A method, system, and medium for information completeness checking and history-driven inquiry in industrial equipment fault diagnosis.

Abstract

This invention discloses a method and system for information completeness checking and history-driven follow-up questioning in industrial equipment fault diagnosis. The method includes: defining an information completeness checklist; verifying the completeness of key information of the equipment to be diagnosed before starting the diagnosis; pausing the diagnosis process and triggering a follow-up questioning process when key information is missing; generating targeted follow-up questions based on the equipment's historical processing records; resuming the diagnosis process after obtaining user feedback; and releasing the diagnosis from pause if no response is received after the follow-up questioning validity period expires, and performing a conservative diagnosis output based on existing information. This invention solves the technical problems of blind inference, inaccurate templated follow-up questioning, and low information completion efficiency in predictive maintenance scenarios for industrial equipment when diagnostic information is missing, through an information completeness checking and history-driven follow-up questioning mechanism. It also avoids permanent suspension of the diagnosis process due to missing information through a follow-up questioning fallback mechanism.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology for industrial equipment, specifically to a method, system, and medium for information completeness checking and history-driven inquiry in predictive maintenance scenarios for industrial equipment. It is particularly suitable for predictive maintenance systems for industrial equipment based on an intelligent agent architecture, wherein the intelligent agent has the ability to self-check information completeness and accurately inquire based on historical experience. Background Technology

[0002] In industrial equipment monitoring scenarios, the health status diagnosis of rotating machinery relies on complete and accurate equipment information and operating parameters. The following specific technical problems exist in industrial settings, which existing diagnostic systems cannot effectively address: (1) Blind reasoning by the system when diagnostic information is missing. Traditional diagnostic systems typically fill in missing default values, continue reasoning based on assumptions, or output low-confidence conclusions when key parameters are missing, resulting in unreliable diagnostic results. In high-risk industries such as chemical engineering, diagnostic conclusions based on incomplete information may mislead on-site decision-making and bring safety hazards.

[0003] (2) The follow-up questions are templated and inaccurate. Although some systems support the follow-up question function when information is missing, the follow-up questions use fixed templates and do not take into account the equipment's historical experience and the current alarm scenario, resulting in poor follow-up questions, low user response rate and low information completion efficiency.

[0004] (3) Lack of closed-loop management for information completion. User-replyed information is used directly for diagnosis without verification, which may lead to problems such as input errors and unit confusion, resulting in secondary misdiagnosis based on erroneous information. At the same time, effective feedback is not updated in sync with equipment files, causing repeated loss of similar information.

[0005] (4) Lack of optimization mechanism for follow-up questioning strategy. The system does not record the effect of follow-up questions and the quality of user responses, and cannot optimize the content and strategy of follow-up questions based on historical data, resulting in a consistently poor follow-up questioning experience.

[0006] (5) The diagnostic process is permanently suspended when information is missing. Some systems do not have a fallback mechanism after follow-up inquiries fail, resulting in a long-term standstill in the diagnostic process. Users neither receive guidance on follow-up inquiries nor any diagnostic output, creating a service blind spot.

[0007] In summary, how to achieve a solution in predictive maintenance scenarios for industrial equipment that combines self-checking of information completeness, historical experience-driven follow-up questions, feedback information verification and file linkage updates, and information supplementation as a fallback when follow-up questions fail is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0008] In view of this, the present invention provides a method and system for information completeness checking and history-driven follow-up questioning in industrial equipment fault diagnosis, in order to solve the technical problems existing in the prior art, such as blind reasoning when diagnostic information is missing, inaccurate templated follow-up questioning content, lack of closed-loop management for information completion, lack of optimization mechanism for follow-up questioning strategy, and suspension of the diagnostic process when follow-up questioning is unsuccessful.

