Knowledge Graph-Based Fault Tracing and Maintenance Decision System and Method for Printing Equipment

By acquiring the operating data and historical fault data of printing equipment, updating the relationship confidence of the knowledge graph, and constructing a fault tracing knowledge graph, the problem of inaccurate fault tracing of printing equipment is solved, and the accuracy and effectiveness of fault identification are improved.

CN120348065BActive Publication Date: 2026-11-17SHENZHEN JINFENG PACKAGING TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510643365.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-19
Publication Date
2026-11-17
Estimated Expiration
2045-05-19

AI Technical Summary

Technical Problem

In existing technologies, when using general knowledge graphs for printing equipment fault identification, the functional differences between different models of printing equipment lead to inaccurate fault tracing results.

Method used

By acquiring the operating data and historical fault data of printing equipment, the relationship confidence in the fault tracing knowledge graph is updated. The initial node is used to search for relationship chains, determine the target relationship chain and the cause of the fault, and combine maintenance data to construct a fault tracing knowledge graph to improve the accuracy of fault tracing.

Benefits of technology

This technology enables the knowledge graph to be adjusted according to the actual working status of the printing equipment, thereby improving the accuracy of fault identification and ensuring the accuracy and effectiveness of fault tracing results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120348065B_ABST
    Figure CN120348065B_ABST
Patent Text Reader

Abstract

This invention discloses a knowledge graph-based fault tracing and maintenance decision-making system and method for printing equipment, belonging to the field of printing equipment technology. This invention acquires the operating data, historical fault data, and current fault information of the printing equipment; updates the confidence levels of relationships in the fault tracing knowledge graph based on the operating data and historical fault data; and determines the cause of the fault based on the fault information and the fault tracing knowledge graph, thereby improving the accuracy of fault tracing results for printing equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of printing equipment, and in particular to a knowledge graph-based system and method for tracing and repairing printing equipment faults. Background Technology

[0002] The electrical control systems and mechanical components of printing equipment are generally quite complex. The realization of a single function often requires accurate electrical control based on high-precision mechanical components to achieve accurate printing. Currently, in the process of using general knowledge graphs for fault identification, some similar models of printing equipment have slight differences in function, resulting in hardware that is not completely identical. Therefore, the use of knowledge graphs cannot completely correspond to the printing equipment, which leads to inaccurate results in tracing faults.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a knowledge graph-based system and method for tracing and repairing printing equipment faults, aiming to improve the accuracy of fault tracing results for printing equipment.

[0005] To achieve the above objectives, this invention provides a knowledge graph-based method for tracing and repairing printing equipment faults, comprising the following steps:

[0006] Acquire the printing equipment's operating data, historical fault data, and current fault status information;

[0007] The confidence level of the relationships in the fault tracing knowledge graph is updated based on the operational data and the historical fault data.

[0008] The cause of the fault is determined based on the fault information and the fault tracing knowledge graph.

[0009] Optionally, the step of determining the cause of the fault based on the fault information and the fault tracing knowledge graph includes:

[0010] Based on the fault information, an initial node is determined in the fault tracing knowledge graph, and a relationship chain is searched based on the initial node to obtain multiple relationship chains;

[0011] The target relation chain is determined based on the multiple relation chains and the confidence level of each relation, and the cause of the failure is determined based on the target relation chain.

[0012] Optionally, the step of updating the confidence of relationships in the fault tracing knowledge graph based on the operational data and the historical fault data further includes:

[0013] Obtain maintenance data of printing equipment, and extract multiple triplet data based on the maintenance data. The structure of the triplet data is: first object, relation, second object;

[0014] A fault tracing knowledge graph is constructed based on multiple triplet data sets.

[0015] Optionally, the historical fault data includes: a historical fault relationship chain, which is a fault relationship chain determined before the current time based on fault prompts in the fault tracing knowledge graph. The step of updating the confidence of the relationships in the fault tracing knowledge graph according to the operational data and the historical fault data includes:

[0016] Determine the operational relationship chain based on the operational data;

[0017] Based on the historical fault relationship chain and the operational relationship chain, a first type of relationship and a second type of relationship are determined in the fault tracing knowledge graph. The first type of relationship is a relationship belonging to the historical fault relationship chain or a relationship belonging to the operational relationship chain. The second type of relationship is a relationship other than the first type of relationship in the fault tracing knowledge graph.

