressor, condenser, and refrigerant circulation system; exception codes are associated with the control board and sensor components; and technical parameters and a fault mode library are associated with each component.

[0021] The problem description is broken down into trigger conditions, including equipment operation phases (startup, stable operation, shutdown), environmental conditions (temperature, humidity, voltage range), and operational behaviors (parameter adjustment, mode switching). Keyword matching and semantic understanding are used to create a structured list. Cause-and-effect relationships are used to identify associated components, such as abnormal noise associated with the fan assembly and compressor mounting structure, and display anomalies associated with the display assembly and driver circuitry. The breakdown results are stored in a feature vector database.

[0022] Step S122: Call the operation and maintenance field parameter library associated with the virtual artificial intelligence agent, and match the operation parameter type that needs to be monitored by the corresponding operation and maintenance object under normal operating conditions according to the field to which the operation and maintenance object belongs and the core component type of the operation and maintenance object.

[0023] The O&M parameter library stores key parameter standards hierarchically by device domain and component type. The agent performs multi-level searches based on the commercial smart air conditioning domain and core component type (compressor, condenser, control system). Compressor components match parameters such as operating current, operating pressure, exhaust temperature, and operating frequency; condenser components match parameters such as heat sink temperature, fan speed, and inlet / outlet temperature difference; and control systems match parameters such as the deviation between set and actual temperature, mode switching response time, and sensor feedback values.

[0024] Each parameter type is associated with a monitoring location, normal range, measurement tool requirements, and the permissible range of operating condition fluctuations. Extracted parameters are then re-matched with the device model characteristics, eliminating common parameters and retaining targeted parameters, such as the pressure monitoring parameters specific to a particular compressor brand, to ensure the parameter types are practical and effective.

[0025] Step S123: Based on the problem triggering conditions and problem-related components in the operation and maintenance problem description, combined with the fault impact analysis logic of the virtual artificial intelligence agent, determine the scope of adjacent components that may be affected by the operation and maintenance problem and the risk level of service interruption, and form the fault impact scope corresponding to the operation and maintenance problem.

[0026] Fault impact analysis logic is based on a fault propagation model that includes component connectivity, functional dependencies, and fault transmission paths. The initial fault point is determined by inputting the problem trigger conditions and associated components. Based on physical connections and functional collaboration, the scope of adjacent components to which the fault propagates is deduced. For example, a compressor failure could affect the condenser and evaporator, while a control board failure could affect sensors and actuators.

[0027] The service interruption risk level is assessed based on a comprehensive assessment of store operating hours (peak / off-peak), equipment importance (core area / auxiliary area), and repair difficulty. Air conditioning failures in core areas during peak hours are associated with a higher risk level, while failures in auxiliary areas during off-peak hours are associated with a lower risk level. The scope of the failure is described by combining the adjacent component ranges and risk levels.

[0028] Step S1231: Input the problem triggering conditions and problem-related components in the operation and maintenance problem description into the fault impact analysis unit of the virtual artificial intelligence agent.

[0029] After receiving the problem triggering conditions and the associated components, the Fault Impact Analysis Unit performs a time series analysis of the triggering conditions to determine the duration of the problem, its frequency characteristics, and whether there is a periodic pattern. It also creates a 3D model of the associated components, reconstructing their installation position within the overall device structure, their connection ports, and their physical distance from other components.

[0030] Step S1232: Call the component association map of the virtual artificial intelligence agent, locate the connection relationship of the problem-related components in the operation and maintenance system according to the problem-related components, and identify the directly connected first-level adjacent components and the indirectly related second-level adjacent components.

[0031] The component association graph stores the connection relationships between device components in a node-edge structure, with nodes representing components and edges representing connection types (e.g., mechanical, electrical, or signal connections). By traversing the graph through the problem-associated component nodes, directly connected first-level adjacent components are extracted (e.g., a compressor directly connected to a refrigerant pipeline). Further traversal through the first-level adjacent component nodes identifies indirectly connected second-level adjacent components (e.g., a condenser connected to a refrigerant pipeline). Each adjacent component is labeled with the connection strength and signal transmission direction. A higher connection strength indicates a greater probability of fault transmission.

[0032] Step S1233: Analyze the functional impact of the problem triggering condition on the problem-related components, determine whether the functional impact will be transmitted to the first-level adjacent components and the second-level adjacent components through the connection relationship, and determine the range of adjacent components that may be affected.

[0033] Analyze the impact of the problem triggering condition (e.g., voltage anomaly) on the functionality of the associated components (e.g., control motherboard) (e.g., signal processing anomalies, unstable power supply). Combined with the connection types and strengths in the component association diagram, simulate the transmission paths of the functional impact: electrical connections may transmit voltage anomalies, while signal connections may transmit data errors. Evaluate the potential for transmission for each of the primary and secondary adjacent components. Components with transmission risks are included in the list of potentially affected adjacent components, and the severity of the impact is noted (e.g., severe impact, minor impact).

[0034] Step S1234: Based on the range of adjacent components that may be affected and the service weight of each component in the operation and maintenance system, the possibility of service interruption and the number of affected users are calculated to determine the service interruption risk level.

[0035] Possibly affected adjacent components are assigned service weights based on their functional importance within the system (e.g., core refrigeration components have a higher weight than auxiliary display components) and their operational dependency on the store (e.g., lobby air conditioning components have a higher weight than warehouse air conditioning components). The probability of service interruption caused by a component failure is calculated using historical failure data, and the number of affected users is estimated using real-time store traffic data. The probability of service interruption (high / medium / low) is combined with the number of affected users (many / medium / few) to map the resulting service interruption risk level (e.g., extremely high, high, medium, low).

[0036] Step S1235: Combine the range of adjacent components that may be affected with the service interruption risk level to form the fault impact range corresponding to the operation and maintenance problem.

[0037] A structured table presents the range of potentially affected adjacent components, including component name, impact level, and transmission path description. The service disruption risk level is accompanied by a summary of the calculation basis, detailing the time period of real-time passenger flow data and historical disruption probabilities used in the assessment. These two factors are combined with time-based information (such as the expected duration of the impact) to form a complete description of the fault's impact scope.

[0038] Step S124: Analyze the operation target expression in the expected guidance direction, combine it with the operation complexity evaluation model of the virtual artificial intelligence agent, and determine the operation complexity level that matches the expected guidance direction based on the number of components involved in the operation, the correlation of the operation steps, and the operation fault tolerance requirements.

[0039] Semantic analysis is used to describe the desired operational objectives, extracting core operational verbs (such as replace, adjust, and test) and operational objects, and clarifying the operation type. The operational complexity assessment model assesses operations along three dimensions: The component count dimension counts the total number of components touched or adjusted; a larger number increases the base value; the step relevance dimension analyzes step dependencies and sequence constraints; the stronger the relevance, the greater the difficulty in decomposition; and the operational fault tolerance requirement dimension determines the fault tolerance threshold based on the severity of the consequences of a failure (equipment damage, safety risk); higher requirements indicate increased complexity.

[0040] The model calculates a weighted comprehensive score and maps it to basic, advanced, and expert complexity levels. Basic corresponds to single-step, low-risk operations; advanced corresponds to multi-step, interdependent operations; and expert corresponds to multi-component coordinated, high-tolerance operations. Each level is associated with different guidance depths and presentation methods.

[0041] Step S125: Summarize the operating parameter types, fault impact ranges, and operation complexity levels to form an operation and maintenance scenario feature recognition result.

[0042] The system summarizes the operating parameter type, fault impact scope, and operational complexity level. The operating parameter types are ranked by criticality and labeled with the weight associated with the fault. The fault impact scope lists the affected components and describes the risk level. The operational complexity level is accompanied by a summary of the assessment basis. The results are stored in a standardized format, including feature identifiers, feature values, confidence scores (based on data fit and model accuracy), and data source identifiers. These serve as the core input for generating dynamic guidance strategies, ensuring that the guidance plan is accurately adapted to the scenario.

[0043] Step S130: Based on the operation and maintenance scenario feature recognition result, the dynamic guidance strategy generation module of the virtual artificial intelligence agent is called to generate an operation and maintenance guidance plan including a scenario-adaptive operation guidance sequence and real-time semantic interaction nodes.