[0009] This invention provides a method for information completeness checking and history-driven inquiry in industrial equipment fault diagnosis, applied to predictive maintenance scenarios for industrial equipment. The method comprises: The information completeness check step involves verifying the completeness of preset key information items required for diagnosis based on the device identification of the device to be diagnosed before the diagnosis is initiated. The preset key information items include at least one or more of the following: device type, current speed, bearing model, and measuring point direction, and can be dynamically configured according to the device type. The follow-up questioning triggers a process where, if any item is missing from the key information, the current diagnostic process is paused, the diagnostic pause status is marked, and the follow-up questioning process is triggered. The history-driven follow-up question generation step involves retrieving the device's historical processing records based on the device's identity identifier, analyzing the historical fault modes, historical processing measures, and historical follow-up question responses in the historical processing records, and generating targeted follow-up question content related to the current alarm scenario. The follow-up question output step involves outputting the targeted follow-up questions to the user interaction terminal and waiting for user feedback. The diagnostic recovery step involves parsing the key parameters in the feedback information after receiving user feedback, adding them to the diagnostic information set, lifting the diagnostic pause status, and resuming the diagnostic process. The follow-up step automatically lifts the diagnostic pause if no valid response is received from the user after the preset follow-up period and the preset maximum number of follow-up questions has been reached. Based on the existing information, a conservative diagnostic output is performed, and the diagnostic conclusion is marked with missing information and corresponding uncertainty prompts.

[0010] Optionally, the information completeness check step further includes a comparison and verification of static and dynamic information.

[0011] Optionally, the history-driven follow-up question generation step includes four sub-steps: historical record retrieval, scene association analysis, missing information analysis, and follow-up question content generation. The scene association analysis employs feature matching, matching the feature parameters of the current alarm with the features of historical alarms in the historical record to filter out related historical records, rather than being limited to a specific similarity calculation algorithm.

[0012] Optionally, the follow-up question generation sub-step generates different types of follow-up questions based on the type of missing item.

[0013] Optionally, the history-driven follow-up question generation step further includes a personalized follow-up question strategy, which employs different follow-up question generation logics based on the existence or absence of associated historical records.

[0014] Optionally, it also includes a follow-up questioning effectiveness evaluation step, recording follow-up questioning effectiveness data and optimizing the follow-up question generation algorithm.

[0015] Optionally, it may also include a follow-up questioning frequency control step, which limits the number of follow-up questions, the intervals between them, and the reminder strategy.

[0016] Optionally, the diagnostic recovery step may further include verifying the rationality of the feedback information and questioning the parameter correction.

[0017] Optionally, it also includes an information update linkage step, which synchronizes the effective feedback to the device file.

[0018] Optionally, the information completeness checklist supports dynamically expandable configuration.

[0019] In addition, the present invention provides an information completeness check and history-driven follow-up system for fault diagnosis of industrial equipment, characterized in that it includes: an information completeness check module, a follow-up trigger module, a history-driven follow-up generation module, a follow-up output module, a diagnosis recovery module, and a follow-up fallback module.

[0020] Furthermore, the present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the method described in any of the preceding claims.

[0021] The beneficial effects of this invention are as follows: The information completeness check step identifies missing information before diagnosis begins, avoiding blind reasoning based on incomplete information; the pause diagnosis mechanism triggered by follow-up questions ensures the completeness of the information required for diagnosis; the history-driven follow-up question generation step generates targeted follow-up questions based on the equipment's historical processing records, solving the problem of inaccurate follow-up question templates; the follow-up question effect evaluation step establishes a continuous optimization mechanism for the follow-up question strategy; the information update linkage step achieves synchronized updates of effective feedback and equipment files, avoiding repeated omissions of similar information; and the follow-up question fallback step automatically lifts the diagnosis pause and executes conservative diagnostic output when follow-up questions are unsuccessful, avoiding service blind spots caused by permanent suspension of the diagnosis process. This invention provides information completion support for predictive maintenance systems for industrial equipment based on an intelligent agent architecture, possessing capabilities of "completeness self-check - precise follow-up question - closed-loop update - fallback output," systematically solving the technical problem of managing missing diagnostic information in industrial equipment monitoring scenarios, with an overall effect greater than the sum of its parts. Attached Figure Description

[0022] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings: Figure 1 The flowchart of an information completeness check and history-driven inquiry method in industrial equipment fault diagnosis according to Embodiment 1 of the present invention is shown. Figure 2 This diagram illustrates the structure of the information completeness checklist in Embodiment 1 of the present invention. Figure 3 The flowchart of history-driven follow-up question generation in Embodiment 1 of the present invention is shown; Figure 4 This diagram illustrates a decision tree for the follow-up question generation strategy in Embodiment 1 of the present invention. Figure 5 This illustrates a closed-loop flowchart of the inquiry effect evaluation and strategy optimization in Embodiment 1 of the present invention; Figure 6 The flowchart illustrating the linkage between diagnosis recovery and information update in Embodiment 1 of the present invention is shown. Figure 7 The flowchart of the follow-up questioning step in Embodiment 1 of the present invention is shown; Figure 8 The diagram shows the structure of an information completeness check and history-driven inquiry system for fault diagnosis of industrial equipment according to Embodiment 2 of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In this embodiment of the invention, the term "and / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following associated objects have an "or" relationship.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0026] In this embodiment of the invention, the term "multiple" refers to two or more, and other quantifiers are similar. Example 1