[0018] The weight value corresponding to each first type of relationship is determined based on the historical fault relationship chain and the operation relationship chain, and the weight value of the second type of relationship is set to a preset value.

[0019] The confidence level of each relation is determined based on all relations and their corresponding weight values.

[0020] Optionally, the step of determining the confidence level of each relation based on all relations and their corresponding weight values ​​includes:

[0021] Based on the fault tracing knowledge graph, at least one target triplet data is determined for each target object, wherein the target triplet data is triplet data with the target object as the first object;

[0022] Calculate the target confidence level of each target relation on the target object based on the target relations and corresponding target weight data of all the target triplet data;

[0023] The target confidence level is used as the confidence level.

[0024] Optionally, the operational data includes: the operational functions and corresponding related components, and the operational time; the step of determining the operational relationship chain based on the operational data includes:

[0025] Based on the relevant components, multiple related objects in the fault tracing knowledge graph are identified;

[0026] The operational relationship chain is determined based on the fault tracing knowledge graph and multiple related objects.

[0027] Optionally, the step of determining the operational relationship chain based on the fault tracing knowledge graph and multiple related objects includes:

[0028] Based on the fault tracing knowledge graph and the multiple related objects, calculate the first shortest path between any two related objects to obtain multiple first shortest paths;

[0029] The running relationship chain is determined based on multiple first shortest paths and the Held-Karp algorithm.

[0030] Optionally, the step of determining the target relation chain based on the plurality of relation chains and the confidence level of each relation includes:

[0031] The probability of each relation chain is determined based on a preset calculation formula, the confidence level of each relation, and each relation chain.

[0032] The target relation chain is determined based on the relation probability of each relation chain.

[0033] Optionally, after the step of determining the cause of the fault based on the fault condition information and the fault tracing knowledge graph, the method further includes:

[0034] A corresponding maintenance decision is generated based on the cause of the fault.

[0035] Furthermore, to achieve the above objectives, the present invention also provides a knowledge graph-based printing equipment fault tracing and maintenance decision-making system, the knowledge graph-based printing equipment fault tracing and maintenance decision-making system comprising:

[0036] The acquisition module is used to acquire the printing equipment's operating data, historical fault data, and current fault information.

[0037] An update module is used to update the confidence level of relationships in the fault tracing knowledge graph based on the operational data and the historical fault data.

[0038] The identification module is used to determine the cause of the fault based on the fault information and the fault tracing knowledge graph.

[0039] This invention proposes a knowledge graph-based method for fault tracing and maintenance decision-making in printing equipment. This method acquires the printing equipment's operational data, historical fault data, and current fault status information. It then updates the confidence levels of relationships in the fault tracing knowledge graph based on the operational and historical fault data. This allows for adjustments to the knowledge graph to suit the actual operating state of each printing device, resulting in a knowledge graph directly related to the actual operation of the printing equipment. This improves the accuracy of the knowledge graph reasoning process and, consequently, the accuracy of fault identification. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the structure of a printing device in the hardware operating environment involved in the embodiments of the present invention;

[0041] Figure 2 This is a flowchart illustrating the first embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention.

[0042] Figure 3 This is a flowchart illustrating the second embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention.

[0043] Figure 4 This is a flowchart illustrating the third embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention.

[0044] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0045] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0046] Reference Figure 1 , Figure 1 This is a schematic diagram of the printing equipment structure of the hardware operating environment involved in the embodiments of the present invention.

[0047] like Figure 1As shown, the printing equipment may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, an interactive device 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The interactive device 1003 may include a display screen and an input unit such as a keyboard. Optionally, the interactive device 1003 may also be connected to the communication bus via a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0048] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the printing equipment and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0049] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and a knowledge graph-based printing equipment fault tracing and maintenance decision-making program.

[0050] exist Figure 1 In the printing equipment shown, the network interface 1004 is mainly used for data communication with other devices; the interactive device 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the printing equipment of the present invention can be set in the printing equipment. The printing equipment calls the knowledge graph-based printing equipment fault tracing and maintenance decision program stored in the memory 1005 through the processor 1001, and executes the knowledge graph-based printing equipment fault tracing and maintenance decision method provided in the embodiment of the present invention.