[0044] The dynamic guidance strategy generation module receives the feature recognition results and initiates a multi-dimensional generation process. It prioritizes actions based on the scope of the fault's impact, prioritizing high-risk faults for emergency response. It also aligns the granularity and depth of step decomposition with the complexity of the operation. Monitoring logic is embedded within the operational steps based on the type of operating parameters. Reference is made to historical case studies to optimize the sequence of steps and reduce redundancy.

[0045] Real-time semantic interaction nodes are set up before and after key steps, including parameter monitoring and confirmation, operational effect feedback, and risk warning nodes, with clear feedback information types and formats. The final guidance plan includes a checklist of operational steps, interaction node configuration, supporting documentation, and risk warnings, forming a complete guidance system.

[0046] Step S131: Input the operation and maintenance scenario feature recognition result into the dynamic guidance strategy generation module of the virtual artificial intelligence agent, trigger the strategy generation rule engine in the dynamic guidance strategy generation module, and extract the basic guidance framework that matches the operation parameter type, fault impact range and operation complexity level.

[0047] The strategy generation rule engine includes multiple pre-set rule sets, each corresponding to a specific scenario feature combination. After inputting the feature recognition results, the engine uses a feature matching algorithm to select the most suitable rule combination and extract the corresponding basic guidance framework from the guidance framework library. The framework includes the overall structure of the operation process, the logical sequence of core steps, and the safety specification module. For example, a high-complexity refrigeration system operation includes the steps of safety pressure relief, component inspection, parameter calibration, and functional verification.

[0048] Step S132: Adjust the operation priority order in the basic guidance framework according to the fault impact scope, advance the operation steps corresponding to the associated components whose fault impact scope covers multiple core service components, and supplement the risk avoidance statements corresponding to the corresponding associated component operations.

[0049] Analyze the number and importance of core components affected by the fault and re-prioritize the steps in the basic guidance framework. Pre-empt operations for related components that affect multiple core components. For example, if a fault affects the compressor, condenser, and temperature control system, pre-empt refrigerant pressure testing and power supply safety isolation steps to control the spread of the fault. Supplement the pre-emptive steps with risk avoidance statements, including potential risk points (refrigerant leakage, high-voltage electric shock), preventative measures (wearing protective equipment, using specialized tools), and emergency plans (leakage ventilation, electric shock first aid). Convert these statements into concise operational instructions embedded in the step-by-step instructions.

[0050] Step S133: In combination with the operation complexity level, match the detail description depth for each operation step in the basic guidance framework, add component collaborative operation instructions for steps corresponding to detail requirements of the operation complexity level, and simplify redundant descriptions for steps corresponding to non-detail requirements of the operation complexity level to form a scenario-adaptive operation guidance sequence.

[0051] A pre-set mapping relationship is established between the level of operation complexity and the depth of description, with higher complexity corresponding to finer-grained descriptions. A description depth parameter is defined for each step to control information density and level of detail. For steps requiring high detail, component coordination instructions are added, including operation sequence constraints (open valve A before pump B), parameter linkage relationships (adjust parameter X while monitoring parameter Y), and coordination precautions (avoid operating component D while component C is running). Tool specifications and environmental control standards are also supplemented.

[0052] Low-detail steps are streamlined, removing duplicate prompts, redundant background, and irrelevant extended instructions. The operation objects, core actions, and key results indicators are retained. The adjusted steps are arranged sequentially to form a scenario-adaptive operation guide sequence.

[0053] Step S1331: extract all operation steps in the basic guidance framework, each operation step including an operation object, an operation action and an operation target.

[0054] Go through the steps in the basic guidance framework and analyze the three elements of each operation step: the operation object specifies the specific component or parameter (e.g., "compressor power switch," "refrigerant pressure value"); the operation action defines the specific execution behavior (e.g., "close," "adjust," "test"); and the operation objective describes the desired result (e.g., "cut off power," "pressure stabilizes within the normal range," "confirm no leaks"). Complete any steps that lack these three elements to ensure that the operational intent of each step is clear and explicit.

[0055] Step S1332: Associate the operation complexity level with the description depth matching rules of the virtual artificial intelligence agent to determine the detailed description dimensions corresponding to different operation complexity levels. When the operation complexity level corresponds to detailed requirements, it includes component collaboration logic, operation timing requirements, and parameter adjustment gradients. When the operation complexity level corresponds to non-detailed requirements, only the core operation actions and operation targets are retained.

[0056] Describe the mapping relationship between complexity levels and detail dimensions by defining depth matching rules: the expert level (high-detail requirements) corresponds to component coordination logic (coordination between multiple components), operation timing requirements (the time interval and sequence of step execution), and parameter adjustment gradients (the amplitude and range of each adjustment); the advanced level corresponds to some detail dimensions; the basic level (no detail requirements) retains only the core operation actions and operation objectives. Based on the current operation complexity level, determine the detail description dimensions required for each operation step.

[0057] Step S1333: For the operation steps corresponding to the detailed requirements of the operation complexity level, supplement the collaborative triggering conditions between the multiple components involved in the operation steps, the sequential order of operation execution and the gradual adjustment gradient description of key parameters to improve the details of the operation steps.

[0058] For expert-level operation steps, supplement component coordination trigger conditions (e.g., "The compressor can only be turned on after the condenser fan has started and has been running stably for 30 seconds"); clarify the sequential order of operation execution (e.g., "The first step is to close the high-pressure valve, the second step is to close the low-pressure valve, and the third step is to disconnect the main power supply"); and detail the step-by-step adjustment gradient for key parameters (e.g., "Each temperature setpoint is adjusted in 1°C increments, and after each adjustment, observe for 5 minutes before making the next adjustment"). Supplementary content should reference equipment technical manuals and expert operating experience to ensure the accuracy and operability of detailed descriptions.

[0059] Step S1334: For operation steps with non-detailed requirements corresponding to the operation complexity level, delete repeated operation target statements, redundant safety prompts, and background descriptions irrelevant to the operation actions, and simplify the operation step descriptions.

[0060] For basic-level operation steps, review and remove repeated operation objectives (e.g., multiple references to "make sure the device is powered off"), streamline redundant safety instructions (e.g., retain only the core prompt "make sure the device is powered off before operation" and remove repeated warnings about the risk of electric shock), and remove irrelevant background information (e.g., explanations of the device's operating principles). Simplified step descriptions should be concise, logically coherent, and highlight the core operation information.

[0061] Step S1335: Integrate all the supplemented, improved and simplified operation steps according to the initial sequence of the basic guidance framework to form a scenario-adapted operation guidance sequence.

[0062] Arrange the simplified and detailed steps in the original logical order of the basic guidance framework. Check the connections between the steps to ensure that the operation objectives of the previous steps provide the necessary conditions for the subsequent steps. Simulate the step sequence to verify the rationality of the operation sequence and make partial adjustments if any logical conflicts exist. The resulting scenario-adapted operation guidance sequence includes step number, operation object, operation action, operation goal, and corresponding detailed descriptions, fully presenting the operation process.

[0063] Step S134: In the scenario-adaptive operation guidance sequence, key steps involving operation parameter monitoring, component status confirmation and operation result verification are selected, and real-time semantic interaction nodes are set. Each real-time semantic interaction node contains the core information type and feedback triggering timing that require feedback from operation and maintenance personnel.

[0064] Traverse the scenario-adaptive operation guidance sequence and identify key steps: operating parameter monitoring steps (such as testing refrigerant pressure and measuring current values), component status confirmation steps (such as inspecting component appearance and verifying connection tightness), and operation result verification steps (such as evaluating trial operation results and observing stability after parameter adjustment). These steps directly affect the correctness of subsequent operations and require the establishment of real-time semantic interaction nodes.

[0065] Each real-time semantic interaction node contains clear core information type requirements, such as the operation parameter monitoring node needs to feedback the specific parameter measurement value and measurement tool model; the component status confirmation node needs to feedback the component's appearance characteristics, connection status and function indications; the operation result verification node needs to feedback the phenomenon changes, abnormal conditions and next operation suggestions after the operation.

