Fault analysis method and apparatus, computer device, and storage medium

Abstract

The present application relates to a fault analysis method and apparatus, a computer device, and a storage medium. The method comprises: constructing a product structure tree of a production line; for each node in the product structure tree, determining at least one fault factor capable of causing a failure of the node and an initial weight of each fault factor; correcting the initial weight of each fault factor to obtain a target weight of each fault factor; and on the basis of the at least one fault factor corresponding to each node and the target weight corresponding to each fault factor, constructing an attribution analysis archive with respect to a failure mode of the production line, and performing fault analysis on the production line by means of the attribution analysis archive. By means of the method, when any node in the production line fails, fault analysis can be performed on the failed node by means of the attribution analysis archive, thereby allowing for accurate maintenance, and overcoming the defect of time and labor consuming of static maintenance.

Description

Fault analysis method and device, computer device, and storage medium TECHNICAL FIELD

[0001] The present application relates to the technical field of fault processing, and in particular to a fault analysis method and device, a computer device, a storage medium, and a computer program product. BACKGROUND

[0002] In the field of intelligent manufacturing, maintenance of equipment is a necessary process. The traditional method of maintaining equipment is static maintenance, that is, maintenance and care work is performed when the equipment is not running or in use. For example, regular inspection, cleaning, lubrication, repair, and replacement of equipment components, and the like, to ensure that the equipment can operate normally when needed and extend its service life. However, this method requires maintenance of all equipment, which consumes a large amount of manpower and resources.

[0003] SUMMARY

[0004] Therefore, it is necessary to provide a fault analysis method and device, a computer device, a computer readable storage medium, and a computer program product to solve the technical problem that the above method consumes a large amount of manpower and resources.

[0005] In a first aspect, the present application provides a fault analysis method. The method comprises:

[0006] constructing a product structure tree of a production line; the product structure tree has multiple levels, and each level includes at least one node of the production line;

[0007] for each node in the product structure tree, determining at least one fault factor that can cause the node to fail, and an initial weight of each fault factor;

[0008] correcting the initial weight of each fault factor to obtain a target weight of each fault factor;

[0009] constructing an attribution analysis file on failure modes of the production line according to the at least one fault factor corresponding to each node and the target weight of each fault factor corresponding to each fault factor, and performing fault analysis on the production line through the attribution analysis file.

[0010] In one embodiment, the correcting the initial weight of each fault factor to obtain a target weight of each fault factor comprises:

[0011] determining a correction coefficient corresponding to each fault factor, respectively;

[0012] multiplying the initial weight of each fault factor by the corresponding correction coefficient to obtain the target weight of each fault factor.

[0013] In one of the embodiments, the initial weight of each failure factor is modified to obtain the target weight of each failure factor, and the method further comprises:

[0014] The initial weight of each failure factor is normalized to obtain the normalized weight of each failure factor.

[0015] The normalized weight of each failure factor is modified to obtain the modified weight of each failure factor.

[0016] The modified weight of each failure factor is normalized to obtain the target weight of each failure factor.

[0017] In one of the embodiments, the product structure tree of the production line is constructed, and the method comprises:

[0018] Collecting the sub-production line architecture information of each edge node; each edge node corresponds to the production of a sub-production line of the production line;

[0019] According to the sub-production line architecture information of each edge node, an initial product structure tree of the production line is constructed.

[0020] According to the relationship between the business information of the production line and the production line architecture information, a corresponding business node is mounted on the initial product structure tree to obtain the product structure tree of the production line.

[0021] In one of the embodiments, the initial product structure tree of the production line is constructed according to the sub-production line architecture information of each edge node, and the method comprises:

[0022] The attributes and capabilities of the product structure tree of each level are defined by an automatic modeling language, and the component relationship of the product structure tree of each level is described by a graph database, and the initial product structure tree of the production line is constructed according to the sub-production line architecture information of each edge node.

[0023] In one of the embodiments, the production line is analyzed for failure by using the attribution analysis archive, and the method comprises:

[0024] In the case that a target node of the production line fails, the attribution analysis archive is searched to determine each failure factor causing the failure of the target node and the target weight of each failure factor;

[0025] According to the target weight of each failure factor, each failure factor is investigated to determine the target failure factor causing the failure of the target node.

