A search-enhanced method, system, electronic device, and storage medium for assisting in troubleshooting maintenance and repair.

By employing a dual-path retrieval method and a knowledge closed-loop update mechanism, the problems of low knowledge retrieval efficiency, insufficient accuracy, and lack of confidence assessment in operation and maintenance fault handling are solved, thereby achieving efficient and accurate operation and maintenance fault handling and knowledge reuse.

CN123118080APending Publication Date: 2026-11-17SHENZHEN PUHUI ZHIWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611619859.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-20
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing methods for handling operational and maintenance faults rely on human experience, which leads to problems such as low knowledge retrieval efficiency, insufficient accuracy, difficulty in knowledge reuse, and lack of confidence assessment, resulting in low efficiency in handling operational and maintenance faults and knowledge loss.

Method used

A dual-path retrieval method combining vector retrieval and keyword retrieval is adopted. By obtaining work order fields and equipment files, a retrieval query is constructed, the retrieval score and confidence level are calculated, and different levels of handling guidance are pushed. After handling difficult cases, a knowledge closed loop is formed for updating.

Benefits of technology

It improved the efficiency and accuracy of operation and maintenance fault handling, enabled the efficient reuse of knowledge, solved the problem of knowledge loss, and enhanced the ability of operation and maintenance personnel to judge the credibility of information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123118080A_ABST
    Figure CN123118080A_ABST
Patent Text Reader

Abstract

This invention discloses a retrieval-enhanced method, system, electronic device, and storage medium for handling maintenance faults, relating to the field of intelligent maintenance technology. The method includes: S10, obtaining the work order fields and equipment files from on-site maintenance work orders; S20, constructing a retrieval query based on the work order fields and equipment files; S30, performing a dual-path retrieval of vector retrieval and keyword retrieval in a pre-built maintenance knowledge base, and calculating the retrieval score; S40, selecting main and secondary reference fragments, extracting handling steps sequentially, and associating each step with the corresponding original text source; S50, calculating the original text citation coverage rate and the retrieval confidence score; S60, pushing handling guidance of different levels to the on-site terminal based on the comparison result of the retrieval confidence score and a preset threshold; S70, collecting actual handling steps to generate new knowledge fragments, which are then reviewed and written into the maintenance knowledge base. The beneficial effects of this invention are: it can significantly improve the efficiency, accuracy, and knowledge reuse level of handling maintenance faults.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology, and more specifically, to a search-enhanced operation and maintenance fault handling auxiliary method, system, electronic device, and storage medium. Background Technology

[0002] With the continuous improvement of enterprise informatization and intelligentization, the scale of various production and operation systems is becoming increasingly large, with diverse equipment types and complex operating environments, leading to a corresponding increase in the frequency of operation and maintenance failures. When equipment fails, operation and maintenance personnel typically need to troubleshoot and handle the problem based on historical experience or standard operating procedures. However, traditional operation and maintenance failure handling methods mainly rely on manual experience and paper documents, which have the following significant shortcomings: Low knowledge retrieval efficiency: Existing operation and maintenance knowledge bases are mostly static document structures, lacking efficient retrieval mechanisms based on the context of work orders. After receiving a work order, operation and maintenance personnel often need to manually search for matching solutions in a large number of documents, which is time-consuming and can easily delay the handling of emergency failures.

[0003] Insufficient retrieval accuracy: Relying solely on keyword matching cannot understand the semantics of work orders, easily missing highly relevant knowledge fragments or returning a large number of irrelevant results. Furthermore, single-modal retrieval (such as text-only or vector-only retrieval) struggles to simultaneously meet the needs of precise matching and semantic generalization.

[0004] Difficulty in knowledge reuse: Many troubleshooting experiences are stored in personal memories or unstructured documents, lacking a systematic mechanism for collection, verification, and storage. For new and challenging faults encountered for the first time, it is often impossible to formulate effective handling guidelines in a timely manner, leading to duplication of effort and knowledge loss.

[0005] Lack of confidence assessment: When pushing search results, the existing system lacks a quantitative assessment of the reliability of the search results. It is difficult for operation and maintenance personnel to judge the accuracy of the information obtained, which may lead to erroneous operations due to the adoption of low-quality suggestions.