[0066] Feedback triggering is determined by the nature of the step, including pre-step preparation confirmation (e.g., "Please provide feedback on the tool calibration status before starting measurement"), key milestones during step execution (e.g., "Please provide feedback on the current reading after the pressure stabilizes"), and post-step result feedback (e.g., "Please provide feedback on the device response after the operation is complete"). Real-time semantic interaction nodes are uniquely associated with corresponding operation steps to ensure that interaction information accurately maps to the operational process.

[0067] Step S135: Integrate the scenario-adaptive operation guidance sequence and the real-time semantic interaction node, add association identifiers between the interaction node and the operation steps, and generate an operation and maintenance guidance plan.

[0068] The scenario-adaptive operation guidance sequence and real-time semantic interaction nodes are integrated according to chronological order and logical relationships to form a unified operation and maintenance guidance solution structure. Each real-time semantic interaction node is bound to the corresponding operation step through an association identifier. The identifier includes the step number, node type, and interaction sequence number, ensuring the accurate location of the node during the guidance execution process.

[0069] The O&M guidance plan also includes metadata, such as the plan's creation time, applicable device models, plan version number, associated fault type tags, and referenced technical document numbers. This metadata facilitates plan management and traceability. Furthermore, a format conversion module is embedded within the plan, automatically adjusting the content layout based on the display characteristics of the operator's terminal, ensuring clear presentation across different devices.

[0070] The generated operation and maintenance guidance plan undergoes a compliance check to verify that the operational steps meet safety regulations and technical standards and that the interaction node settings cover all key links. Once the check passes, the plan is stored in an encrypted format and a unique solution identifier is generated for subsequent updates and tracking.

[0071] Step S140: Push the operation and maintenance guidance plan to the operation and maintenance personnel terminal, receive in real time the operation feedback description returned by the operation and maintenance personnel after performing the operation based on the operation and maintenance guidance plan, and input the operation feedback description into the feedback semantic parsing module of the virtual artificial intelligence agent.

[0072] The virtual AI agent's solution push unit sends the O&M guidance plan to the operator's mobile terminal via an encrypted transmission channel. Data fragmentation and verification mechanisms are used during transmission to ensure the plan's integrity. After receiving the O&M guidance plan, the terminal application automatically parses the plan structure and displays a scenario-adapted operational guidance sequence on the interface, chronologically displaying the location of the real-time semantic interaction node next to the corresponding step, such as the "Feedback Required" label after the "Refrigerant Pressure Detection" step.

[0073] As operators perform each operation step, the terminal application prompts based on the triggering conditions of real-time semantic interaction nodes. For example, when the step reaches the "Please provide feedback on the current reading after the pressure stabilizes" node, a feedback input box pops up on the interface, supporting multiple feedback methods such as text input, voice recording, and image upload. The operator's operation feedback description includes the actual operation execution (such as "The pressure gauge has been connected according to the steps"), the observed component status (such as "The pressure pointer is stable in a certain range"), the measured parameter value (such as "The current pressure value is a certain state"), and any abnormal phenomena encountered (such as "The pressure gauge reading continues to fluctuate").

[0074] The terminal application performs real-time preprocessing on the input operation feedback description, including format standardization (unifying professional terminology) and completeness verification (checking for the inclusion of core information required by the node). If any information is missing, the operation and maintenance personnel are prompted to supplement it. The verified operation feedback description is sent to the feedback semantic parsing module of the virtual artificial intelligence agent via an instant messaging protocol. During transmission, the feedback timestamp and corresponding node identifier are recorded to ensure that the feedback information is accurately associated with the operation step. After receiving the operation feedback description, the feedback semantic parsing module stores it in a temporary data buffer, awaiting the next step of semantic analysis.

[0075] Step S141: Through the interactive push unit of the virtual artificial intelligence agent, the operation step description and the feedback prompt of the corresponding real-time semantic interaction node are synchronously pushed to the operation and maintenance personnel terminal in accordance with the step order of the scenario-adaptive operation guidance sequence in the operation and maintenance guidance plan.

[0076] The interactive push unit generates a push queue based on the sequential order of the O&M guidance steps, pushing content sequentially according to the O&M personnel's progress. Each step description includes the action object, core action, precautions, and reference diagrams. For example, when pushing the "turn off the compressor power" step, a diagram of the power switch location and the core action of "confirming the switch is off" are included.

[0077] When pushing to a step containing a real-time semantic interaction node, the interactive push unit simultaneously sends a feedback prompt, clearly informing the operator of the type of information required and the specific requirements, such as "Please provide feedback on the current refrigerant pressure measurement value and the model of the measuring tool." This prompt appears in a prominent pop-up window, including a deadline (set based on the complexity of the operation, such as within 5 minutes) and a sample format (such as "Pressure value: certain state, tool: digital pressure gauge"), guiding operators in providing standardized feedback.

[0078] The push unit monitors the terminal's reception status in real time. If it doesn't receive a confirmation from the terminal, it will re-push after 30 seconds to ensure that the operation and maintenance personnel can obtain the operation instructions in a timely manner. At the same time, the push unit records the push time of each step, the terminal's reception time, and the operation and maintenance personnel's viewing status, forming a push track log.

[0079] Step S142: When each real-time semantic interaction node is triggered, the feedback receiving channel of the virtual artificial intelligence agent is started to receive the operation execution status, component response phenomenon and operation question statement input by the operation and maintenance personnel through the terminal to form an operation feedback description.

[0080] Once a real-time semantic interaction node is triggered, the feedback receiving channel is automatically activated, establishing a two-way communication link between the terminal and the virtual AI agent. Operations personnel use the terminal to input the status of the operation, including the completion status of the step (e.g., "Completed," "In Progress," or "Not Executed"), the tool model used during the execution (e.g., "Multimeter model is X," "Pressure gauge range is X"), and the operation duration (e.g., "It took 3 minutes from connecting the tool to obtaining a reading").

[0081] Feedback on component responses includes detailed descriptions of changes in the component's physical state (e.g., "no leakage sound after valve closure," "smooth operation after fan startup"), indicator light status (e.g., "power indicator is always on," "fault indicator is flashing"), and abnormal behavior (e.g., "slight frost on pipe connections," "abnormal motor noise"). Operational questions are provided by O&M personnel, who may have encountered confusion during the execution of a procedure, such as "does pressure need to be relieved immediately if the pressure falls below the normal range?" and "what should I do if the tool display value doesn't match the standard unit?"

[0082] The feedback receiving channel supports multiple rounds of input. Operations personnel can provide multiple rounds of feedback on the same node. The terminal application automatically consolidates these multiple entries into a complete operational feedback description, timestamping each addition. After completing the feedback input, the operator clicks "Submit." The terminal application encrypts the operational feedback description and sends it to the feedback receiving channel. The channel then confirms receipt and returns a successful receipt to the terminal.

[0083] Step S143: The operation feedback description is transmitted in real time, so that the operation feedback description is immediately sent to the feedback semantic parsing module of the virtual artificial intelligence agent after the operation and maintenance personnel completes the input.

[0084] After the operation feedback description is submitted, the terminal application encapsulates the data, adding metadata such as the terminal device identifier, the operator's identity, the corresponding real-time semantic interaction node identifier, and the feedback timestamp to form a complete feedback data packet. The data packet is encrypted using a symmetric encryption algorithm and sent to the virtual artificial intelligence agent's feedback receiving server via a dedicated communication port.

[0085] A breakpoint-resume mechanism is used during transmission. If a network interruption causes a transmission failure, the terminal application automatically resends the unsuccessfully transmitted data packet after the network is restored. Upon receiving the data packet, the feedback receiving server first decrypts and verifies its integrity. Once the integrity check passes, the packet is forwarded to the feedback semantic parsing module and a confirmation message is returned to the terminal, including the packet reception time and verification results.

[0086] After receiving the data packet, the feedback semantic parsing module extracts the operation feedback description and associated metadata, stores them in the parsing buffer, and simultaneously logs the data received, including information such as the reception time, data size, and source identifier, to ensure the traceability of the feedback data. The parsing module queues the operation feedback descriptions in the order they were received, awaiting semantic segmentation.

[0087] Step S144: After receiving the operation feedback description, the feedback semantic analysis module starts semantic word segmentation processing to separate factual expressions and interrogative expressions in the operation feedback description, and mark them as factual feedback segments and interrogative feedback segments respectively.