[0051] This invention provides a knowledge graph-based method for fault tracing and maintenance decision-making in printing equipment, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of a knowledge graph-based method for tracing and making maintenance decisions for printing equipment.

[0052] In this embodiment, the knowledge graph-based printing equipment fault tracing and maintenance decision-making method includes:

[0053] Step S1: Obtain the printing equipment's operating data, historical fault data, and current fault information;

[0054] The fault information here can be machine-generated prompts or actual problem information identified by printing equipment administrators after a fault is discovered. This is achieved by recording the printing equipment's operation logs and extracting data related to printer operation from these logs. This operation data does not include user personal privacy data. Optionally, after a fault occurs, maintenance personnel record maintenance logs and determine the historical fault data based on these logs. Alternatively, fault data or anomaly data recorded by the equipment can also be used as the historical fault data.

[0055] Step S2: Update the confidence of the relationships in the fault tracing knowledge graph based on the operational data and the historical fault data;

[0056] It is important to clarify that updating the confidence level of relationships in the fault tracing knowledge graph based on the operational data and historical fault data does not add new data to the fault tracing knowledge graph; that is, it does not add triplet data to the fault tracing knowledge graph, i.e., it does not add objects or relationships to the fault tracing knowledge graph. Each relationship in the fault tracing knowledge graph has a set confidence level, reflecting the reliability of that relationship for the two connected objects. Furthermore, in this field, the data in triplet data can also be called head entity, relation, and tail entity. However, in this embodiment, the first and second objects can be hardware entities or concepts outside of entities. For example, printing process, ink state, paper parameters, fault type, etc. Here, the relation is the edge connecting two points in the knowledge graph. Updating the confidence level of relationships in the fault tracing knowledge graph using the operational data and historical fault data is actually dynamically adjusting the knowledge graph based on the current printing equipment. For example, when printing equipment operates in a single mode for an extended period, the confidence level of relevant relationships in the fault tracing knowledge graph can be adjusted to improve the accuracy of subsequent fault detection using the knowledge graph.

[0057] Step S3: Determine the cause of the fault based on the fault information and the fault tracing knowledge graph.

[0058] In this embodiment, optionally, an initial node in the fault tracing knowledge graph is determined based on the fault information, and a relationship chain is searched based on the initial node to obtain multiple relationship chains; a target relationship chain is determined based on the multiple relationship chains and the confidence level of each relationship, and the cause of the fault is determined based on the target relationship chain. The algorithm for determining the cause of the fault in the fault tracing knowledge graph based on the fault information is not limited; for example, the Dijkstra algorithm or the A* algorithm can be used to determine the target relationship chain, and the cause of the fault can be identified based on the target relationship chain.

[0059] In this embodiment, by acquiring the operating data, historical fault data, and current fault information of the printing equipment, and updating the confidence of the relationships in the fault tracing knowledge graph based on the operating data and historical fault data, the knowledge graph can be adjusted for the actual working state of each printing equipment, thereby obtaining a knowledge graph that is directly related to the actual printing equipment, thus improving the accuracy of the knowledge graph reasoning process and improving the accuracy of fault identification.

[0060] Furthermore, based on the first embodiment, a second embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention is proposed. In this embodiment, referring to... Figure 3 Before the step of updating the confidence of relationships in the fault tracing knowledge graph based on the operational data and the historical fault data, the method further includes:

[0061] Step S201: Obtain maintenance data of the printing equipment, and extract multiple triplet data based on the maintenance data. The structure of the triplet data is: first object, relation, second object;

[0062] Specifically, the process involves obtaining relevant materials such as instruction manuals and maintenance manuals for printing equipment, extracting triplet data based on these materials, extracting the text information from the manuals and maintenance manuals, removing noise information such as page numbers from the text, identifying entity objects after cleaning, and extracting relationships according to preset rules after obtaining the entity objects. Regular expressions can be used for extraction to obtain triplet data of the first object, relationship, and second object.

[0063] Step S202: Construct a fault tracing knowledge graph based on the multiple triplet data.