[0026] In one of the embodiments, the troubleshooting of the respective failure factors according to the target weight of the respective failure factors comprises:

[0027] The priority of the troubleshooting of the respective failure factors is determined according to the target weight of the respective failure factors; the priority is in positive correlation with the target weight;

[0028] The respective failure factors are troubleshooted in order of the priority from high to low.

[0029] In a second aspect, the present application further provides a failure analysis device. The device comprises:

[0030] A construction module is configured to construct a product structure tree of a production line; the product structure tree has multiple levels, and each level comprises at least one node of the production line;

[0031] A determination module is configured to determine, for each node in the product structure tree, at least one failure factor that can cause failure of the node and an initial weight of each failure factor;

[0032] A correction module is configured to correct the initial weight of each failure factor to obtain a target weight of each failure factor;

[0033] An analysis module is configured to construct, according to the at least one failure factor corresponding to each node and the target weight of each failure factor, an attribution analysis file about a failure mode of the production line, and perform failure analysis on the production line through the attribution analysis file.

[0034] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0035] A product structure tree of a production line is constructed; the product structure tree has multiple levels, and each level comprises at least one node of the production line;

[0036] For each node in the product structure tree, at least one failure factor that can cause failure of the node and an initial weight of each failure factor are determined;

[0037] The initial weight of each failure factor is corrected to obtain a target weight of each failure factor;

[0038] According to the at least one failure factor corresponding to each node and the target weight of each failure factor, an attribution analysis file about a failure mode of the production line is constructed, and failure analysis on the production line is performed through the attribution analysis file.

[0039] In a fourth aspect, the present application provides a computer readable storage medium. The computer readable storage medium has a computer program stored thereon, and the computer program, when executed by a processor, implements the following steps:

[0040] constructing a product structure tree of the production line; the product structure tree has multiple levels, and each level includes at least one node of the production line;

[0041] determining, for each node in the product structure tree, at least one failure factor that can cause the node to fail, and an initial weight of each failure factor;

[0042] correcting the initial weight of each failure factor to obtain a target weight of each failure factor;

[0043] constructing, according to the at least one failure factor corresponding to each node and the target weight of each failure factor, an attribution analysis file about the failure mode of the production line, and performing failure analysis on the production line through the attribution analysis file.

[0044] In a fifth aspect, the present application provides a computer program product. The computer program product includes a computer program, and the computer program, when executed by a processor, implements the following steps:

[0045] constructing a product structure tree of the production line; the product structure tree has multiple levels, and each level includes at least one node of the production line;

[0046] determining, for each node in the product structure tree, at least one failure factor that can cause the node to fail, and an initial weight of each failure factor;

[0047] correcting the initial weight of each failure factor to obtain a target weight of each failure factor;

[0048] constructing, according to the at least one failure factor corresponding to each node and the target weight of each failure factor, an attribution analysis file about the failure mode of the production line, and performing failure analysis on the production line through the attribution analysis file.

[0049] The fault analysis method, device, computer device, storage medium and computer program product determine at least one fault factor that can cause a node fault and an initial weight of each fault factor for each node in the product structure tree of the production line, correct the initial weight of each fault factor to obtain a target weight of each fault factor, and construct an attribution analysis file of the failure mode of the production line according to at least one fault factor corresponding to each node and the target weight of each fault factor. The method constructs the product structure tree of the production line, constructs the attribution analysis file of the production line based on the fault factors corresponding to each node in the product structure tree and the target weight of each fault factor, and performs fault analysis on the fault node through the attribution analysis file when a fault occurs in any node in the subsequent production line, so that accurate maintenance can be performed, and the defect of time-consuming and labor-consuming static maintenance is overcome. Meanwhile, the initial weight of each fault factor of each node is corrected through the forgetting curve, the accuracy of the weight of each fault factor determined to cause the node fault is improved, and the accuracy of subsequent maintenance is improved. BRIEF DESCRIPTION OF DRAWINGS

[0050] FIG. 1 is a flowchart of a fault analysis method in an embodiment;