[0006] Therefore, the present invention provides a retrieval-enhanced auxiliary method, system, electronic device and storage medium for handling operation and maintenance faults, which can significantly improve the efficiency, accuracy and knowledge reuse level of handling operation and maintenance faults. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, this invention provides a retrieval-enhanced auxiliary method, system, electronic device, and storage medium for handling maintenance and operation faults, which can significantly improve the efficiency, accuracy, and knowledge reuse level of handling maintenance and operation faults.

[0008] The technical solution adopted by this invention to solve its technical problem is: a search-enhanced operation and maintenance fault handling assistance method, the improvement of which is that the search-enhanced operation and maintenance fault handling assistance method includes the following steps: S10, obtain the work order field of the on-site maintenance work order and the equipment file of the target device corresponding to the work order; S20, construct a retrieval query based on the work order field and equipment file, the retrieval query including query text, keyword set and query vector; S30, perform dual-path retrieval of vector retrieval and keyword retrieval in the pre-built maintenance knowledge base, and calculate the retrieval score for each candidate knowledge fragment according to the fusion weight; S40: Select the main reference fragment and the auxiliary reference fragment from the candidate knowledge fragments, extract the processing steps in sequence and associate the original text sources corresponding to each step to obtain the step-by-step processing operation; S50, calculate the original text citation coverage rate based on the proportion of steps with original text sources in the step-by-step processing operation, and calculate the retrieval confidence based on the highest retrieval score and the original text citation coverage rate; S60, based on the comparison result between the search confidence level and the preset threshold, push different levels of handling instructions to the on-site terminal. Among them, when the search confidence level is lower than the preset first threshold, the work order is marked as a difficult case. S70, in response to a work order being marked as a difficult case and having been handled on-site and passed customs clearance, the actual handling steps are collected to generate a new knowledge fragment, which is then written into the maintenance knowledge base after review.

[0009] Furthermore, in step S20, the specific method for constructing the retrieval query based on the work order field and equipment file is as follows: The query text is obtained by combining the work order description, fault code, work order location, and category, model, and alias from the equipment file using fixed fields. Extract fault codes, equipment models, product categories, and phenomenon terms from the query text to form a keyword set; The query vector is obtained by encoding the query text using an encoder that is of the same origin as the fragment vector in the knowledge base.

[0010] Furthermore, in step S30, the calculation expression for the retrieval score is: In the formula, This represents the retrieval score of the k-th knowledge fragment; This indicates the weight of vector similarity in the retrieval score; The similarity between the query vector and the k-th knowledge fragment is expressed as follows: ,in, This represents the L2 norm of the query vector q; This represents the vector of the k-th knowledge fragment in the maintenance knowledge base. The second norm; Represents the query vector q and the knowledge base fragment vector The inner product; Vector similarity is delimited to prevent division by zero. The keyword hit rate of the k-th knowledge fragment is expressed by the following formula: ,in, This represents the number of query keywords matched by the k-th knowledge fragment; This indicates the keyword hit rate, excluding zero decimals. This indicates the number of keywords in the search query.

[0011] Furthermore, in step S40, the specific method for selecting the main reference fragment and the secondary reference fragment from the candidate knowledge fragments is as follows: Sort by search score in descending order, and take the segment corresponding to the highest score as the main reference segment; Select several segments that are secondary reference segments and are related to the current work order in terms of fault code, model, or phenomenon, and retrieve them in order of priority to the main reference segment. When multiple high-scoring segments exist, the segment with the most complete fault codes is selected first. If the scores are still the same, the segment with the more recent shutdown time is selected first.

[0012] Furthermore, in step S50, the expression for calculating the original text citation coverage is: ; In the formula, Indicates the original text citation coverage; This indicates the number of steps in the distributed processing steps that include the original source. This indicates the number of distributed processing steps obtained from the assembly; This indicates the original text citation coverage, excluding zero decimals.

[0013] Furthermore, the expression for calculating the retrieval confidence is as follows: ; In the formula, Indicates the confidence level of the retrieval; Indicates the highest retrieval score weight; This indicates the retrieval score of the selected main reference fragment; This indicates the weight of the original text citation coverage.