[0027] This embodiment provides a method for information completeness checking and history-driven inquiry in industrial equipment fault diagnosis, such as... Figure 1 As shown, the method includes: S1: Information completeness check step. Before the diagnosis is started, the key information required for the diagnosis is checked based on the device identity of the device to be diagnosed. S2: Follow-up questioning triggers the process. When there are missing items in the key information, the current diagnostic process is paused, the diagnostic pause status is marked, and the follow-up questioning process is triggered. S3: History-driven follow-up question generation step, which retrieves the device's historical processing records based on the device's identity and generates targeted follow-up questions related to the current alarm scenario; S4: Follow-up Question Output Steps: Output targeted follow-up questions to the user interaction platform and wait for user feedback; S5: Diagnostic recovery step. After receiving user feedback information, the key parameters in the feedback information are parsed, added to the diagnostic information set, the diagnostic pause status is lifted, and the diagnostic process is resumed. S6: Follow-up questioning step. If no valid response is received from the user after the preset follow-up questioning period, the diagnosis pause status will be automatically lifted, and conservative diagnosis output will be performed based on the existing information.

[0028] In this embodiment, the missing information is identified before the diagnosis is initiated through an information completeness check step, the diagnosis is paused through a follow-up question trigger step to avoid blind reasoning through a follow-up question generation step based on historical experience to generate accurate follow-up questions through a history-driven follow-up question generation step, the diagnosis is closed-loop after information is completed through a diagnosis recovery step, and the follow-up question fallback step ensures that the diagnosis service is not interrupted when follow-up questions are unsuccessful.

[0029] Each step of this embodiment will be described in detail below.

[0030] In this embodiment S1, the system performs an information completeness check.

[0031] Based on the equipment identification of the device to be diagnosed, the system reads the static information of the equipment from the equipment file database, including equipment type, rated speed, installed bearing model, measuring point configuration information, equipment tag number, equipment name, and workshop to which it belongs. At the same time, the system reads the dynamic information of the equipment from the real-time monitoring system, including the current actual speed, current operating conditions, and current load status.

[0032] The system compares and verifies static and dynamic information. If there is a discrepancy between the dynamic and static information that exceeds the preset tolerance range, the dynamic information will be used and an error message will be displayed. For example, if the rated speed in the equipment file is 1480 rpm, but the real-time monitoring shows the current speed is 1450 rpm, the deviation is within the preset tolerance range and is considered normal fluctuation. If the current speed is 1200 rpm, the deviation exceeds the tolerance range, then 1200 rpm will be used and a message will be displayed: "Speed ​​deviates from rated value, please confirm whether to adjust speed."

[0033] The information completeness checklist should include at least the following key information items: Equipment type: Used to determine the applicable diagnostic rule base and fault mode base; Current rotational speed: used to calculate characteristic frequency and determine operating condition; Bearing model: Used to query bearing characteristic frequency parameters and fault characteristic database; Measurement point direction: used to determine the orientation of vibration measurement and the focus of diagnosis.

[0034] The information completeness checklist supports dynamic expansion, allowing administrators to configure different sets of check items and verification rules based on equipment type and diagnostic needs. For example, for gearbox equipment, check items such as the number of gear teeth and transmission ratio can be added; for pump equipment, check items such as the number of impeller blades and media type can be added.

[0035] In this embodiment S2, when there are missing items in the key information, the system pauses the current diagnostic process.

[0036] The system marks the diagnostic process as paused, recording the reason for the pause (missing information), a list of missing items, and the current alarm flag. While paused, the diagnostic engine does not perform inference calculations to avoid outputting unreliable conclusions based on incomplete information.

[0037] Simultaneously, the system triggers a follow-up questioning process, adding the follow-up questioning task to an asynchronous task queue for execution by the follow-up questioning engine. The follow-up questioning engine reads the list of missing items and prepares to generate targeted follow-up questions.

[0038] In this embodiment S3, the system retrieves the device's historical processing records based on the device's identity identifier and generates targeted follow-up questions.