[0051] FIG. 2 is a schematic diagram of the hierarchical description of the product structure tree and the composition of the AML of the final factory level through the AML language in an embodiment;

[0052] FIG. 3 is a schematic diagram of the relationship among the production line layer, the equipment layer and the part layer in an embodiment;

[0053] FIG. 4 is a flowchart of a fault analysis method in another embodiment;

[0054] FIG. 5 is a schematic diagram of an assembly production line of an automobile in an embodiment;

[0055] FIG. 6 is a schematic diagram of a product structure tree (BOM tree) in an embodiment;

[0056] FIG. 7 is a schematic diagram of mounting a fault factor in a product structure tree in an embodiment;

[0057] FIG. 8 is a structural block diagram of a fault analysis device in an embodiment;

[0058] FIG. 9 is an internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0060] In one embodiment, as shown in FIG. 1, a fault analysis method is provided, and this embodiment is exemplified by the method applied to a cloud platform. It can be understood that the method can also be applied to a server or other devices with processing functions. In this embodiment, the method comprises the following steps:

[0061] Step S110, constructing a product structure tree of the production line; the product structure tree has multiple levels, and each level comprises at least one node of the production line.

[0062] The product structure tree refers to a BOM tree (Bill of Material), which is a tree diagram describing the hierarchical structure of product materials.

[0063] The production line can be an industrial production line.

[0064] In a specific implementation, the entity architecture information of the production line to be analyzed for faults, i.e., the entities of the devices and the association relationship between the devices, etc., can be obtained. Based on the entity architecture information, a multi-level product structure tree is constructed. On this basis, the non-entity nodes of the production line can be further mounted on the product structure tree, thereby obtaining the final product structure tree.

[0065] Step S120, determining at least one fault factor that can cause a fault of each node in the product structure tree and an initial weight of each fault factor.

[0066] The initial weight of the fault factor represents the possibility or probability of the fault factor causing the fault of the corresponding node.

[0067] Specifically, after the construction of the product structure tree is completed, fault analysis can be performed for each node to construct an attribution analysis archive of the production line. Specifically, for each node in the product structure tree, the historical fault data corresponding to the node is determined, at least one fault factor that can cause a fault of the node is determined according to the historical fault data, and the initial weight of each fault factor is calculated.

[0068] For example, for the cylinder node, the fault factors that can cause a fault of the cylinder include compressor failure, valve leakage, and pipeline leakage. The initial weights of the fault factors can be: compressor failure-20%, valve leakage-32%, and pipeline leakage-48%.

[0069] Step S130, correcting the initial weights of the fault factors to obtain target weights of the fault factors.

[0070] In a specific implementation, as time goes by, the probability of a node failure caused by the same failure factor will change. For example, for a cylinder, the weight of a cylinder failure caused by a compressor failure on the first day after the cylinder is installed will be different from the weight of a cylinder failure caused by a compressor failure on the 20th day. Therefore, the initial weight of each failure factor needs to be corrected by considering the impact of the change of the weight of each failure factor over time.

[0071] More specifically, the initial weight of each failure factor can be corrected by a forgetting curve to obtain the target weight of each failure factor.

[0072] The forgetting curve, i.e., Ebbinghaus forgetting curve, can be expressed by the following formula:

[0073] In the formula, R represents the retention of memory after time t, e is the base of natural logarithm, t is the elapsed time, and s represents the sensitivity to specific information. The value of s will be different for different information.

[0074] Based on the above formula of the forgetting curve, it can be determined that the rate of information forgetting is the fastest at the beginning after learning, and then the rate gradually slows down.

[0075] In step S140, an attribution analysis file of the failure mode of the production line is constructed according to at least one failure factor corresponding to each node and the target weight of each failure factor, and the failure analysis of the production line is performed through the attribution analysis file.

[0076] The attribution analysis file is constructed based on the attribution analysis method, which is also known as the fishbone diagram analysis method or the cause-and-effect analysis method. It is a method for finding the root cause of a problem. Based on the visual fishbone diagram, the problem analysis result and the solution idea can be displayed concisely and intuitively. Generally, the problem to be analyzed is first determined, and after in-depth analysis of various causes of the problem, the fishbone diagram structure is arranged in groups according to levels.