[0014] Furthermore, in step S60, the preset threshold includes an upper threshold. and lower threshold ,when When, push standard operating procedures; when When, push out step-by-step processing tasks and mark references; when When this happens, it is marked as a difficult example and only reference steps are pushed.

[0015] An enhanced retrieval-based maintenance fault handling assistance system, improved in that it includes: The work order and file acquisition module is used to acquire the work order fields of on-site maintenance work orders and the equipment file of the target device corresponding to the work order; The query construction module is used to construct a retrieval query based on the work order field and equipment file. The retrieval query includes query text, keyword set and query vector. The dual-path retrieval module is used to perform dual-path retrieval of vector retrieval and keyword retrieval in the pre-built maintenance knowledge base, and calculate the retrieval score for each candidate knowledge fragment according to the fusion weight; The step-by-step assembly module is used to select the main reference fragment and the auxiliary reference fragment from the candidate knowledge fragments, extract the processing steps in sequence and associate the original text sources corresponding to each step to obtain the step-by-step processing job, and calculate the original text citation coverage rate. The confidence-based traffic splitting module is used to calculate the retrieval confidence based on the highest retrieval score and the original text citation coverage, and to push different levels of handling guidance to the on-site terminal based on the comparison result of the retrieval confidence with the preset threshold. When the retrieval confidence is lower than the preset first threshold, the work order is marked as a difficult case. The difficult case write-back module is used to respond to a work order that has been marked as a difficult case and whose on-site handling has been completed and passed the customs clearance acceptance. It collects the actual handling steps to generate new knowledge fragments, which are then written into the maintenance knowledge base after review.

[0016] An electronic device, improved in that it includes at least one processor and at least one memory, wherein, The memory stores computer-readable instructions; The computer-readable instructions are executed by one or more processors, enabling the electronic device to implement the retrieval-enhanced maintenance fault handling assistance method described above.

[0017] A storage medium storing computer-readable instructions, wherein the improvement is that the computer-readable instructions are executed by one or more processors to implement the retrieval-enhanced operation and maintenance fault handling assistance method as described above.

[0018] The beneficial effects of this invention are as follows: Firstly, by acquiring the work order fields of on-site maintenance work orders and the equipment files of the target equipment corresponding to those work orders, and constructing a retrieval query based on the work order fields and equipment files—the retrieval query including query text, keyword set, and query vector—a deep understanding of the work order context is achieved, avoiding the inefficiency of simply relying on manual document retrieval. Secondly, by performing dual-path retrieval of vector retrieval and keyword retrieval in a pre-built maintenance knowledge base, and calculating retrieval scores for each candidate knowledge fragment according to fusion weights, the advantages of semantic matching and exact matching are combined, effectively improving retrieval accuracy and overcoming the limitations of a single retrieval mode. Thirdly, by implementing step-by-step processing... The invention calculates the original text citation coverage rate by including the proportion of steps with original text sources in the work process. It then calculates the retrieval confidence level based on the highest retrieval score and the original text citation coverage rate. Based on the comparison between the retrieval confidence level and a preset threshold, it pushes different levels of handling guidance to the field terminal, achieving quantitative reliability assessment and tiered push of retrieval results to help maintenance personnel judge the credibility of information. Finally, in response to work orders marked as difficult cases that have been completed on-site and passed customs clearance, the actual handling steps are collected to generate new knowledge fragments. After review, these fragments are written into the maintenance knowledge base, forming a closed-loop knowledge update mechanism. This allows difficult case experience to be accumulated and reused, fundamentally solving the problem of difficult knowledge reuse. Therefore, this invention can significantly improve the efficiency, accuracy, and knowledge reuse level of maintenance fault handling. Attached Figure Description

[0019] Figure 1 This is an overall flowchart of a search-enhanced operation and maintenance fault handling auxiliary method according to the present invention; Figure 2 This is a block diagram of a search-enhanced operation and maintenance fault handling auxiliary system according to the present invention; Figure 3 A hardware structure diagram of an electronic device is shown as an exemplary embodiment; Figure 4 A block diagram illustrating an electronic device as an exemplary embodiment. Detailed Implementation

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0021] The following will clearly and completely describe the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the scope of protection of the present invention. Furthermore, all connections / linkages involved in the patent do not simply refer to direct contact between components, but rather to the ability to form a better connection structure by adding or reducing connecting accessories according to specific implementation conditions. The various technical features in this invention can be combined interactively without contradicting each other.