[0039] Historical record retrieval sub-step: The system retrieves the device's historical processing records based on the device's identity identifier, sorts them in reverse chronological order, and extracts the most recent preset number of historical records. Historical processing records include historical alarm times, historical alarm characteristic parameters, historical diagnostic conclusions, historical user feedback, historical processing measures, historical follow-up questions, and historical follow-up question responses.

[0040] Scene association analysis sub-step: The system performs feature matching between the feature parameters of the current alarm and the features of historical alarms in the historical records. Feature matching comprehensively considers dimensions such as feature frequency, type of exceeding indicator, trend label, and alarm level, and filters out historical records with a matching degree exceeding a preset association threshold as associated historical records. The feature matching is not limited to a specific similarity calculation algorithm and can be implemented in various ways such as feature vector distance, rule matching, or machine learning models. For example, if the current alarm is "the vertical speed at the pump coupling end is slowly increasing, and the high-frequency envelope bearing energy is significantly exceeding the standard," and a certain historical record is "the horizontal speed at the pump coupling end is slowly increasing, and the high-frequency envelope bearing energy is exceeding the standard," the feature matching result exceeds the threshold and is determined to be an associated historical record.

[0041] Missing Information Analysis Sub-step: The system analyzes the currently missing key information items to determine the specific parameters that need to be investigated. If there are multiple missing items, they are sorted according to information dependency priority. The information dependency priority is determined based on the needs of the diagnostic rule base. For example, the bearing model has the greatest impact on the diagnostic conclusion and has the highest priority; the measurement point direction has a relatively smaller impact and has a lower priority.

[0042] The follow-up question generation sub-step involves the system generating targeted follow-up questions based on historical processing measures and responses from the associated historical records. These targeted follow-up questions reference specific historical scenarios from the associated historical records, forming a comparative inquiry.

[0043] The generation of follow-up questions also includes personalized follow-up questioning strategies: If a similar alarm exists in the historical records and the historical follow-up responses contain a valid value for the missing item, then the historical value will be directly used as the reference for follow-up inquiries. For example: "When this device last triggered a similar alarm (April 15, 2026), the engine speed was 1480 rpm. Is the current engine speed still 1480 rpm?" If a similar alarm exists in the associated historical records and the historical handling measures mention inspection actions related to the missing item, then a follow-up question related to the action is generated. For example: if the historical records show that similar faults usually require checking the bearing lubrication status, and the bearing model is currently missing, then the question would be: "The last maintenance record for the bearing at the coupling end of this pump shows 6314. Is it still 6314? Also, the last recommendation was to check the lubrication status. Has that been checked?" If no similar alarm is found in the associated historical records, a general follow-up question is generated, requesting only the specific value of the missing item. For example: "Please provide the current operating speed of this device so that we can accurately calculate its characteristic frequency." The follow-up questions employ different generation strategies depending on the type of missing item: When the missing item is the equipment type or bearing model, a confirmation question is generated to inquire whether the current equipment configuration matches the file. For example: "The equipment file lists the bearing model for the pump coupling end as 6314. Does the actual installed model match this?" When the missing item is the current speed, a follow-up question related to the operating condition is generated to inquire about the current operating condition and speed adjustment status. For example: "The rated speed of this equipment is 1480 rpm. What is the current actual operating speed? Has there been any recent speed adjustment?" When the missing item is the measurement point direction, a location confirmation follow-up question is generated to inquire about the specific installation orientation of the abnormal measurement point. For example: "The current alarm is from the measurement point at the pump coupling end. Is the installation orientation of this measurement point vertical (V) or horizontal (H)?" In this embodiment S4, the system outputs the generated targeted follow-up questions to the user interaction terminal.

[0044] Follow-up questions are delivered via instant messaging channels such as WeChat Work, DingTalk, and SMS. The questions should be phrased politely and professionally, clearly explaining the importance of the missing information for diagnosis, and providing sample responses to lower the barrier to entry for users.

[0045] The system simultaneously starts a follow-up inquiry validity period timer. During the follow-up inquiry validity period, the diagnostic process remains paused, awaiting user feedback.

[0046] A frequency control mechanism for follow-up questions ensures the rationality of follow-up questions: The number of follow-up inquiries for the same alarm from the same device shall not exceed the preset maximum number of follow-up inquiries, so as to avoid excessively disturbing the user; The time interval between two consecutive follow-up questions is no less than the preset minimum follow-up question interval, giving users ample time to respond; If the user does not reply within the preset follow-up question validity period, the system will automatically send a follow-up question reminder, and the number of reminders will not exceed the preset maximum number of reminders; If no response is received after the preset follow-up period expires, the system will trigger a fallback follow-up step and will not continue to wait.