[0077] In a specific implementation, after the failure factors of each node in the production line to be analyzed and the target weight of the failure factors are determined, the attribution analysis file of the failure mode of the production line can be constructed according to the relationship of each node on the product structure tree, and the failure analysis of the production line can be performed through the attribution analysis file subsequently.

[0078] In the fault analysis method, a product structure tree of the production line is constructed, for each node in the product structure tree, at least one fault factor that can cause the node to fail is determined, and an initial weight of each fault factor is determined and corrected to obtain a target weight of each fault factor; and an attribution analysis file about failure modes of the production line is constructed according to at least one fault factor corresponding to each node and the target weight of each fault factor. The method constructs an attribution analysis file about the production line based on the fault factors corresponding to each node in the product structure tree and the target weight of each fault factor, and when a fault occurs at any node in the subsequent production line, the fault analysis of the node can be performed through the attribution analysis file, so that accurate maintenance can be performed, and the defects of time-consuming and labor-consuming static maintenance are overcome. Meanwhile, the initial weight of each fault factor of each node is corrected through the forgetting curve, which can improve the accuracy of the weight of each fault factor that causes the node to fail, thereby improving the accuracy of subsequent maintenance.

[0079] In an exemplary embodiment, the step S130 of correcting the initial weight of each fault factor to obtain the target weight of each fault factor includes: determining a correction coefficient corresponding to each fault factor respectively; and multiplying the initial weight of each fault factor by the corresponding correction coefficient to obtain the target weight of each fault factor.

[0080] In a specific implementation, the correction coefficient corresponding to each fault factor can be determined through the forgetting curve, and the initial weight of each fault factor is multiplied by the corresponding correction coefficient to obtain the target weight of each fault factor. The formula of the forgetting curve is: wherein R represents the memory retention after time t, e is the base of natural logarithm, t is the elapsed time, and s represents the sensitivity to specific information. The correction coefficient is denoted as the R value in the forgetting curve, i.e. the memory retention after time t. The forgetting curve is used to correct each fault factor, specifically, the dependent variable R, i.e. the memory retention, in the forgetting curve is used as the correction coefficient to correct the initial weight of each fault factor.

[0081] More specifically, the s value of each fault factor, i.e. the sensitivity of each fault factor, can be determined first, and substituted into the formula of the forgetting curve to obtain the memory retention formula of each fault factor changing with time t. Further, the initial weight of each fault factor is multiplied by the memory retention formula of each fault factor changing with time t to obtain a target weight curve of each fault factor, and the target weight curve changes with time t.

[0082] For example, still taking the cylinder node as an example, it is assumed that the fault factors causing the cylinder fault include: compressor failure, valve leakage and pipeline leakage. The initial weights of the fault factors are: compressor failure-20%, valve leakage-32%, pipeline leakage-48%, and the s values are s1, s2 and s3. The target weights of the fault factors corresponding to the cylinder node can be respectively represented as:

[0083] In this embodiment, the forgetting curve is introduced to correct the initial weights of the fault factors, considering the influence of time on the weights of the fault factors, so that the accuracy and reliability of the determined target weights of the fault factors can be improved.

[0084] In an exemplary embodiment, the step S130 of correcting the initial weights of the fault factors to obtain the target weights of the fault factors further includes: normalizing the initial weights of the fault factors to obtain normalized weights of the fault factors; correcting the normalized weights of the fault factors to obtain corrected weights of the fault factors; and normalizing the corrected weights of the fault factors to obtain the target weights of the fault factors.

[0085] In a specific implementation, for any node, after the initial weights of the fault factors are determined, the fault factors can be normalized first, and the maximum-minimum value normalization method can be used for normalization. For example, it is assumed that the initial weights of the fault factors are w1, w2, w3, …, and wn, respectively. If w1+w2+w3+…+wn=P, the normalized weights of the fault factors are w1 / P, w2 / P, w3 / P, …, and wn / P, respectively. i i i i i Further normalizing the corrected weights of the fault factors to obtain the target weights of the fault factors, that is, processing the corrected weights so that the sum of all the corrected weights:∑(R i ×w i / P) is 1. It is assumed that (R1×w1 / P)+(R2×w2 / P)+(R3×w3 / P)+…+(R i ×w i / P) = Q, and the target weights of the fault factors are (R1×w1 / P) / Q, (R2×w2 / P) / Q, (R3×w3 / P) / Q, …, and (R i ​​​​​Xw i / P) / Q.