[0064] Standardizing objects within each triple involves aligning different objects expressing the same meaning. This is done by recording the similarity between objects using a string similarity algorithm, and grouping objects with similarity scores within a certain range into a single object. Similarly, relationships need to be standardized. For example, error codes, fault codes, and alarm codes should be standardized; print settings, print options, and print parameters should also be standardized.

[0065] In this embodiment, a fault tracing knowledge graph is constructed using the maintenance data of the printing equipment, thereby obtaining a complete knowledge graph that can be used to trace faults.

[0066] Furthermore, based on the first or second embodiment, a third embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention is proposed. In this embodiment, referring to... Figure 4 The historical fault data includes: historical fault relationship chains, which are fault relationship chains determined before the current time based on fault prompts in the fault tracing knowledge graph. The step of updating the confidence of the relationships in the fault tracing knowledge graph according to the operational data and the historical fault data includes:

[0067] Step S21: Determine the operational relationship chain based on the operational data;

[0068] The operational data here can involve the hardware or virtual concepts of one or more printing devices, and the objects whose states change are determined based on the operational data, such as changes in ink content, printhead position, print settings, drum assembly temperature, and power supply temperature. There can be multiple operational chains, and there is no restriction on whether these chains need to be connected.

[0069] Step S22: Determine the first type of relationship and the second type of relationship in the fault tracing knowledge graph based on the historical fault relationship chain and the operational relationship chain. The first type of relationship is a relationship belonging to the historical fault relationship chain or a relationship belonging to the operational relationship chain. The second type of relationship is a relationship other than the first type of relationship in the fault tracing knowledge graph.

[0070] The historical fault relationship chain here is a relationship chain generated based on the fault cause determined in the fault tracing knowledge graph at the current moment. It can include only the fault tracing of the current device, or it can include the historical fault relationship chain of printing devices of the same model. Here, after distinguishing between the first type of relationship and the second type of relationship, their weight values ​​can be updated according to different methods.

[0071] Step S23: Determine the weight value corresponding to each of the first type of relationship based on the historical fault relationship chain and the operation relationship chain, and set the weight value of the second type of relationship to a preset value;

[0072] The number of times each of the first type of relationship appears in the historical fault relationship chain and the running relationship chain is counted, and the corresponding weight value is determined based on the statistical results;

[0073] Optionally, the formula for calculating the weight values ​​is:

[0074] w i =αN(r i )+βM(r i )

[0075] w i Let r be the weight value of the i-th first-class relation. i For the i-th first-class relation, N(r) i ) represents the number of times the first-type relation of the i-th element appears in the historical fault relation chain, M(r) i Let be the number of times the i-th first-type relation appears in the running relation chain, and α and β be the coefficients corresponding to the historical fault relation chain and the running relation chain, respectively. The weight value of the second-type relation is set to a preset value. Preferably, this preset value is less than the weight value of any of the first-type relations.

[0076] Optionally, α and β can be determined by historical faults and operational conditions. Specifically, when the frequency of the same fault occurring on the same type of printing equipment increases, the corresponding α coefficient can increase accordingly. Furthermore, in special cases, the α coefficient can be negative. For example, if a fault causes some components to be removed during repair, the α coefficient corresponding to this historical fault relationship chain will be negative. For printing equipment, some models, due to minor improvements, do not actually have separate instruction manuals. This results in not all relationships in the fault tracing knowledge graph corresponding to the current printing equipment. For example, some hardware optimizations, driver updates, and changes in the cooling fan position settings may occur. The weight formula in this embodiment, determined based on operational data, can effectively reduce the fault determination errors caused by erroneous relationships in the fault tracing knowledge graph due to the aforementioned problems.

[0077] Furthermore, in some embodiments, different historical fault relationship chains correspond to different coefficients, i.e., α. j Let be the coefficient corresponding to the j-th historical fault relationship chain.

[0078] The confidence level of each relation is determined based on all relations and their corresponding weight values.

[0079] After obtaining the weight values, the weight data is not directly used as the confidence level of each relation in the fault tracing knowledge graph. Instead, the number of relations related to the object in the fault tracing knowledge graph needs to be adjusted to obtain the weight values.