[0022] Reference Figure 1 As shown, the present invention provides a retrieval-enhanced operation and maintenance fault handling assistance method, which includes the following steps: S10, obtain the work order field of the on-site maintenance work order and the equipment file of the target device corresponding to the work order; It should be noted that in this embodiment, on-site maintenance work orders are dispatched from the service desk to the on-site engineer's APP. Work order fields include at least the work order number, a description of the problem, a fault code, and the work order location. Equipment files are obtained from the one-item-one-file system and include asset identifier, category, model, alias, installation location, and historical fault records. Work order fields and equipment files together form the basic input for subsequent searches and queries. The system obtains the above data in real time through an API interface to ensure the timeliness of information. When equipment files are missing (e.g., new equipment has not yet been filed), only the work order fields are used to construct the query, but guessing the model is not allowed to avoid search bias.

[0023] S20, construct a retrieval query based on the work order field and equipment file, the retrieval query including query text, keyword set and query vector; It should be noted that, in this embodiment, the specific method of constructing the query is as follows: The query text is obtained by concatenating the work order phenomenon description, fault code, work order location, and the category, model, and alias from the equipment file according to fixed fields. For example, if the work order phenomenon is "screen flickering," the fault code is "IMG-BLUR," the location is "East Gate," and the equipment file lists the category as "gun-type" and the model as "E12," then the concatenated query text would be: "East Gate Gun-type E12 Screen Flickering IMG-BLUR."

[0024] The fault codes, equipment models, product categories, and related terms are extracted from the query text to form a keyword set. The number of keywords is denoted as [missing information]. For example, the keyword set is: {IMG-BLUR, E12, bolt action, screen flickering}.

[0025] The query text is encoded using an encoder that shares the same origin as the knowledge base fragment vectors (such as the same Sentence-BERT model), resulting in the query vector q. The encoder is identical to the model used in the offline pre-encoding on the knowledge base side to ensure consistency of the vector space. Searching solely by phenomenon or solely by model number does not constitute a complete query construction.

[0026] S30, perform dual-path retrieval of vector retrieval and keyword retrieval in the pre-built maintenance knowledge base, and calculate the retrieval score for each candidate knowledge fragment according to the fusion weight; It should be noted that in this embodiment, the maintenance knowledge base pre-stores fragments of historical shutdown work orders, as well as paragraphs from manuals and work instructions. Each knowledge fragment includes the phenomenon, model number, fault code, step-by-step text, and the source work order number or manual section. The knowledge base encodes each fragment offline into a vector. An inverted index is created for fault codes, model numbers, and symptom terms. New fragments written back from difficult examples are also added to this knowledge base, and their encoding vectors are synchronized and an inverted index is created.

[0027] During the dual-path retrieval process, vector retrieval requires calculating the query vector q and the knowledge base fragment vector. The cosine similarity is calculated as follows: In the formula, This represents the similarity between the query vector and the k-th knowledge fragment; Represents the query vector q and the knowledge base fragment vector The inner product, i.e. ; This represents the L2 norm of the query vector q, i.e. ; Represents knowledge base fragment vectors The 2-norm, i.e. ; Vector similarity is calculated by removing zero decimals; while keyword retrieval requires calculating keyword hit rate, the formula for which is: In the formula, This represents the keyword hit rate of the k-th knowledge fragment; This represents the number of query keywords matched by the k-th knowledge fragment; This indicates the keyword hit rate, excluding zero decimals. This indicates the number of keywords in the search query.

[0028] The union of the two search results yields a candidate set K. For each segment in K, a search score is calculated based on the fusion weight, expressed as: In the formula, This represents the retrieval score of the k-th knowledge fragment; This indicates the weight of vector similarity in the retrieval score; The value range is [0,1]. If set K is empty, the retrieval confidence is 0, and the process jumps directly to the difficult case file in step S60.