[0086] In this embodiment, by normalizing the initial weight or the corrected weight of each fault factor after each determination, the differences between the weight values of different features can be eliminated, the calculation cost can be reduced, and the calculation efficiency can be improved.

[0087] In an example embodiment, the step S110 of constructing the product structure tree of the production line includes: collecting the sub-production line architecture information of each edge node; constructing an initial product structure tree of the production line according to the sub-production line architecture information of each edge node; and mounting the corresponding business node on the initial product structure tree according to the relationship between the business information and the production line architecture information of the production line, to obtain the product structure tree of the production line.

[0088] It should be noted that constructing the product structure tree of the production line mainly includes two steps: the first step is to construct a BOM tree, that is, to convert each entity device in the production line into a BOM tree. The relationship between the leaf nodes and the upper nodes of the BOM tree is an immutable relationship, so the main BOM tree needs to be constructed first. The second step is to mount the nodes of the BOM tree. The BOM tree can realize the mounting of non-entity nodes. This form enables the subsequent use of knowledge graph derivative computing schemes.

[0089] In a specific implementation, in a large production line, the production line can involve multiple sub-production lines, and each sub-production line can be deployed in different areas and controlled through different edge nodes. Therefore, before constructing the product structure tree of the production line, the architecture information of the sub-production line can be collected from each edge node, and then the architecture information of each sub-production line can be summarized to construct an initial product structure tree of the production line. Further, considering that some business nodes do not have entity devices, the mounting of non-entity business nodes is needed on the basis of the constructed initial product structure tree, thereby obtaining the product structure tree of all nodes affecting the fault analysis of the production line, which facilitates subsequent fault analysis of the production line and improves the accuracy of the analysis results.

[0090] In this embodiment, the initial product structure tree is first constructed by the entity nodes of the production line, and then the mounting of non-entity business nodes is performed on this basis, thereby obtaining the product structure tree of all nodes affecting the fault analysis of the production line, which facilitates subsequent fault analysis of the production line and improves the accuracy of the analysis results.

[0091] Further, in an example embodiment, the initial product structure tree of the production line is constructed according to the sub-production line architecture information of each edge node, including: defining the attributes and capabilities of the product structure tree at each level in an automated modeling language, and describing the component relationship of the product structure tree at each level through a graph database, and constructing the initial product structure tree of the production line according to the sub-production line architecture information of each edge node.

[0092] Among them, the Automation Markup Language (AML) is an industrial modeling language compatible with xml format, which provides modeling method based on BOM level. It can be applied to all industrial fields requiring data exchange, such as discrete industry or process industry.

[0093] Among them, Neo in the graph database (Neo4j) is a network-oriented database, which is an embedded, disk-based, full-transaction Java persistence engine, but it stores structured data on the network rather than in tables. The network (called graph from the mathematical point of view) is a flexible data structure that can be applied to more agile and fast development mode. It not only supports graph-based data expression, but also supports graph-derived functions such as searching nodes according to relevance.

[0094] Specifically, AML is mainly used to realize the description of the main body, interface, role and parameter of BOM level (parts or devices, etc.) and the description of all constituent elements, and can realize the nesting of elements, so that the composition scheme of upper and lower BOMs can be realized. After defining the AML description of each level of BOM, the description of AML language can be further simplified, and the AML description of each level of BOM is simplified to attributes and sub-element groups, so as to realize the whole factory with a single AML description. For example, referring to FIG. 2, a schematic diagram of describing each level of BOM and the composition of AML of the final factory level by AML is provided.