[0080] In this embodiment, the weight value corresponding to each first type of relationship is determined by the historical fault relationship chain and the operation relationship chain, and the weight value of the second type of relationship is set to a preset value. This can effectively determine the relationship changes caused by operation or fault, thereby improving the accuracy of subsequent fault cause determination based on fault tracing knowledge graph.

[0081] Furthermore, the step of determining the confidence level of each relation based on all relations and their corresponding weight values ​​includes:

[0082] Based on the fault tracing knowledge graph, at least one target triplet data is determined for each target object, wherein the target triplet data is triplet data with the target object as the first object;

[0083] The target object here is any object in the obstacle tracing knowledge graph that can serve as the first object, i.e., the head entity. Objects that can only serve as the tail or second object cannot be the target object.

[0084] Calculate the target confidence level of each target relation on the target object based on the target relations and corresponding target weight data of all the target triplet data;

[0085] The target confidence level is used as the confidence level.

[0086] In this embodiment, each target object often corresponds to more than one relation. Therefore, it is necessary to determine the confidence level between the various relations of the target object based on all the target triplet data as a whole. The formula for calculating the confidence level is as follows:

[0087]

[0088] Specifically, P t w represents the target confidence of the target relation in the target triplet data. t denoted as target weight data, and T represents the number of target triplet data.

[0089] In this embodiment, at least one target triplet data associated with each target object is determined through the fault tracing knowledge graph. Based on the target relationships and corresponding target weight data of all the target triplet data, the target confidence score for each target relationship on the target object is calculated; this target confidence score is then used as the overall confidence score. This allows for updating the fault tracing knowledge graph through the historical fault relationship chain and the operational relationship chain, ensuring targeted corrections for each type of printing equipment, thereby more accurately identifying the corresponding fault for the anomaly.

[0090] Furthermore, based on the third embodiment, a fourth embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention is proposed. The operational data includes: the operational function and corresponding related components, and the operational time. The step of determining the operational relationship chain based on the operational data includes:

[0091] Based on the relevant components, multiple related objects in the fault tracing knowledge graph are identified;

[0092] The operational relationship chain is determined based on the fault tracing knowledge graph and multiple related objects.

[0093] Specifically, each relevant component is matched with an object in the fault tracing knowledge graph, and the closest matching object is identified as the relevant object. Optionally, the relevant objects are grouped according to their operational functions, and an operational relationship chain is determined for each group, thereby obtaining multiple operational relationship chains. In other embodiments, for each relevant component, when the running time of the relevant component is greater than or equal to a preset running time, the relevant component is matched with an object in the fault tracing knowledge graph, and the closest matching object is identified as the relevant object. When the running time of the relevant component is less than the preset running time, it is not necessary to determine the corresponding relationship.

[0094] For components with different operating times, components that have been used for a long time can be selected based on the length of their operating time. This can effectively identify components that are operating under overload. Using these components to determine the operating relationship chain can more accurately update the knowledge graph and improve the accuracy of subsequent abnormal fault identification.

[0095] Furthermore, the step of determining the operational relationship chain based on the fault tracing knowledge graph and multiple related objects includes:

[0096] Based on the fault tracing knowledge graph and the multiple related objects, calculate the first shortest path between any two related objects to obtain multiple first shortest paths;

[0097] The running relationship chain is determined based on multiple first shortest paths and the Held-Karp algorithm.

[0098] In other embodiments, the Neo4j graph database can also be selected to construct the knowledge graph, and the Neo4j built-in algorithm can be used to determine the running relationship chain.

[0099] In this embodiment, multiple related objects in the fault tracing knowledge graph are determined by the related components, and the operation relationship chain is determined based on the fault tracing knowledge graph and the multiple related objects, so that the operation relationship chain can be accurately obtained.

[0100] Furthermore, based on any embodiment, a fifth embodiment of the knowledge graph-based printing equipment fault tracing and maintenance decision-making method of the present invention is proposed, wherein the step of determining the target relation chain based on the plurality of relation chains and the confidence level of each relation includes:

[0101] The probability of each relation chain is determined based on a preset calculation formula, the confidence level of each relation, and each relation chain.

[0102] The target relation chain is determined based on the relation probability of each relation chain.

[0103] All the relationship chains are sorted according to the relationship probability. The higher the relationship probability, the higher the ranking of the corresponding relationship chain. The relationship chain ranked first is taken as the target relationship chain.