[0029] S40: Select the main reference fragment and the auxiliary reference fragment from the candidate knowledge fragments, extract the processing steps in sequence and associate the original text sources corresponding to each step to obtain the step-by-step processing operation; It should be noted that, in this embodiment, the main reference fragment is selected as the fragment with the highest retrieval score. If multiple segments have the highest scores, the segment that matches the fault code is selected first; if they are still the same, the segment with the more recent shutdown time is selected first.

[0030] The auxiliary reference fragments are selected from several fragments that are ranked after the main reference fragment and are related to the current work order in terms of fault code, model, or phenomenon. The auxiliary reference fragments are used to supplement any steps that may be missing from the main reference fragment or to provide alternative solutions.

[0031] From the main and secondary cited fragments, extract the processing steps in their original order. Steps with similar semantics are merged into one step, resulting in a step-by-step processing task. Each step must include the original source: the source fragment number and the original sentence. Steps that cannot be linked to the original source will not be included in the original source citation coverage and should not be displayed as standard work.

[0032] S50, calculate the original text citation coverage rate based on the proportion of steps with original text sources in the step-by-step processing operation, and calculate the retrieval confidence based on the highest retrieval score and the original text citation coverage rate; It should be noted that, in this embodiment, the expression for calculating the original text citation coverage is: ; In the formula, Indicates the original text citation coverage; This indicates the number of steps in the distributed processing steps that include the original source. This indicates the number of distributed processing steps obtained from the assembly; This indicates the original text citation coverage, excluding zero decimals. The value range is [0,1]. The lower the value, the greater the likelihood that the steps will not return to the original text.

[0033] The formula for calculating the retrieval confidence score is: ; In the formula, Indicates the confidence level of the retrieval; Indicates the highest retrieval score weight; This indicates the retrieval score of the selected main reference fragment; This indicates the weight of the original text citation coverage.

[0034] S60, based on the comparison result between the search confidence level and the preset threshold, push different levels of handling instructions to the on-site terminal. Among them, when the search confidence level is lower than the preset first threshold, the work order is marked as a difficult case. It should be noted that, in this embodiment, the preset threshold includes an upper threshold. and lower threshold ,when When the standard operating procedure is pushed out, the APP processes the auxiliary tasks step by step, and the steps can be regarded as the recommended order of tasks for this order; when When a step-by-step handling task is pushed out and marked with references, the interface indicates the nature of the reference, and on-site engineers can actively click the "Difficult Case" button; when When a case is marked as difficult, only reference steps will be provided, and the actual operation steps must be filled in when closing the order. Field engineers can also actively click the "Difficult Case" button to process the case according to the difficult case procedure, regardless of the C rating.

[0035] S70, in response to a work order being marked as a difficult case and having been handled on-site and passed customs clearance, the actual handling steps are collected to generate a new knowledge fragment, which is then written into the maintenance knowledge base after review.

[0036] It should be noted that in this embodiment, the difficult case write-back is only performed when all of the following conditions are met: the work order has been marked as a difficult case (or a difficult case is actively identified on-site), the work order has been handled on-site, and it has been accepted by the foreman or knowledge administrator. The write-back content includes at least: phenomenon, fault code, model, practical steps (listed in sequence), and on-site records (text or photo descriptions). The system assembles this information into a new knowledge fragment, writes it into the maintenance knowledge base after review, and encodes it with a vector and establishes an inverted index.

[0037] If the write-back content lacks steps or model numbers, the segment will not be entered into the database but will only be retained in the work order record to prevent dirty data from being hit by subsequent searches. Regular high-confidence closing orders are not subject to mandatory write-back to avoid the same type of steps repeatedly occupying knowledge base space. After a successful write-back, when similar work orders pass through S20-S40 again, they should be able to find this new segment, thus achieving closed-loop reuse of experience.