[0095] AML is mainly used to define the attributes and capabilities of each level of BOM, and the AML language can only describe the hierarchical relationship and attribute relationship of each level of BOM of the product structure tree, but it is difficult to provide intuitive description for the horizontal relationship. Therefore, the graph database Neo4j is also introduced in the present application, and the BOM structure composed based on the graph database Neo4j can not only reflect the combination relationship of the upper layer to the lower layer, but also reflect the assembly relationship between components. As shown in FIG. 3, there are multiple parts in the same device, so there is actually an assembly relationship between the parts, for example, part 1 and part 2 and part 3 have an assembly relationship. As for the device, device 1 and device 2 and device 3, each device has a cooperation relationship between devices, and the display of these horizontal relationships can be realized by using the graph database Neo4j. Therefore, the initial product structure tree of the production line is constructed by combining the automation modeling language with the graph database.

[0096] In the embodiment, the automated modeling language is combined with the graph database, the relationship between the levels is reflected by the automated modeling language, and the horizontal relationship between the components in the same level is reflected by the graph database, so that the accuracy of the product structure tree constructed can be improved. Meanwhile, the modeling method of the scheme has good generalization and high applicability by the automated modeling language which can be applied to all industrial fields requiring data exchange.

[0097] In an example embodiment, in the step S140, the production line is analyzed for failure by the attribution analysis file, including: in the case that a target node of the production line fails, searching the attribution analysis file to determine each failure factor causing the target node to fail and a target weight of each failure factor; and according to the target weight of each failure factor, investigating each failure factor to determine a target failure factor causing the target node to fail.

[0098] In a specific implementation, after the construction of the attribution analysis file is completed, if a target node of a subsequent production line fails, the pre-constructed attribution analysis file can be directly searched to determine each failure factor causing the target node to fail and a target weight of each failure factor. Each failure factor and the target weight of each failure factor can be displayed, so that each failure factor can be further investigated according to the target weight of each failure factor to determine a target failure factor causing the target node to fail.

[0099] Further, in an example embodiment, the investigating each failure factor according to the target weight of each failure factor includes: determining a priority of investigating each failure factor according to the target weight of each failure factor; the priority and the target weight have a positive correlation; and investigating each failure factor in a descending order of the priority.

[0100] In a specific implementation, when investigating the failure according to the target weight of each failure factor, the investigation can be performed according to the weight, and specifically, the investigation can be performed according to the maximum weight first rule, that is, the greater the target weight, the greater the possibility of causing the cylinder to fail, and therefore, the higher the priority of the investigation. Thus, according to the positive correlation between the priority and the target weight, the priority of investigating each failure factor can be determined, and each failure factor can be investigated in a descending order of the priority.

[0101] For example, when a cylinder failure occurs, an attribution result of the cylinder failure is obtained: [compressor failure: 9.3%, valve leakage: 52.8%, pipeline leakage: 37.9%]. According to the target weight, the priority of each failure factor is ranked from high to low: valve leakage > pipeline leakage > compressor failure, and therefore, when troubleshooting is performed, the valve leakage, the pipeline leakage, and the compressor failure can be sequentially checked.

[0102] In the above embodiment, the troubleshooting of the failure in the production line is performed through the attribution analysis archive, which can improve the maintenance efficiency. In addition, the priority of each failure factor is distinguished according to the target weight, which is beneficial to improve the overall work efficiency, concentrates resources and manpower on the main failure point, and avoids high labor costs caused by searching the archive and relying too much on experience.

[0103] As shown in FIG. 4, a flowchart of a failure analysis method provided by another exemplary embodiment is shown in FIG. 4, which includes the following steps:

[0104] In step S401, the sub-production line architecture information of each edge node is collected.

[0105] In step S402, the attributes and capabilities of the product structure tree of each level are defined by an automatic modeling language, the component relationship of the product structure tree of each level is described by a graph database, and the initial product structure tree of the production line is constructed according to the sub-production line architecture information of each edge node.

[0106] In step S403, the corresponding business node is mounted on the initial product structure tree according to the relationship between the business information and the production line architecture information of the production line, and the product structure tree of the production line is obtained.

[0107] In step S404, for each node in the product structure tree, at least one failure factor that can cause the node failure is determined, and an initial weight of each failure factor is determined.

[0108] In step S405, the initial weight of each failure factor is normalized to obtain a normalized weight of each failure factor.

[0109] In step S406, a forgetting curve is used to determine a correction coefficient corresponding to each failure factor.