[0047] In this embodiment S5, after receiving user feedback information, the system parses the key parameters in the feedback information.

[0048] The system uses natural language processing technology to extract key parameter values, including value recognition, unit extraction, and contextual understanding. For example, if a user replies "the RPM is currently around 1450", the system extracts the value 1450, the unit rpm (inferred from the context), and records the approximate range represented by "around".

[0049] The system performs a validity check on the extracted key parameters: Numerical range verification: Check whether the parameter value is within the normal operating range of the equipment. For example, a speed of 1450 rpm is within the reasonable fluctuation range of the rated speed of 1480 rpm; if the user replies that the speed is 500 rpm, it is significantly deviating from the normal range, and the verification will fail.

[0050] Unit consistency check: Check whether the units provided by the user match the units expected by the system. If they do not match, perform unit conversion or request user confirmation.

[0051] Logical consistency check: This checks whether the logical relationship between parameters is reasonable. For example, if the user replies that the current speed is 0 rpm but the device status is "running", there is a logical contradiction, and the check will fail.

[0052] If the rationality check passes, the key parameters are added to the diagnostic information set, the diagnostic pause is lifted, and the diagnostic process resumes. The diagnostic engine then performs normal diagnostic reasoning based on the completed information set.

[0053] If the validity check fails, a parameter correction question will be generated, requesting user confirmation or correction. For example: "The speed of 500 rpm you replied with does not match the operating status of this equipment. Please confirm whether it is indeed 500 rpm, or provide the current actual operating speed." In addition, this embodiment also includes an information update linkage step: When key parameters reported by the user are inconsistent with the static information in the equipment file, the system generates an equipment file update suggestion and pushes it to the equipment administrator. For example: "The user reports that the current bearing model is 6316, which is inconsistent with 6314 in the file. Please confirm whether the file needs to be updated." Once the key parameters provided by the user are verified as valid, the system automatically updates the corresponding fields in the device file to ensure the timeliness of the file information. The updated device file is used for subsequent alarm diagnostic information completeness checks to avoid repeated omissions of similar information.

[0054] In addition, this embodiment also includes a follow-up questioning effect evaluation step: The system records the timestamp, content, user response time, and response content for each follow-up question. It calculates the follow-up question response rate, statistically analyzing the percentage of follow-up questions that receive valid responses within a preset validity period. The system then determines the validity of each response: if the response contains the parameters requested in the follow-up question, it is marked as a valid response; otherwise, it is marked as an invalid response.

[0055] The system evaluates the effectiveness of different follow-up question generation strategies based on the response rate and effective response rate. For example, it compares the response rate difference between history-driven follow-up questions and general templates to evaluate the effectiveness of the history-related strategy; it also compares the response rate difference between different follow-up question tones to optimize the expression of follow-up questions. The evaluation results are used to continuously optimize the follow-up question generation algorithm and improve information completion efficiency.

[0056] In this embodiment S6, when no valid response is received from the user after the preset follow-up question validity period has expired and the preset maximum number of follow-up questions has been reached, the system executes the follow-up question fallback step.

[0057] The system first checks whether the follow-up questioning trigger conditions have been met: the number of follow-up questions has reached the preset maximum number of follow-up questions, and the last follow-up question has exceeded the preset follow-up question validity period. If the conditions are met, the system automatically releases the diagnostic pause state and marks the alarm as "information missing and pending processing".

[0058] The system performs conservative diagnostic output based on existing complete information (i.e., information items confirmed to be complete in the information completeness check). Characteristics of conservative diagnostic output include: For confirmed information items, use normal logic to participate in diagnostic reasoning; For missing information items, the diagnostic conclusion should clearly state "Due to the lack of [specific missing item] information, this diagnosis is uncertain"; The diagnostic confidence level is downgraded by one level (e.g., if it was originally high confidence, it is downgraded to medium confidence; if it was originally medium confidence, it is downgraded to low confidence). The diagnostic conclusion only provides a broad range of possible faults, without specifying the exact location of the fault. Add a prompt to the output: "It is recommended to supplement and confirm [missing items] to improve diagnostic accuracy."