[0104] Furthermore, after the step of determining the cause of the failure based on the target relationship chain, the method further includes:

[0105] A corresponding maintenance decision is generated based on the cause of the fault.

[0106] In this embodiment, maintenance decisions are made by maintenance personnel or by pre-set maintenance plans based on different causes of failure.

[0107] Furthermore, this invention also proposes a knowledge graph-based printing equipment fault tracing and maintenance decision-making system, which includes:

[0108] The acquisition module is used to acquire the printing equipment's operating data, historical fault data, and current fault information.

[0109] An update module is used to update the confidence level of relationships in the fault tracing knowledge graph based on the operational data and the historical fault data.

[0110] The identification module is used to determine the cause of the fault based on the fault information and the fault tracing knowledge graph.

[0111] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0112] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0113] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0114] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A knowledge graph based printing equipment failure trace and maintenance decision method, characterized in that, The method includes the following steps: Acquire the printing equipment's operating data, historical fault data, and current fault status information; The confidence level of the relationships in the fault tracing knowledge graph is updated based on the operational data and the historical fault data. The cause of the fault is determined based on the fault information and the fault tracing knowledge graph. Generate corresponding maintenance decisions based on the causes of the faults; The step of updating the confidence of relationships in the fault tracing knowledge graph based on the operational data and the historical fault data also includes: Obtain maintenance data of printing equipment, and extract multiple triplet data based on the maintenance data. The structure of the triplet data is: first object, relation, second object; A fault tracing knowledge graph is constructed based on multiple triplet data sets; The historical fault data includes: historical fault relationship chains, which are fault relationship chains determined before the current time based on fault prompts in the fault tracing knowledge graph. The step of updating the confidence of the relationships in the fault tracing knowledge graph based on the operational data and the historical fault data includes: Determine the operational relationship chain based on the operational data; Based on the historical fault relationship chain and the operational relationship chain, a first type of relationship and a second type of relationship are determined in the fault tracing knowledge graph. The first type of relationship is a relationship belonging to the historical fault relationship chain or a relationship belonging to the operational relationship chain. The second type of relationship is a relationship other than the first type of relationship in the fault tracing knowledge graph. The weight value corresponding to each first type of relationship is determined based on the historical fault relationship chain and the operation relationship chain, and the weight value of the second type of relationship is set to a preset value, which is less than the weight value of any first type of relationship. The confidence level of each relation is determined based on all relations and their corresponding weight values; The step of determining the confidence level of each relation based on all relations and their corresponding weight values ​​includes: Based on the fault tracing knowledge graph, at least one target triplet data is determined for each target object, wherein the target triplet data is triplet data with the target object as the first object; Calculate the target confidence level of each target relation on the target object based on the target relations and corresponding target weight data of all the target triplet data; The target confidence level is used as the confidence level; The operational data includes: the operational functions and corresponding related components, and the operational time. The step of determining the operational relationship chain based on the operational data includes: Based on the relevant components, multiple related objects in the fault tracing knowledge graph are identified; The operational relationship chain is determined based on the fault tracing knowledge graph and multiple related objects; For each relevant component, when the running time of the relevant component is greater than or equal to the preset running time, the relevant component is matched with the object in the fault tracing knowledge graph, and the closest matching object is the relevant object. The step of determining the cause of the fault based on the fault information and the fault tracing knowledge graph includes: Based on the fault information, an initial node is determined in the fault tracing knowledge graph, and a relationship chain is searched based on the initial node to obtain multiple relationship chains; The target relation chain is determined based on the multiple relation chains and the confidence level of each relation, and the cause of the failure is determined based on the target relation chain.

2. The knowledge graph-based fault tracing and maintenance decision-making method for printing equipment as described in claim 1, characterized in that, The step of determining the target relation chain based on the plurality of relation chains and the confidence level of each relation includes: The probability of each relation chain is determined based on a preset calculation formula, the confidence level of each relation, and each relation chain. The target relation chain is determined based on the relation chain probability of each relation chain.

Citation Information

Patent Citations

  • Equipment fault cause tracing method based on reinforcement learning and knowledge graph

    CN112100392A

  • Machine room fault diagnosis method and device based on knowledge graph, equipment and medium

    CN112231493A