[0110] In step S407, the normalized weight of each failure factor is multiplied by the corresponding correction coefficient to obtain a corrected weight of each failure factor.

[0111] In step S408, the corrected weight of each failure factor is normalized to obtain a target weight of each failure factor.

[0112] Step S409, according to at least one failure factor corresponding to each node and the target weight corresponding to each failure factor, a cause analysis file about the failure mode of the production line is constructed;

[0113] Step S410, in the case that the target node of the production line fails, the cause analysis file is searched to determine each failure factor causing the failure of the target node and the target weight of each failure factor;

[0114] Step S411, according to the target weight of each failure factor, the priority of checking each failure factor is determined; the priority is in a positive correlation with the target weight;

[0115] Step S412, each failure factor is checked in the order from high to low priority.

[0116] In the embodiment, the graph and the BOM are combined, so that the maintenance of the BOM is deeply mined by combining the maintenance business point and the algorithm model, and accurate operation and maintenance message pushing is realized. The modeling of the multi-level BOM and the preliminary construction of the technical file of the BOM are realized through the graph database Neo4j, the cost of operation and maintenance can be gradually evaluated and weighed, and the data is self-polished and optimized through the Ebbinghaus forgetting curve, and finally the construction of the BOM failure mode cause analysis file under the massive production line data system is realized. On the basis of this file, the four-quadrant method can be combined with the maintenance cost for expansion. It can also provide relevant basis for accurate operation and maintenance business. Based on this system, a graph algorithm scheme (early warning model, process model, diagnosis model, mechanism model, diagnosis model) based on failure mode data can be gradually constructed, which provides an implementation method for the expansion of subsequent dynamic maintenance related business and the construction of a digital lean production system.

[0117] It should be noted that the existing degradation model and failure mode of BOM are difficult to directly determine, and under the condition of the diversity of the existing BOM, how to accurately maintain the BOM remains in great uncertainty. Moreover, how to realize the linkage analysis of the BOM and integrate the entire production line system is also a difficult problem. In addition, the existing cause analysis system relies too much on data files, especially the FMECA (Failure Mode, Effects and Criticality Analysis) file construction mode used in the aviation industry for a long time, which requires a large amount of data injection and search.

[0118] Therefore, based on AML language modeling, the application combines maintenance business knowledge and graph technology and algorithm library technology to jointly construct a graph-based precision maintenance system. Through the combination of BOM failure mode data and the Ebbinghaus forgetting curve, a BOM failure mode data system is formed. Then, through the cause analysis method, the causes of BOM failure modes are found, and weights are assigned to related nodes and the data is corrected by the Ebbinghaus forgetting curve, thereby realizing the construction of a BOM failure mode cause analysis technology system. This provides an implementation method for the expansion of subsequent dynamic maintenance related businesses and the construction of a digital lean production system.

[0119] In one embodiment, in order to facilitate the understanding of the embodiments of the application by those skilled in the art, the following will be described with reference to the specific examples of the assembly line of the automobile in combination with the accompanying drawings.

[0120] Referring to FIG. 5, a schematic diagram of an assembly line of an automobile is shown. By disassembling the BOM of the production line, a schematic diagram of a product structure tree (BOM tree) as shown in FIG. 6 can be obtained. As shown in FIG. 6, the product structure tree includes three levels, each level includes at least one node, and the nodes between adjacent levels have an association relationship.

[0121] When constructing the cause analysis archive, as shown in FIG. 7, taking the cylinder node as an example, three fault factors that cause the cylinder failure can be mounted in a one-way node.

[0122] After completing the node mounting, the target weights of the three fault factors that cause the cylinder failure can be determined in combination with the historical data and the forgetting curve. The correction process from the initial weight to the target weight is shown in Table 1.

[0123] Table 1 Target weights of the fault factors corresponding to the cylinder

[0124] When the cylinder failure occurs subsequently, the cause analysis archive can be searched to obtain the cause result of the cylinder failure: [compressor failure: 9.3%, valve leakage: 52.8%, pipeline leakage: 37.9%].

[0125] Further, according to the target weights, the fault factors are investigated in order. First, according to the target weights, the priority of each fault factor is sorted as follows: valve leakage > pipeline leakage > compressor failure. When investigating, whether the valve leaks, whether the pipeline leaks, and whether the compressor fails can be investigated in turn.