[0059] For example, if the missing item is the bearing model, the system outputs: "Based on the existing information, the vibration at the pump coupling end of this equipment is abnormal, with a slow upward trend and excessive high-frequency envelope energy. Due to the lack of bearing model information, the characteristic frequency of the bearing cannot be accurately located. Early degradation in the bearing area is initially suspected (low confidence). It is recommended to supplement and confirm the bearing model before re-diagnosing." This follow-up questioning mechanism ensures that even if the user does not respond, the system can still provide valuable reference output, avoiding service blind spots caused by the permanent suspension of the diagnostic process. At the same time, by explicitly marking missing information and lowering the confidence level, it reminds users of the uncertainty of the current diagnosis, avoiding the risk of misdiagnosis due to missing information.

[0060] The complete information completeness check and history-driven inquiry method described in this embodiment has the following beneficial effects: (1) It solves the technical problem of blind reasoning when diagnostic information is missing: the missing information is identified before the diagnosis is started by checking the completeness of the information, and the pause diagnosis mechanism triggered by the follow-up questioning is used to forcibly prevent incomplete information from entering the diagnostic engine, thus avoiding unreliable diagnosis based on assumptions and default values.

[0061] (2) Solved the technical problem of inaccurate templated follow-up questions: By driving the follow-up question generation process through history, targeted follow-up questions are generated based on real scenarios in the device's historical processing records. Historical cases are cited to form comparative inquiries, which significantly improves the relevance of follow-up questions and the user response rate. By adopting the higher-level "feature matching" expression, the follow-up question generation logic is not limited to a specific algorithm, thus expanding the scope of protection.

[0062] (3) Solved the technical problem of lack of closed-loop management of information completion: Through the rationality verification mechanism of the diagnosis and recovery steps, the secondary misdiagnosis caused by user input errors was avoided; through the information update linkage steps, the effective feedback was synchronously updated to the equipment file, realizing the closed-loop management of information completion.

[0063] (4) The technical problem of lack of optimization mechanism in the follow-up questioning strategy is solved: through the follow-up questioning effect evaluation step, a continuous evaluation and optimization closed loop of the follow-up questioning strategy is established, so that the follow-up questioning generation algorithm can be continuously improved based on real data.

[0064] (5) Solved the technical problem of the diagnosis process being permanently suspended when follow-up questions fail: Through the follow-up question fallback step, the diagnosis suspension state is automatically lifted after the follow-up question validity period has expired and the maximum number of follow-up questions has been reached. Based on the existing information, a conservative diagnosis output is executed, which ensures the continuity of the diagnosis service and avoids service blind spots caused by user non-response.

[0065] (6) It realizes a human-like "information confirmation-experience reference-precise inquiry-backstop service" probing style: This invention realizes the human-like information confirmation habit through information completeness checks, realizes the human-like experience reference ability through history-driven probing, realizes the human-like precise inquiry style through personalized probing strategies, and realizes the human-like flexible adaptability through probing backstop service. This probing behavior is highly anthropomorphic and fundamentally different from the traditional mechanical probing with fixed templates. Example 2

[0066] An information completeness check and history-driven inquiry system for fault diagnosis of industrial equipment, such as Figure 8 As shown, the system includes an information completeness check module, a follow-up question triggering module, a history-driven follow-up question generation module, a follow-up question output module, a diagnosis and recovery module, and a follow-up question fallback module: The information completeness check module is used to verify the completeness of key information required for diagnosis before the diagnosis is initiated. The follow-up questioning triggering module is used to pause the diagnostic process and trigger the follow-up questioning process when key information is missing. The history-driven follow-up question generation module is used to generate targeted follow-up questions based on the device's historical processing records; The follow-up question output module is used to output targeted follow-up questions to the user interaction terminal; The diagnostic recovery module is used to parse key parameters, verify their rationality, supplement the diagnostic information set, and restore the diagnostic process after receiving user feedback information. The follow-up questioning module is used to lift the diagnostic pause state and execute conservative diagnostic output when no response is received after the follow-up questioning period has expired.

[0067] In this embodiment, a pre-diagnosis information quality control is achieved through an information completeness check module; a diagnostic process pause protection is achieved through a follow-up question trigger module; accurate follow-up questions based on historical experience are achieved through a history-driven follow-up question generation module; follow-up question output is achieved through multi-channel follow-up question reach; a diagnosis recovery module achieves a closed-loop diagnosis after information completion; and a follow-up question fallback module ensures service continuity when follow-up questions are unsuccessful. The system also includes interfaces with the equipment file database, a historical memory retrieval interface, a user interaction interface, and a file update interface, forming a complete information completion closed loop.