[0088] Upon receiving the feedback description, the feedback semantic parsing module immediately activates its built-in semantic processing engine, which includes multiple functional units such as word segmentation, matching, and verification, working together to complete the semantic parsing. The original text is first normalized to remove interfering information. A professional vocabulary matching tool is then used to distinguish between factual and interrogative statements. Finally, the segmentation results are verified for integrity, ensuring that subsequent analysis is based on accurate feedback information.

[0089] Step S1441: calling the semantic word segmentation model of the virtual artificial intelligence agent to perform lexical segmentation on the operation feedback description to obtain multiple semantic word units.

[0090] The semantic word segmentation model is loaded with a special word segmentation dictionary for the operation and maintenance field. This dictionary covers core field vocabulary such as equipment component terminology, operation action vocabulary, parameter type vocabulary, and fault phenomenon vocabulary. The operation feedback description is scanned sentence by sentence, and a two-way maximum matching algorithm is used for word segmentation, matching the longest word sequence from the beginning and end of the text at the same time. For example, "the pressure of the compressor continues to drop and is accompanied by abnormal noise when it is running" is segmented into multiple semantic word units such as "compressor / operation / pressure / continues to drop / and / accompanied by / abnormal noise". Each word unit is assigned a part-of-speech tag (such as "noun", "verb", "adjective", "fault word"), and its starting and ending positions in the original text are recorded, which facilitates the subsequent tracing of the source of the word.

[0091] Step S1442: Match the semantic vocabulary unit with the factual expression vocabulary of the virtual artificial intelligence agent, filter out vocabulary units containing factual information such as operation results, component status, and parameter values, and combine them to form a factual feedback segment.

[0092] The fact description vocabulary is stored in a hierarchical structure. The base layer contains parameter value vocabulary and unit vocabulary, the middle layer contains status description vocabulary and component status vocabulary, and the high layer covers operation result vocabulary. The matching engine calculates the similarity between each semantic vocabulary unit and the fact description vocabulary, uses the edit distance algorithm to quantify the vocabulary difference, and selects vocabulary units with a matching degree higher than the preset threshold as fact candidate vocabulary. Through dependency syntactic analysis, the semantic association between words, such as subject-predicate relationship and verb-object relationship, is identified. "Pressure / continues to drop" is combined into the fact feedback segment of "pressure continues to drop", and "compressor / operation is normal" is combined into the fact feedback segment of "compressor operation is normal". Each fact feedback segment is labeled with the corresponding information type, such as parameter change class, component status class, operation result class, etc.

[0093] Step S1443: Match the remaining semantic vocabulary units with the question expression vocabulary of the virtual artificial intelligence agent, filter out vocabulary units containing question particles, uncertain expressions, and operation confusion descriptions, and combine them to form a question feedback segment.

[0094] The remaining semantic vocabulary units that have not been matched to the factual expression vocabulary enter the question expression vocabulary matching process. The question expression vocabulary is stored by question type, including a subset of question particles, a subset of uncertain expressions, and a subset of help-seeking types. The correlation between the vocabulary and the question expression vocabulary is determined through keyword matching and semantic vector calculation, and vocabulary units containing question particles such as "how", "whether", and "why", uncertain expressions such as "maybe", "suspected", and descriptions of operational confusion such as "I don't know how to operate" and "request guidance" are screened out. The semantic role labeling technology is used to extract the core elements of the question, and the question feedback segment "How to deal with the continuous decrease in pressure" is combined into "Consultation on the treatment method of continuous decrease in pressure", clarifying the operational scenario and core demands pointed to by the question.

[0095] Step S1444: perform integrity check on the factual feedback segment to check whether it contains the core result information corresponding to the operation steps. If missing, generate a supplementary query prompt and push it to the operation and maintenance personnel terminal. After receiving the supplementary feedback, complete the factual feedback segment.

[0096] The integrity judgment unit checks whether the fact feedback segment contains the necessary key elements based on the core information requirements corresponding to the real-time semantic interaction node. For example, the fact feedback segment of the operating parameter monitoring node must include information such as parameter name, measurement value, and measurement tool; the component status confirmation node must include information such as component name, appearance characteristics, and function indications. If the inspection finds that the fact feedback segment is missing core information, such as only feedback of "abnormal pressure" without specifying the specific value, a standardized supplementary inquiry prompt is generated, such as "Please add the specific measurement value of the current refrigerant pressure", and sent to the operation and maintenance personnel terminal through the interactive push unit. After receiving the supplementary feedback from the operation and maintenance personnel, the newly added information will be integrated into the original fact feedback segment to ensure the completeness of the segment information.

[0097] Step S1445: extract the core questions of the query feedback segment, determine the specific confusion points of the operation and maintenance personnel regarding the operation steps, form a core identifier of the query feedback segment, and associate it with the corresponding real-time semantic interaction node identifier.

[0098] The core question extraction unit uses semantic understanding technology to parse query feedback segments, identifying the query type and key concerns. Through keyword extraction and syntactic analysis, the core confusion point from the sentence "After adjusting the expansion valve, the pressure is still unstable, and I don't know whether to continue adjusting it" is extracted as "Subsequent operation judgment for unstable pressure after expansion valve adjustment." A core identifier is generated for each query feedback segment, including the query type (such as operation method, exception judgment, parameter standard), the operation object involved, and a description of the core confusion. The core identifier is bound to the corresponding real-time semantic interaction node identifier to ensure that the query feedback is accurately linked to the specific operation steps and interaction nodes.

[0099] Step S145: Associating the fact feedback segment with the question feedback segment with corresponding real-time semantic interaction node identifiers to form initial input data for feedback semantic parsing.

[0100] The feedback semantic parsing module extracts the real-time semantic interaction node identifiers associated with the operational feedback description and binds them to the factual feedback segment and the question feedback segment, ensuring that each feedback segment accurately corresponds to a specific node location in the operation and maintenance guidance plan. During the binding process, information such as the segment number, segment type, and the step number of the associated node are added to form structured linked data.

[0101] Initial input data is organized as data frames. Each data frame contains a list of factual feedback segments, a list of question feedback segments, a node ID, a feedback timestamp, and an operator ID. The module performs integrity checks on data frames to ensure that all required fields are present. If any fields are missing, default values ​​are automatically added or the data is marked as abnormal.

[0102] The initial input data that passes the check is stored in the parsing result database, and a unique parsing task identifier is generated for tracking the subsequent parsing process. The module also pushes the initial input data to the feedback analysis unit, triggering a comparison analysis of the factual feedback segment with the expected results and extracting the core questions from the question feedback segment.

[0103] Step S150: Associating the real-time semantic interaction nodes in the operation and maintenance guidance plan through the feedback semantic parsing module, adjusting the step content and presentation order of the scenario-adaptive operation guidance sequence, obtaining an updated operation and maintenance guidance plan, pushing the updated operation and maintenance guidance plan to the operation and maintenance personnel terminal and storing the guidance plan evolution trajectory and operation feedback description during this interaction process.

[0104] The feedback semantic parsing module associates and matches the factual and question feedback segments in the initial input data with the corresponding real-time semantic interaction nodes, locating the corresponding step in the scenario-adaptive operation guidance sequence. The analysis unit accesses the expected result database, extracts the expected result description for the step, and compares it with the factual feedback segment to determine whether the operation execution result meets expectations.

[0105] If the factual feedback segment is consistent with the expected result, the subsequent steps remain unchanged. If there are inconsistencies (such as parameter anomalies or the operation did not achieve the target), the abnormal result description is extracted and the abnormal response strategy library is used to generate operational adjustment suggestions. For the core confusion points in the question feedback segment, the question answering knowledge base is used to match the corresponding answers and incorporate them into the adjusted step description.

[0106] Based on the operational adjustment suggestions and answers, the module revises the corresponding steps in the scenario-adaptive operational guidance sequence, such as adding exception handling substeps and adjusting parameter monitoring standards. It also reorders subsequent steps based on the scope of the exception's impact. Once the revisions are complete, an updated operational guidance plan is generated and sent to the operator's terminal via a push unit. The terminal also displays the updated plan instructions.

[0107] The module also records the initial guidance plan, updated guidance plan, the rationale for each adjustment, and the corresponding operational feedback during the interaction, forming a chronological evolution trajectory of the guidance plan. This evolution trajectory is associated with all operational feedback and stored in the interaction archive, along with metadata such as the operation and maintenance object identifier and the start and end times of the interaction.