[0038] Reference Figure 2 As shown, the present invention also discloses a retrieval-enhanced operation and maintenance fault handling auxiliary system, comprising: The work order and file acquisition module 601 is used to acquire the work order field of the on-site maintenance work order and the equipment file of the target device corresponding to the work order; The query construction module 602 is used to construct a retrieval query based on the work order field and the equipment file. The retrieval query includes query text, keyword set and query vector. The dual-path retrieval module 603 is used to perform dual-path retrieval of vector retrieval and keyword retrieval in the pre-built maintenance knowledge base, and calculate the retrieval score for each candidate knowledge fragment according to the fusion weight; Step-by-step assembly module 604 is used to select main reference fragments and auxiliary reference fragments from candidate knowledge fragments, extract processing steps in sequence and associate the original text sources corresponding to each step to obtain step-by-step processing operations, and calculate the original text citation coverage rate. The confidence-based traffic splitting module 605 is used to calculate the retrieval confidence based on the highest retrieval score and the original text citation coverage, and to push different levels of handling guidance to the field terminal based on the comparison result of the retrieval confidence with the preset threshold. When the retrieval confidence is lower than the preset first threshold, the work order is marked as a difficult case. The difficult case write-back module 606 is used to respond to a work order that has been marked as a difficult case and whose on-site handling has been completed and passed the customs clearance acceptance. It collects the actual handling steps to generate new knowledge fragments, which are then written into the maintenance knowledge base after review.

[0039] It should be noted that the search-enhanced operation and maintenance fault handling assistance strategy provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed. That is, the internal structure of the search-enhanced operation and maintenance fault handling assistance system 600 will be divided into different functional modules to complete all or part of the functions described above. Furthermore, the search-enhanced operation and maintenance fault handling assistance system 600 and the search-enhanced operation and maintenance fault handling assistance method provided in the above embodiments belong to the same concept. The specific way each module performs operations has been described in detail in the method embodiments, and will not be repeated here.

[0040] Figure 3 A schematic diagram of the structure of an electronic device according to an exemplary embodiment is shown.

[0041] It should be noted that this electronic device is merely an example adapted to the present invention and should not be construed as providing any limitation on the scope of use of the present invention. Furthermore, this electronic device should not be interpreted as requiring or depending on having... Figure 3 One or more components of the exemplary electronic device 2000 shown.

[0042] The hardware structure of electronic devices 2000 can vary significantly due to differences in configuration or performance, such as... Figure 3 As shown, the electronic device 2000 includes: a power supply 210, an interface 230, at least one memory 250, and at least one central processing unit (CPU) 270.

[0043] Specifically, power supply 210 is used to provide operating voltage for various hardware devices on electronic device 2000.

[0044] Interface 230 includes at least one wired or wireless network interface 231 for interacting with external devices. Of course, in other examples adapted to this invention, interface 230 may further include at least one serial-to-parallel conversion interface 233, at least one input / output interface 235, and at least one USB interface 237, etc. Figure 3 As shown, this does not constitute a specific limitation.

[0045] The memory 250 serves as a carrier for resource storage and can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored on it include the operating system 251, application programs 253, and data 255, etc., and the storage method can be temporary storage or permanent storage.

[0046] The operating system 251 is used to manage and control the various hardware devices and application programs 253 on the electronic device 2000, so as to enable the central processing unit 270 to perform calculations and processing on the massive data 255 in the memory 250. It can be Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0047] Application 253 is a computer-readable instruction based on operating system 251 that performs at least one specific task, and may include at least one module ( Figure 3 (Not shown), each module may contain computer-readable instructions for the electronic device 2000. For example, the enhanced maintenance troubleshooting assistance device can be considered as application 253 deployed on the electronic device 2000.

[0048] Data 255 may be signal information, etc., and is stored in memory 250.

[0049] The central processing unit 270 may include one or more processors and is configured to communicate with the memory 250 via at least one communication bus to read computer-readable instructions stored in the memory 250, thereby enabling the computation and processing of massive amounts of data 255 in the memory 250. For example, an enhanced maintenance fault handling assistance method can be implemented by the central processing unit 270 reading a series of computer-readable instructions stored in the memory 250.

[0050] Furthermore, the present invention can also be implemented through hardware circuits or a combination of hardware circuits and software. Therefore, the implementation of the present invention is not limited to any specific hardware circuit, software, or combination thereof.

[0051] Please see Figure 4This invention provides an electronic device 4000, which may include: a desktop computer, a laptop computer, a server, etc., with sensor recognition capabilities.

[0052] exist Figure 4 In this context, the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.