[0068] The functions of each part in the information completeness check and history-driven inquiry system described in this embodiment are the same as the methods and steps described in Embodiment 1. Therefore, for details not covered in this embodiment, please refer to Embodiment 1 and... Figures 1 to 7 The specific details will not be elaborated here. Example 3

[0069] This embodiment also provides an information completeness check and history-driven inquiry device for industrial equipment fault diagnosis, including a processor, a memory, and a computer program stored in the memory and run on the processor. When the computer program runs, it implements the method steps of the information completeness check and history-driven inquiry method in Embodiment 1.

[0070] By using a computer program stored in memory and running on a processor, it is possible to perform information integrity checks, follow-up question triggering, history-driven follow-up question generation, follow-up question output, diagnostic recovery, and follow-up question fallback, effectively solving the technical problem of managing missing diagnostic information in industrial equipment monitoring scenarios.

[0071] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the computer device, connecting all parts of the computer device through various interfaces and lines.

[0072] Memory can be used to store computer programs and / or models. The processor performs various functions of the computer device by running or executing the computer programs and / or models stored in the memory, and by accessing data stored in the memory. Memory can primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on usage. Furthermore, memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0073] It should be understood that each block of a flowchart and / or block diagram, and combinations of blocks in a flowchart and / or block diagram, can be implemented by a computer program. These computer programs can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that instructions executable by the processor of the computer or other programmable data processing device generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] These computer programs may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0075] These computer programs may also be loaded onto a computer or other programmable data processing equipment, causing a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0076] This embodiment also provides a computer-readable storage medium, which includes at least one instruction that, when executed by a computer, implements the method steps in the information completeness check and history-driven inquiry method of Embodiment 1.

[0077] By executing a computer-readable storage medium containing at least one instruction, it is possible to achieve information completeness checks, follow-up question triggering, history-driven follow-up question generation, follow-up question output, diagnostic recovery, and follow-up question fallback, effectively solving the technical problem of managing missing diagnostic information in industrial equipment monitoring scenarios.

[0078] Similarly, for details not covered in this embodiment, please refer to Embodiment 1, Embodiment 2, and... Figures 1 to 8 The specific details will not be elaborated here.

[0079] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. An information completeness check and history driven interrogation method in industrial equipment failure diagnosis, applied to industrial equipment predictive maintenance scenarios, characterized in that, include: The information completeness check step involves verifying the completeness of preset key information items required for diagnosis based on the device identity of the device to be diagnosed before the diagnosis is initiated. The preset key information items include at least one or more of the following: equipment type, current speed, bearing model, and measuring point direction, and can be dynamically configured according to the equipment type; The follow-up questioning triggers a process where, if any item is missing from the key information, the current diagnostic process is paused, the diagnostic pause status is marked, and the follow-up questioning process is triggered. The history-driven follow-up question generation step involves retrieving the device's historical processing records based on the device's identity identifier, analyzing the historical fault modes, historical processing measures, and historical follow-up question responses in the historical processing records, and generating targeted follow-up question content related to the current alarm scenario. The follow-up question output step involves outputting the targeted follow-up questions to the user interaction terminal and waiting for user feedback. The diagnostic recovery step involves parsing the key parameters in the feedback information after receiving user feedback, adding them to the diagnostic information set, lifting the diagnostic pause status, and resuming the diagnostic process. The follow-up step automatically lifts the diagnostic pause if no valid response is received from the user after the preset follow-up period and the preset maximum number of follow-up questions has been reached. Based on the existing information, a conservative diagnostic output is performed, and the diagnostic conclusion is marked with missing information and corresponding uncertainty prompts.

2. The method of claim 1, wherein, The information completeness check step also includes: Retrieve static information of the equipment from the equipment file database, including equipment type, rated speed, bearing model, and measuring point configuration information; Read dynamic information about the equipment from the real-time monitoring system, including the current actual speed and current operating conditions; The static information is compared and verified with the dynamic information. If there is a deviation between the dynamic information and the static information and the deviation exceeds the preset tolerance range, the dynamic information shall prevail and an information abnormality prompt shall be marked.

3. The method according to claim 1 or 2, characterized in that, The history-driven follow-up question generation steps include: The historical record retrieval sub-step retrieves the historical processing records of the device based on the device identification, sorts them in reverse chronological order, and extracts the most recent preset number of historical records. The scene association analysis sub-step performs feature matching between the feature parameters of the current alarm and the features of historical alarms in the historical records, and filters out the historical records with a matching degree exceeding the preset association threshold as the associated historical records. The missing information analysis sub-step analyzes the currently missing key information items and determines the specific parameters that need to be followed up. The follow-up question generation sub-step generates targeted follow-up questions based on historical processing measures and historical follow-up question responses in the associated historical records; the targeted follow-up questions reference specific historical scenarios in the associated historical records to form a comparative inquiry.