[0108] Step S151: extracting the fact feedback segment, the question feedback segment and the associated real-time semantic interaction node identifier corresponding to the operation feedback description through the feedback semantic parsing module.

[0109] The extraction unit of the feedback semantic parsing module reads the factual feedback segments, question feedback segments, and associated real-time semantic interaction node identifiers from the initial input data and performs a secondary verification of the segment content to ensure the integrity and semantic coherence of the segments. If the verification process finds any semantic gaps or missing key information in the segment, it is automatically marked as a low-confidence segment and the type of missing information (such as missing parameter values ​​or incomplete status description) is recorded.

[0110] The extraction unit categorizes high-confidence factual feedback segments by information type, such as parameter measurement, state observation, and operation execution. Each segment type is associated with a corresponding feature tag (e.g., "pressure parameter," "leakage status," and "switch operation"). Question feedback segments are categorized by question type, such as operation method, anomaly determination, and tool usage, to facilitate subsequent question answering.

[0111] The extraction unit rebinds the classified segments with the real-time semantic interaction node identifiers to generate a segment-node association table containing information such as segment number, node identifier, segment type, feature label, and confidence score. This segment-node association table serves as the basis for subsequent positioning and content adjustment and is stored in a temporary analysis database.

[0112] Step S152: Match the real-time semantic interaction node identifier with the real-time semantic interaction node in the operation and maintenance guidance plan, and locate the target operation step in the scenario-adaptive operation guidance sequence corresponding to the real-time semantic interaction node.

[0113] The matching unit receives the real-time semantic interaction node identifier from the fragment-node association table, traverses the real-time semantic interaction node list in the operation and maintenance guidance plan, and finds the corresponding node record through identifier comparison. The node record contains information such as the node number, the corresponding step number, the node type, and the associated operation target. The matching unit locates the specific step in the scenario-adaptive operation guidance sequence based on the node's corresponding step number, i.e., the target operation step.

[0114] After location is complete, the matching unit extracts detailed information about the target operation step, including the step number, operation object, action, expected result description, and associated operating parameter types, to form a target step information table. If a node identifier fails to match (e.g., due to an identifier error or the node has been deleted), a matching exception log is generated, recording the error identifier and the reason for the match failure. This log also triggers a manual intervention prompt to ensure accurate step location.

[0115] After the target step information table is generated, it is sent to the analysis unit as benchmark data for comparing the fact feedback segment with the expected result. At the same time, it is linked to the answer matching process of the question feedback segment to ensure that the adjustment and answer content accurately corresponds to the target step.

[0116] Step S153: Analyze whether the factual feedback segment is consistent with the expected result description of the target operation step. If not, extract the abnormal result description in the factual feedback segment, call the abnormal response strategy library of the virtual artificial intelligence agent, and generate operation adjustment suggestions for the abnormal result description.

[0117] In this step, the expected result description of the target operation step must be accurately located, and this will be used as a benchmark for comparing the consistency of the factual feedback fragments. The expected result description usually includes multi-dimensional verification indicators, such as the parameter range of the component operating status, the physical phenomenon characteristics after the operation is executed, and the hallmark performance of functional recovery. Taking the refrigerant pressure adjustment step of the chain store smart air conditioner as an example, the expected result description may be "After adjustment, the low-pressure side pressure is stabilized within the standard range, the pressure gauge pointer fluctuation amplitude does not exceed the allowable range, and the air conditioner operating noise is reduced to the normal decibel range."

[0118] The specific statements in the actual feedback segment are compared point by point with the description of the expected results. The comparison dimensions include parameter value matching, phenomenon characteristic consistency, and time series stability. If the actual feedback segment contains content that does not match expectations, such as "pressure continues to drop below the lower limit of the standard range after adjustment" or "indicator fluctuates violently for more than the specified period", the results are considered inconsistent and the exception handling process is triggered.

[0119] Step S1531: extract the expected result description of the target operation step in the operation and maintenance guidance plan, and determine the component status parameters, operating parameter ranges and operation completion identifiers included in the expected result description.

[0120] The core elements of the expected results are separated from the target operation steps description of the operation and maintenance guidance plan. Component status parameters include the physical state of the component (such as "no signs of leakage after the valve is closed"), the indicator light display status (such as "the operation indicator light is always on without flickering") and the mechanical action response (such as "the fan rotates smoothly without jamming after starting"). The operating parameter range clearly defines the normal range of key indicators, such as "the difference between the return air temperature and the set temperature does not exceed 2 degrees Celsius" and "the compressor operating current is maintained within the rated range." The operation completion mark is a landmark end signal, such as "the buzzer sounds a prompt sound" and "the display returns to the normal working interface" and other directly observable results. These elements are stored through structured fields to form a checklist for expected results.

[0121] Step S1532: compare the operation execution status and component response phenomenon in the fact feedback segment with the expected result description point by point, mark the inconsistent difference items, and determine the abnormal result description.

[0122] Perform structured analysis of factual feedback fragments to extract operation execution status (e.g., "Three adjustments have been completed according to the steps"), component response phenomena (e.g., "Pressure continues to drop" and "Abnormal vibration occurs"), and related parameter data (e.g., "Current pressure value, fluctuation frequency"). Perform point-by-point comparisons according to the order of the expected result checklist, for example, comparing the actual pressure value with the standard range and the noise description with normal characteristics. Items with deviations are marked as difference items, such as "Pressure value is 20% lower than the lower limit" and "Continuous fluctuation duration is three times longer than expected." The difference items are summarized to form a description of the abnormal result, ensuring that each abnormal point is supported by a specific description.

[0123] Step S1533: Send a query request to the abnormal response strategy library of the virtual artificial intelligence agent, so as to form an initial response suggestion based on the historical response strategy corresponding to the abnormal result description and the target operation step identifier matched with the query request, and the operating parameter type supplementary parameter monitoring requirements combined with the operating parameter type. The query request includes the abnormal result description, the target operation step identifier and the operating parameter type in the operation and maintenance scenario feature recognition result.

[0124] The abnormal response strategy library adopts a three-dimensional index structure of "phenomenon-step-strategy", and the query request must carry complete search elements. The abnormal result description is used as the core search term, such as "the refrigerant pressure continues to drop"; the target operation step identifier clearly defines the operation type, such as "expansion valve adjustment step"; the operating parameter type limits the associated monitoring indicators, such as "low-pressure side pressure, compressor current". The system matches the response strategies for the same or similar scenarios in historical cases in the library, such as "gradually close the stop valve to observe the pressure change" and "check the valve port sealing status" and other operational suggestions. At the same time, targeted monitoring requirements are supplemented according to the operating parameter type, such as "record the pressure value every 1 / 4 turn of adjustment" and "synchronously monitor the compressor exhaust temperature" to form an initial response suggestion that includes operating actions, monitoring nodes, and judgment conditions.

[0125] Step S1534: Perform scenario adaptability analysis on the initial response suggestion to ensure that the operation adjustment content in the initial response suggestion conforms to the fault impact scope and operation complexity level of the current operation and maintenance scenario, and generate a final operation adjustment suggestion.

[0126] The scenario adaptability analysis is carried out from two dimensions: the scope of fault impact and the complexity of operation. Regarding the scope of fault impact, if the abnormality may affect multiple core components (such as pressure abnormalities that may affect the compressor and condenser), the initial recommendations will strengthen the safety isolation operation, such as "turn off the main power first and then check the components"; if the scope of impact is limited to a single component, then focus on precise adjustment recommendations. Combined with the level of operational complexity, detailed tool specifications and coordination requirements are added to the expert-level operation steps, such as "use a special torque wrench to tighten to the specified torque"; the basic-level steps are simplified and the core actions are retained. During the analysis process, general strategies that do not match the current scenario are eliminated, and scenario-specific constraints are added, such as "avoid long-term downtime during store business hours and adopt a segmented adjustment method", and finally form operational adjustment recommendations that adapt to the current scenario.

[0127] Step S154: For the core confusing points in the question feedback segment, call the question answering knowledge base of the virtual artificial intelligence agent, match the answer statements related to the target operation steps, and integrate them into the adjusted operation step description.