[0053] The data interaction between the processor 4001 and the memory 4003 can be achieved through at least one communication bus 4002. This communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0054] Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0055] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0056] The memory 4003 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program instructions or code in the form of instructions or data structures and accessible by the electronic device 4000, but not limited thereto.

[0057] The memory 4003 stores computer-readable instructions, and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002.

[0058] The computer-readable instructions are executed by one or more processors 4001 to implement the retrieval-enhanced operation and maintenance fault handling assistance method in the above embodiments.

[0059] Furthermore, this embodiment of the invention provides a storage medium storing computer-readable instructions, which are executed by one or more processors to implement the retrieval-enhanced operation and maintenance fault handling assistance method described above.

[0060] Compared with existing technologies, the beneficial effects of this invention are as follows: By acquiring the work order fields of the on-site maintenance work order and the equipment file of the target equipment corresponding to the work order, and constructing a retrieval query based on the work order fields and equipment file, the retrieval query includes query text, keyword set, and query vector, achieving a deep understanding of the work order context and avoiding the inefficiency of simply relying on manual document retrieval; secondly, by performing dual-path retrieval of vector retrieval and keyword retrieval in a pre-built maintenance knowledge base, and calculating the retrieval score for each candidate knowledge fragment according to the fusion weight, the advantages of semantic matching and exact matching are combined, effectively improving retrieval accuracy and overcoming the limitations of a single retrieval mode; then, by proceeding step by step... The invention calculates the original text citation coverage rate by considering the proportion of steps with original text sources in the work process. It then calculates the retrieval confidence rate based on the highest retrieval score and the original text citation coverage rate. Based on the comparison between the retrieval confidence rate and a preset threshold, it pushes different levels of handling guidance to the field terminal, achieving quantitative reliability assessment and tiered push of retrieval results to help maintenance personnel judge the credibility of information. Finally, in response to work orders marked as difficult cases that have been completed on-site and passed customs clearance, the actual handling steps are collected to generate new knowledge fragments. After review, these fragments are written into the maintenance knowledge base, forming a closed-loop knowledge update mechanism. This allows difficult case experience to be accumulated and reused, fundamentally solving the problem of difficult knowledge reuse. Therefore, this invention can significantly improve the efficiency, accuracy, and knowledge reuse level of maintenance fault handling.

[0061] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A search-enhanced auxiliary method for handling operational and maintenance faults, characterized in that, The enhanced retrieval method for handling operational and maintenance faults includes the following steps: S10, obtain the work order field of the on-site maintenance work order and the equipment file of the target device corresponding to the work order; S20, construct a retrieval query based on the work order field and equipment file, the retrieval query including query text, keyword set and query vector; S30, perform dual-path retrieval of vector retrieval and keyword retrieval in the pre-built maintenance knowledge base, and calculate the retrieval score for each candidate knowledge fragment according to the fusion weight; S40: Select the main reference fragment and the auxiliary reference fragment from the candidate knowledge fragments, extract the processing steps in sequence and associate the original text sources corresponding to each step to obtain the step-by-step processing operation; S50, calculate the original text citation coverage rate based on the proportion of steps with original text sources in the step-by-step processing operation, and calculate the retrieval confidence based on the highest retrieval score and the original text citation coverage rate; S60, based on the comparison result between the search confidence level and the preset threshold, push different levels of handling instructions to the on-site terminal. Among them, when the search confidence level is lower than the preset first threshold, the work order is marked as a difficult case. S70, in response to a work order being marked as a difficult case and having been handled on-site and passed customs clearance, the actual handling steps are collected to generate a new knowledge fragment, which is then written into the maintenance knowledge base after review.

2. The enhanced retrieval method for handling maintenance faults according to claim 1, characterized in that, In step S20, the specific method for constructing the retrieval query based on the work order field and equipment file is as follows: The query text is obtained by combining the work order description, fault code, work order location, and category, model, and alias from the equipment file using fixed fields. Extract fault codes, equipment models, product categories, and phenomenon terms from the query text to form a keyword set; The query vector is obtained by encoding the query text using an encoder that is of the same origin as the fragment vector in the knowledge base.