4. The method according to any one of claims 1 to 3, characterized in that, The sub-step for generating follow-up questions also includes: When the missing item is the equipment type or bearing model, a confirmation question is generated to ask whether the current configuration of the equipment is consistent with the file. When the missing item is the current speed, generate a follow-up question related to the operating condition to inquire about the current operating condition and speed adjustment status. When the missing item is the direction of the measuring point, a follow-up question of the location confirmation type is generated to inquire about the specific installation location of the abnormal measuring point; When there are multiple missing items, sort them according to information dependency priority, and first generate follow-up questions about the missing items that have the greatest impact on the diagnostic conclusion.

5. The method according to any one of claims 1-4, characterized in that, The history-driven follow-up question generation step also includes a personalized follow-up question strategy: If there are similar alarms in the associated history and the historical follow-up replies contain a valid value for the missing item, then the historical value is directly used as a reference for follow-up questions to inquire whether the current value is consistent with the historical value. If there are similar alarms in the associated history and the historical handling measures mention inspection actions related to the missing item, then an action association follow-up question is generated to ask whether the relevant inspection actions have been completed. If no similar alarm is found in the associated historical records, a general follow-up question is generated, which only asks for the specific value of the missing item.

6. The method according to any one of claims 1-5, characterized in that, It also includes follow-up questioning and evaluation steps: Record the timestamp of each follow-up question, the content of the follow-up question, the user's reply time and reply content; Calculate the follow-up question response rate and count the percentage of follow-up questions that received valid responses within the preset follow-up question validity period; The validity of the response is determined. If the response contains the parameter information requested by the follow-up question, it is marked as a valid response; otherwise, it is marked as an invalid response. Based on the follow-up question response rate and effective response rate, the effectiveness of different follow-up question generation strategies is evaluated, which is used to optimize the follow-up question generation algorithm.

7. The method according to any one of claims 1-6, characterized in that, It also includes follow-up questioning on frequency control steps: The number of follow-up inquiries for the same alarm from the same device shall not exceed the preset maximum number of follow-up inquiries; The time interval between two consecutive follow-up questions shall not be less than the preset minimum follow-up question interval; If the user does not reply within the preset follow-up question validity period, the system will automatically send a follow-up question reminder, and the number of reminders will not exceed the preset maximum number of reminders; If no response is received within the preset follow-up period, the system will automatically mark the alarm as an information missing pending processing status and trigger the follow-up step.

8. The method according to any one of claims 1-7, characterized in that, The diagnostic recovery steps also include: Analyze user feedback information and use natural language processing technology to extract key parameter values; Perform rationality verification on the extracted key parameters, including numerical range verification, unit consistency verification, and logical consistency verification; If the rationality check passes, the key parameters will be added to the diagnostic information set. If the validity check fails, a parameter correction question will be generated, requesting the user to confirm or correct it.

9. The method according to any one of claims 1-8, characterized in that, It also includes information update linkage steps: When key parameters reported by users are inconsistent with the static information in the device profile, a device profile update suggestion is generated. Once the key parameters provided by the user are verified to be valid, the corresponding fields in the device profile will be automatically updated. The updated device profile is used to check the completeness of diagnostic information for subsequent alarms.

10. The method according to any one of claims 1-9, characterized in that, The information completeness checklist supports dynamic expansion, and administrators can configure different sets of check items and verification rules according to device type and diagnostic needs.

11. A system for information completeness checking and history-driven inquiry in industrial equipment fault diagnosis, characterized in that, include: The information completeness check module is used to verify the completeness of key information required for diagnosis before the diagnosis is initiated. The follow-up questioning trigger module is used to pause the diagnostic process and trigger a follow-up questioning process when key information is missing. The history-driven follow-up question generation module is used to generate targeted follow-up questions based on the device's historical processing records. The follow-up question output module is used to output targeted follow-up questions to the user interaction terminal; The diagnostic recovery module is used to resume the diagnostic process after receiving user feedback. The follow-up question module is used to lift the diagnostic pause state and execute conservative diagnostic output when no response is received after the follow-up question validity period has expired.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-10.

Images