[0128] The question parsing unit extracts core confusion points from the query feedback fragments and uses semantic understanding technology to identify the core demand of the question. For example, the core confusion point of "Is pressure relief necessary immediately when pressure falls below the normal range?" is "Determining when to address abnormal pressure." After extracting the core confusion points, it generates a query request containing the target operation step identifier, confusion point keywords, and operation and maintenance scenario characteristics, and sends it to the question answering knowledge base.

[0129] The Q&A knowledge base utilizes a multi-level index structure, categorizing and storing answers by operation step type, device component, fault type, and other dimensions. Upon receiving a query, the knowledge base uses keyword matching and scenario similarity calculations to retrieve answers that are highly relevant to the core confusion point, such as "When the pressure falls below a certain lower limit and continues to fluctuate, close the intake valve first and then perform pressure relief operations after the pressure stabilizes."

[0130] The retrieved answers are verified for scenario suitability, ensuring they match the equipment model, operational complexity, and fault impact scope of the current O&M scenario. Once verified, the answers are formatted as natural language instructions and added to the adjusted content of the target operation steps. For example, after the "Check refrigerant pressure" step, the following supplementary instruction is added: "If the pressure is below the normal range, the recommended action is: First close the air inlet valve..."

[0131] Step S155: According to the operation adjustment suggestions and answer statements, the content of the target operation steps in the scenario-adaptive operation guidance sequence is modified, and the presentation order of subsequent operation steps is adjusted based on the impact scope corresponding to the abnormal result statement to form an updated operation and maintenance guidance plan.

[0132] The solution revision unit receives operation adjustment suggestions and answers, and first revises the content of the target operation steps in the scenario-adaptive operation guidance sequence, such as updating the operation action description (revising "record pressure value" to "record pressure value, and perform the following operations if it is lower than the normal range"), adding abnormal handling sub-steps (such as "close the air intake valve" and "wait for the pressure to stabilize") and additional precautions (such as "open the valve slowly when releasing pressure to avoid a sudden drop in pressure").

[0133] At the same time, the solution revision unit analyzes the impact scope of the abnormal result statement and assesses its impact on the execution conditions and risk level of subsequent steps. If the abnormality's impact is small (e.g., only the parameters of the current step need to be adjusted), the order of subsequent steps remains unchanged. If the abnormality may affect multiple subsequent steps (e.g., requiring component replacement before continuing), the subsequent steps are reordered, with the relevant component inspection and replacement steps being placed earlier to ensure the rationality and safety of the operational process.

[0134] Once the revision is complete, the solution revision unit updates the step numbers and associated identifiers in the scenario-adaptive operation guidance sequence and regenerates the location information of the real-time semantic interaction nodes to ensure that the nodes accurately correspond to the adjusted steps. The resulting updated operation and maintenance guidance solution undergoes integrity and compliance checks, is marked as the official version, and is ready to be pushed to the operation and maintenance personnel's terminals.

[0135] Step S156: Through the solution push unit of the virtual artificial intelligence agent, the updated operation and maintenance guidance plan is pushed to the operation and maintenance personnel terminal in the order of the adjusted scenario-adaptive operation guidance sequence, and the solution update logo and adjustment instructions are displayed on the terminal interface of the operation and maintenance personnel terminal.

[0136] The solution push unit receives the updated O&M guidance plan and generates a solution update notification, including the update time, adjustment reason (e.g., "Adjustment based on pressure anomaly feedback"), and a summary of the main adjustments (e.g., "Addition of pressure anomaly handling steps"). This notification, along with the updated guidance plan, is pushed to the O&M personnel's terminal via an encrypted channel. This push utilizes incremental transmission technology, sending only the differences from the previous version to reduce data transmission.

[0137] After receiving the update, the operator's terminal displays the updated plan (e.g., a red "Update" label) at the top of the interface, and a pop-up window displays the reason for the update and the main changes. The terminal application automatically replaces the original guidance plan content and displays the scenario-adapted operation instructions in the adjusted order. New or modified steps are highlighted (e.g., with a yellow background) and labeled "New" or "Modified" next to them.

[0138] The terminal application records the plan update log, including information such as update time, version number, adjustment content summary and reception status. It also supports the operation and maintenance personnel to view the historical version of the plan. The version switching button can be used to compare the differences in steps between different versions to ensure that the operation and maintenance personnel understand the specific content of the plan adjustment.

[0139] Step S157: receiving confirmation information of the operation and maintenance personnel on the updated operation and maintenance guidance plan. If the plan is confirmed to be executed, the generation time, adjustment basis and version identifier of the updated operation and maintenance guidance plan are recorded.

[0140] While displaying the updated O&M guidance plan, the terminal interface provides two buttons: "Confirm Execution" and "Feedback." After reviewing the plan adjustments, the O&M personnel click the "Confirm Execution" button to indicate their approval of the adjusted plan. The terminal then converts this confirmation into structured confirmation information, including the O&M personnel's ID, confirmation timestamp, and plan version ID. This information is then transmitted via an encrypted channel to the interaction recording module of the virtual AI agent. Upon receiving the confirmation information, the interaction recording module immediately associates the metadata of the updated O&M guidance plan and records the current system time in the generation time field, accurate to the millisecond level, to ensure accurate time tracking.

[0141] The Adjustment Basis field details the operational feedback description snippet underlying the solution adjustment, the corresponding real-time semantic interaction node identifier, and the anomaly analysis conclusion. For example, "Based on the 'persistently low refrigerant pressure' feedback from step S143, combined with the pressure monitoring requirements of node N3, the refrigerant leak response strategy numbered STR-2023-045 in the anomaly response strategy library is adjusted." Version identifiers are generated using the format "base version number. number of adjustments." For example, if the initial solution is V1.0, it will become V1.1 after the first adjustment, and so on. Each version identifier uniquely corresponds to a solution update.

[0142] This information is stored in association with the complete content of the updated operation and maintenance guidance plan, forming a complete record of that version of the plan. If the operator clicks the "Feedback Question" button, the terminal will jump to the question input interface, where the operator can enter specific questions about the updated plan. The virtual artificial intelligence agent will restart the feedback semantic parsing process and further optimize the plan based on the question content.

[0143] Step S158: Summarize the initial operation and maintenance guidance plan and the operation and maintenance guidance plans after each update in this interaction process, arrange them in chronological order of generation, mark the real-time semantic interaction nodes and operation feedback descriptions corresponding to each adjustment, and form an evolution trajectory of the guidance plan.

[0144] Retrieve all solution data generated by this interaction from the solution storage module of the virtual AI agent, including the initial operation and maintenance guidance solution and the updated operation and maintenance guidance solutions. Add a timestamp to each solution and create a linear sequence of solutions in chronological order of generation.

[0145] For each updated O&M guidance plan in the sequence, associate the corresponding adjustment trigger information: note the real-time semantic interaction node number that triggered the adjustment, the operation step location of the node, and the full description of the operation feedback received by the node. Arrows connect adjacent plans in the plan sequence, and the adjustment type (such as "Parameter Anomaly Adjustment," "Question and Answer Adjustment," or "Step Optimization Adjustment") is labeled next to the arrows to clearly demonstrate the cause-and-effect relationship of the plan evolution.

[0146] The evolution trajectory of each solution also includes a table comparing the core differences between each solution. This table lists the specific changes in the content of the operation steps, the order of steps, the depth of detailed descriptions, and the settings of interactive nodes, making it easier to trace and adjust the logic. The trajectory data is stored in a visual format, allowing you to view the status of each solution at each stage by sliding the timeline.

[0147] Step S159: Associating the guidance plan evolution trajectory with all operation feedback descriptions, adding the operation and maintenance object identifier, operation and maintenance personnel identifier and interaction start and end time of this interaction, to form a complete interaction record data set.

[0148] Establish an association index between the guidance solution evolution trajectory and the operation feedback description, and accurately bind each operation feedback description to the corresponding solution adjustment node in the trajectory through real-time semantic interaction node identification to ensure that the correspondence between feedback content and solution adjustment is traceable.