3. The enhanced retrieval method for handling maintenance faults according to claim 1, characterized in that, In step S30, the calculation expression for the retrieval score is: In the formula, This represents the retrieval score of the k-th knowledge fragment; This indicates the weight of vector similarity in the retrieval score; The similarity between the query vector and the k-th knowledge fragment is expressed as follows: ,in, This represents the L2 norm of the query vector q; This represents the vector of the k-th knowledge fragment in the maintenance knowledge base. The second norm; Represents the query vector q and the knowledge base fragment vector The inner product; Vector similarity is delimited to prevent division by zero. The keyword hit rate of the k-th knowledge fragment is expressed by the following formula: ,in, This represents the number of query keywords matched by the k-th knowledge fragment; This indicates the keyword hit rate, excluding zero decimals. This indicates the number of keywords in the search query.

4. The enhanced retrieval method for handling maintenance faults according to claim 1, characterized in that, In step S40, the specific method for selecting the main reference fragment and the auxiliary reference fragment from the candidate knowledge fragments is as follows: Sort by search score in descending order, and take the segment corresponding to the highest score as the main reference segment; Select several segments that are secondary reference segments and are related to the current work order in terms of fault code, model, or phenomenon, and retrieve them in order of priority to the main reference segment. When multiple high-scoring segments exist, the segment with the most complete fault codes is selected first. If the scores are still the same, the segment with the more recent shutdown time is selected first.

5. The enhanced retrieval method for handling maintenance faults according to claim 1, characterized in that, In step S50, the expression for calculating the original text citation coverage is: ; In the formula, Indicates the original text citation coverage; This indicates the number of steps in the distributed processing steps that include the original source. This indicates the number of distributed processing steps obtained from the assembly; This indicates the original text citation coverage, excluding zero decimals.

6. The enhanced retrieval method for handling maintenance faults according to claim 5, characterized in that, The formula for calculating the retrieval confidence score is: ; In the formula, Indicates the confidence level of the retrieval; Indicates the highest retrieval score weight; This indicates the retrieval score of the selected main reference fragment; This indicates the weight of the original text citation coverage.

7. The enhanced retrieval method for handling maintenance faults according to claim 6, characterized in that, In step S60, the preset threshold includes an upper threshold. and lower threshold ,when When, push standard operating procedures; when When, push out step-by-step processing tasks and mark references; when When this happens, it is marked as a difficult example and only reference steps are pushed.

8. A search-enhanced operation and maintenance fault handling auxiliary system, characterized in that, include: The work order and file acquisition module is used to acquire the work order fields of on-site maintenance work orders and the equipment file of the target device corresponding to the work order; The query construction module is used to construct a retrieval query based on the work order field and equipment file. The retrieval query includes query text, keyword set and query vector. The dual-path retrieval module is used to perform dual-path retrieval of vector retrieval and keyword retrieval in the pre-built maintenance knowledge base, and calculate the retrieval score for each candidate knowledge fragment according to the fusion weight; The step-by-step assembly module is used to select the main reference fragment and the auxiliary reference fragment from the candidate knowledge fragments, extract the processing steps in sequence and associate the original text sources corresponding to each step to obtain the step-by-step processing job, and calculate the original text citation coverage rate. The confidence-based traffic splitting module is used to calculate the retrieval confidence based on the highest retrieval score and the original text citation coverage, and to push different levels of handling guidance to the on-site terminal based on the comparison result of the retrieval confidence with the preset threshold. When the retrieval confidence is lower than the preset first threshold, the work order is marked as a difficult case. The difficult case write-back module is used to respond to a work order that has been marked as a difficult case and whose on-site handling has been completed and passed the customs clearance acceptance. It collects the actual handling steps to generate new knowledge fragments, which are then written into the maintenance knowledge base after review.

9. An electronic device, characterized in that, Includes at least one processor and at least one memory, wherein, The memory stores computer-readable instructions; The computer-readable instructions are executed by one or more processors, causing the electronic device to implement the retrieval-enhanced operation and maintenance fault handling assistance method as described in any one of claims 1-7.

10. A storage medium having computer-readable instructions stored thereon, characterized in that, The computer-readable instructions are executed by one or more processors to implement the retrieval-enhanced operation and maintenance fault handling assistance method according to any one of claims 1-7.