[0149] Add basic information fields to the interaction record dataset: Operation and maintenance objects are identified using the device's unique hardware code, operation and maintenance personnel are identified by their work number, and interaction start and end times are accurate to the second (including the time the request was initiated and the time the final solution was confirmed). Key time points during the interaction process are also recorded, such as the time the first solution was pushed, the time the first feedback was received, and the time each solution was updated, forming a complete timeline.

[0150] The dataset also includes device information (such as device model and operating system version) and network status records (such as network latency and signal strength at each stage of data transmission). All data is organized in a structured format to ensure complete fields and clear logical connections.

[0151] Step S1510: Storing the interaction record data set in the interaction archive of the virtual artificial intelligence agent.

[0152] After the interaction record dataset is generated, it undergoes a data verification process to check field integrity, relationship accuracy, and data format compliance. Any missing or incorrect data is returned for correction. Once verified, the dataset is encrypted using an encryption algorithm to ensure security during data transmission and storage.

[0153] The interaction archive for the virtual AI agent stores data in a two-level directory structure: "Operation and Maintenance Object - Interaction Time." Multiple interactions with the same operation and maintenance object are grouped into the same primary directory. Within this directory, subdirectories are created based on interaction timestamps to store corresponding datasets. The archive supports multi-dimensional searches based on operation and maintenance object ID, operation and maintenance personnel ID, interaction time range, and fault type tags, facilitating subsequent data statistics and case analysis.

[0154] After storage is complete, the archive automatically generates a data backup, which is stored off-site to prevent data loss. Lifecycle management tags are also added to datasets, with retention periods set based on data importance. Non-core data exceeding the retention period is automatically archived to a historical database, ensuring efficient operation of the archive.

[0155] Figure 2An interactive system 100 for operation and maintenance technical services combined with the assistance of an intelligent agent is shown in an embodiment of the present application, including: a processor 1001 and a memory 1003. The processor 1001 and the memory 1003 are connected, such as through a bus 1002. Optionally, the interactive system 100 for operation and maintenance technical services combined with the assistance of an intelligent agent may further include a transceiver 1004, which may be used for data interaction between the interactive system 100 for operation and maintenance technical services combined with the assistance of an intelligent agent and other interactive systems for operation and maintenance technical services combined with the assistance of an intelligent agent, such as sending data and / or receiving data. It should be noted that in actual scheduling, the transceiver 1004 is not limited to one, and the structure of the interactive system 100 for operation and maintenance technical services combined with the assistance of an intelligent agent does not constitute a limitation on the embodiment of the present application.

[0156] The memory 1003 is used to store program codes for executing the embodiments of the present application, and the execution is controlled by the processor 1001. The processor 1001 is used to execute the program codes stored in the memory 1003 to implement the steps shown in the above method embodiments.

[0157] An embodiment of the present application provides a computer-readable storage medium having program code stored thereon. When the program code is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0158] It should be understood that, although each operation step is indicated by arrows in the flow chart of the embodiment of the present application, the order of implementation of these steps is not limited to the order indicated by the arrows. Unless otherwise clearly stated herein, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flow chart can be performed in other orders based on demand. In addition, some or all of the steps in each flow chart can include multiple sub-steps or multiple stages according to actual implementation scenario, and some or all of these sub-steps or stages can be executed at the same time, and each sub-step or stage in these sub-steps or stages can also be executed at different times respectively. Under different scenarios at the execution time, the order of execution of these sub-steps or stages can be flexibly configured based on demand, and the embodiment of the present application does not limit this.

[0159] The above is only an optional implementation method for some implementation scenarios of this application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the solution of this application, other similar implementation methods based on the technical ideas of this application also fall within the protection scope of the embodiments of this application.

Claims

1. A remote guidance and interactive method for operation and maintenance technical services combined with intelligent agent assistance, characterized in that: The method comprises: Receive a remote guidance interaction request for operation and maintenance technical services initiated by an operation and maintenance personnel, wherein the remote guidance interaction request includes operation and maintenance object information, operation and maintenance problem description, and desired guidance direction; The virtual artificial intelligence agent identifies the operation and maintenance scenario characteristics of the remote guidance interaction request, extracts the operation parameter type associated with the operation and maintenance object, the fault impact range corresponding to the operation and maintenance problem, and the operation complexity level matching the expected guidance direction; Based on the operation and maintenance scenario feature recognition results, a dynamic guidance strategy generation module of the virtual artificial intelligence agent is called to generate an operation and maintenance guidance plan including a scenario-adaptive operation guidance sequence and real-time semantic interaction nodes; Pushing the operation and maintenance guidance plan to the operation and maintenance personnel terminal, receiving in real time the operation feedback description returned by the operation and maintenance personnel after performing the operation based on the operation and maintenance guidance plan, and inputting the operation feedback description into the feedback semantic parsing module of the virtual artificial intelligence agent; The real-time semantic interaction nodes in the operation and maintenance guidance plan are associated with the feedback semantic parsing module, the step content and presentation order of the scenario-adaptive operation guidance sequence are adjusted, and an updated operation and maintenance guidance plan is obtained. The updated operation and maintenance guidance plan is pushed to the operation and maintenance personnel terminal and the guidance plan evolution trajectory and operation feedback description during this interaction process are stored.

2. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 1 is characterized in that: The virtual artificial intelligence agent performs operation and maintenance scenario feature recognition on the remote guidance interaction request, extracts the operation parameter type associated with the operation and maintenance object, the fault impact range corresponding to the operation and maintenance problem, and the operation complexity level matching the expected guidance direction, including: Input the remote guidance interaction request into the scene feature extraction unit of the virtual artificial intelligence agent, and separate the operation and maintenance object field, the operation and maintenance object core component type, and the problem triggering conditions and problem-related components in the operation and maintenance problem description from the operation and maintenance object information; Call the operation and maintenance field parameter library associated with the virtual artificial intelligence agent, and match the operation parameter type that needs to be monitored by the corresponding operation and maintenance object under normal operating conditions according to the field to which the operation and maintenance object belongs and the core component type of the operation and maintenance object; Based on the problem triggering conditions and problem-related components in the operation and maintenance problem description, combined with the fault impact analysis logic of the virtual artificial intelligence agent, the scope of adjacent components that may be affected by the operation and maintenance problem and the risk level of service interruption are determined to form the fault impact scope corresponding to the operation and maintenance problem; Analyze the operation goal expression in the desired guidance direction, combine it with the operation complexity assessment model of the virtual artificial intelligence agent, and determine the operation complexity level that matches the desired guidance direction based on the number of components involved in the operation, the relevance of the operation steps, and the operation fault tolerance requirements; The operating parameter types, fault impact range and operation complexity level are summarized to form the operation and maintenance scenario feature recognition results.

3. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 2 is characterized in that: Based on the problem triggering conditions and problem-related components in the operation and maintenance problem description, combined with the fault impact analysis logic of the virtual artificial intelligence agent, the scope of adjacent components that may be affected by the operation and maintenance problem and the risk level of service interruption are determined to form the fault impact scope corresponding to the operation and maintenance problem, including: Inputting the problem triggering conditions and problem-related components in the operation and maintenance problem description into the fault impact analysis unit of the virtual artificial intelligence agent; Call the component association map of the virtual artificial intelligence agent, locate the connection relationship of the problem-related components in the operation and maintenance system based on the problem-related components, and identify the directly connected first-level adjacent components and the indirectly related second-level adjacent components; Analyze the impact of the problem triggering conditions on the functions of the problem-related components, determine whether the functional impact will be transmitted to the first-level and second-level adjacent components through the connection relationship, and determine the range of adjacent components that may be affected; Based on the range of potentially affected adjacent components and the service weight of each component in the operation and maintenance system, the likelihood of service interruption and the number of affected users are calculated to determine the service interruption risk level. Combine the range of potentially affected adjacent components with the service interruption risk level to form the fault impact range corresponding to the operation and maintenance issue.

4. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 1 is characterized in that: Based on the operation and maintenance scenario feature recognition result, the dynamic guidance strategy generation module of the virtual artificial intelligence agent is called to generate an operation and maintenance guidance plan including a scenario-adaptive operation guidance sequence and a real-time semantic interaction node, including: Inputting the operation and maintenance scenario feature recognition result into the dynamic guidance strategy generation module of the virtual artificial intelligence agent, triggering the strategy generation rule engine in the dynamic guidance strategy generation module to extract a basic guidance framework that matches the operation parameter type, fault impact range, and operation complexity level; Adjust the operation priority order in the basic guidance framework according to the scope of the fault impact, put the operation steps corresponding to the associated components whose fault impact covers multiple core service components in the front, and supplement the risk avoidance statements corresponding to the operations of the corresponding associated components; In combination with the operation complexity level, the depth of detailed description is matched for each operation step in the basic guidance framework. For steps corresponding to the detailed requirements of the operation complexity level, component collaborative operation instructions are added. For steps corresponding to the non-detailed requirements of the operation complexity level, redundant descriptions are simplified to form a scenario-adaptive operation guidance sequence. In the scenario-adaptive operation guidance sequence, key steps involving operation parameter monitoring, component status confirmation, and operation result verification are selected, and real-time semantic interaction nodes are set. Each real-time semantic interaction node contains the core information type and feedback triggering timing required by the operation and maintenance personnel; Integrate the scenario-adaptive operation guidance sequence with the real-time semantic interaction node, add association identifiers between the interaction node and the operation steps, and generate an operation and maintenance guidance plan.

5. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 4 is characterized in that: In combination with the operation complexity level, the depth of detailed description is matched for each operation step in the basic guidance framework. Component collaborative operation instructions are added to the steps corresponding to the detailed requirements of the operation complexity level. Redundant descriptions are simplified for the steps corresponding to the non-detailed requirements of the operation complexity level, forming a scenario-adaptive operation guidance sequence, including: Extract all the operation steps in the basic guidance framework, each operation step including an operation object, an operation action, and an operation target; Associating the operation complexity level with the description depth matching rules of the virtual artificial intelligence agent to determine the detailed description dimensions corresponding to different operation complexity levels. When the operation complexity level corresponds to detailed requirements, it includes component collaboration logic, operation timing requirements, and parameter adjustment gradients. When the operation complexity level corresponds to non-detailed requirements, only the core operation actions and operation goals are retained; For the operation steps with corresponding detailed requirements at the operation complexity level, supplement the description of the coordinated triggering conditions between the multiple components involved in the operation step, the sequential order of operation execution, and the gradual adjustment gradient of key parameters to improve the details of the operation steps; For operation steps that do not meet the detailed requirements of the operation complexity level, delete repeated operation target statements, redundant safety tips, and background descriptions that are not related to the operation actions, and simplify the operation step descriptions; According to the initial order of the basic guidance framework, all the operation steps that have been supplemented, improved and simplified are integrated to form a scenario-adapted operation guidance sequence.

6. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 1 is characterized in that: The method of pushing the operation and maintenance guidance plan to the operation and maintenance personnel terminal, receiving in real time the operation feedback description returned by the operation and maintenance personnel after performing the operation based on the operation and maintenance guidance plan, and inputting the operation feedback description into the feedback semantic parsing module of the virtual artificial intelligence agent includes: Through the interactive push unit of the virtual artificial intelligence agent, according to the step sequence of the scenario-adaptive operation guidance sequence in the operation and maintenance guidance plan, the operation step description and the feedback prompt of the corresponding real-time semantic interaction node are synchronously pushed to the operation and maintenance personnel terminal; When each real-time semantic interaction node is triggered, the feedback receiving channel of the virtual artificial intelligence agent is started to receive the operation execution status, component response phenomenon and operation question expression input by the operation and maintenance personnel through the terminal, forming an operation feedback description; The operation feedback description is transmitted in real time so that the operation feedback description is immediately sent to the feedback semantic parsing module of the virtual artificial intelligence agent after the operation and maintenance personnel completes the input; After receiving the operation feedback description, the feedback semantic parsing module starts semantic word segmentation processing to separate factual expressions and interrogative expressions in the operation feedback description, and marks them as factual feedback segments and interrogative feedback segments respectively; The fact feedback segment and the question feedback segment are associated with corresponding real-time semantic interaction node identifiers to form initial input data for feedback semantic analysis.

7. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 6 is characterized in that: After receiving the operation feedback description, the feedback semantic parsing module starts semantic word segmentation processing to separate the factual statements and interrogative statements in the operation feedback description, marking them as factual feedback segments and interrogative feedback segments respectively, including: Calling a semantic word segmentation model of a virtual artificial intelligence agent to perform word segmentation on the operation feedback description to obtain a plurality of semantic word units; Matching the semantic vocabulary units with the factual expression vocabulary of the virtual artificial intelligence agent, screening out vocabulary units containing factual information such as operation results, component status, and parameter values, and combining them to form factual feedback segments; Match the remaining semantic vocabulary units with the question expression vocabulary of the virtual artificial intelligence agent, filter out vocabulary units containing question particles, uncertain expressions, and descriptions of operational confusion, and combine them to form question feedback segments; The integrity of the factual feedback segment is judged to check whether it contains the core result information corresponding to the operation steps. If it is missing, a supplementary query prompt is generated and pushed to the operation and maintenance personnel terminal. After receiving the supplementary feedback, the factual feedback segment is improved; The core questions of the question feedback segment are extracted to determine the specific confusion points of the operation and maintenance personnel regarding the operation steps, form a core identifier of the question feedback segment, and associate it with the corresponding real-time semantic interaction node identifier.

8. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 1 is characterized in that: The step of associating the real-time semantic interaction nodes in the operation and maintenance guidance plan with the feedback semantic parsing module, adjusting the step content and presentation order of the scenario-adaptive operation guidance sequence, and obtaining an updated operation and maintenance guidance plan includes: Extracting the fact feedback segment, the question feedback segment and the associated real-time semantic interaction node identifier corresponding to the operation feedback description through the feedback semantic parsing module; Matching the real-time semantic interaction node identifier with the real-time semantic interaction node in the operation and maintenance guidance scheme, and locating the target operation step in the scenario-adaptive operation guidance sequence corresponding to the real-time semantic interaction node; Analyze whether the factual feedback segment is consistent with the expected result description of the target operation step. If not, extract the abnormal result description in the factual feedback segment, call the abnormal response strategy library of the virtual artificial intelligence agent, and generate an operation adjustment suggestion for the abnormal result description; For the core confusing points in the question feedback segment, call the question answering knowledge base of the virtual artificial intelligence agent, match the answer statements related to the target operation steps, and incorporate them into the adjusted operation step description; According to the operation adjustment suggestions and answers, the content of the target operation steps in the scenario-adaptive operation guidance sequence is revised, and the presentation order of subsequent operation steps is adjusted based on the impact scope corresponding to the abnormal result description to form an updated operation and maintenance guidance plan.

9. The method for remote guidance and interaction of operation and maintenance technical services combined with intelligent agent assistance according to claim 8, characterized in that: The analyzing whether the fact feedback segment is consistent with the expected result description of the target operation step, if not, extracting the abnormal result description in the fact feedback segment, calling the abnormal response strategy library of the virtual artificial intelligence agent, and generating an operation adjustment suggestion for the abnormal result description, including: Extract the expected result description of the target operation step in the operation and maintenance guidance plan, and determine the component status parameters, operating parameter ranges, and operation completion indicators included in the expected result description; Compare the operation execution status and component response phenomena in the fact feedback segment with the expected result description point by point, mark the inconsistent differences, and determine the abnormal result description; Sending a query request to the abnormal response strategy library of the virtual artificial intelligence agent, so that the historical response strategy corresponding to the abnormal result description and the target operation step identifier is matched based on the query request, and the operating parameter type is supplemented with the parameter monitoring requirements in combination with the operating parameter type to form an initial response suggestion, wherein the query request includes the abnormal result description, the target operation step identifier and the operating parameter type in the operation and maintenance scenario feature recognition result; The initial response suggestions are analyzed for scenario adaptability to ensure that the operational adjustments in the initial response suggestions are consistent with the fault impact scope and operational complexity level of the current operation and maintenance scenario, and a final operational adjustment suggestion is generated.

10. A remote guidance interactive system for operation and maintenance technical services combined with intelligent agent assistance, characterized in that: It includes a processor and a computer-readable storage medium, wherein the computer-readable storage medium stores machine-executable instructions, and when the machine-executable instructions are executed by the processor, the remote guidance interaction method for operation and maintenance technical services combined with intelligent agent assistance as described in any one of claims 1 to 9 is implemented.

Images