[0126] The embodiment analyzes the maintenance system of the aviation industry, and the core points are combined with the intelligent manufacturing industry for optimization and upgrading, thereby constructing the equipment failure mode analysis scheme, and providing support for the construction of the intelligent manufacturing system and the dynamic maintenance system.

[0127] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or steps or stages in other steps.

[0128] Based on the same inventive concept, the embodiment of the present application also provides a fault analysis device for implementing the above-mentioned fault analysis method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more fault analysis device embodiments provided below can refer to the limitations of the fault analysis method in the above text, and will not be repeated here.

[0129] In one embodiment, as shown in FIG. 8, a fault analysis device is provided, including a construction module 810, a determination module 820, a correction module 830 and an analysis module 840, wherein:

[0130] The construction module 810 is configured to construct a product structure tree of the production line; the product structure tree has multiple levels, and each level includes at least one node of the production line;

[0131] The determination module 820 is configured to determine, for each node in the product structure tree, at least one fault factor that can cause a fault of the node, and an initial weight of each fault factor;

[0132] The correction module 830 is configured to correct the initial weight of each fault factor to obtain a target weight of each fault factor;

[0133] The analysis module 840 is configured to construct, according to the at least one fault factor corresponding to each node and the target weight corresponding to each fault factor, an attribution analysis file about the failure mode of the production line, and perform fault analysis on the production line through the attribution analysis file.

[0134] In one of the embodiments, the correction module 830 is further configured to determine a correction coefficient corresponding to each fault factor respectively; and multiply the initial weight of each fault factor with the corresponding correction coefficient to obtain a target weight of each fault factor.

[0135] In one of the embodiments, the correction module 830 is further configured to normalize the initial weight of each fault factor to obtain a normalized weight of each fault factor; correct the normalized weight of each fault factor to obtain a corrected weight of each fault factor; and normalize the corrected weight of each fault factor to obtain a target weight of each fault factor.

[0136] In one of the embodiments, the construction module 810 is further configured to collect sub-production line architecture information of each edge node; each edge node corresponds to production of a sub-production line of the production line; construct an initial product structure tree of the production line according to the sub-production line architecture information of each edge node; and mount a corresponding business node on the initial product structure tree according to the relationship between the business information and the architecture information of the production line to obtain a product structure tree of the production line.

[0137] In one of the embodiments, the construction module 810 is further configured to define attributes and capabilities of the product structure tree of each level in an automatic modeling language, and describe component relationships of the product structure tree of each level through a graph database, and construct an initial product structure tree of the production line according to the sub-production line architecture information of each edge node.

[0138] In one of the embodiments, the analysis module 840 is further configured to, in the case that a target node of the production line fails, search the attribution analysis archive to determine each fault factor causing the failure of the target node and a target weight of each fault factor; and investigate each fault factor according to the target weight of each fault factor to determine a target fault factor causing the failure of the target node.

[0139] In one of the embodiments, the analysis module 840 is further configured to determine a priority of investigating each fault factor according to the target weight of each fault factor; the priority is in a positive correlation with the target weight; and investigate each fault factor in a descending order of the priority.

[0140] Each module in the above fault analysis device can be realized by software, hardware, and a combination thereof in whole or in part. Each module can be embedded in or independent of a processor in a computer device in a hardware form, or stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to each module.

[0141] In an embodiment, a computer device is provided, which can be a server, and an internal structure diagram of the computer device can be as shown in FIG. 9. The computer device includes a processor, a memory and a network interface connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store data in a fault analysis process. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a fault analysis method.

[0142] Those skilled in the art can understand that the structure shown in FIG. 9 is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0143] In an embodiment, a computer device is also provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0144] In an embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0145] In an embodiment, a computer program product is provided, which includes a computer program. The computer program is executed by a processor to implement the steps in the above method embodiments.

[0146] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0147] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0148] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0149] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method of failure analysis, characterized by, The method comprises: constructing a product structure tree of a production line; the product structure tree has multiple levels, and each level comprises at least one node of the production line; for each node in the product structure tree, determining at least one failure factor that can cause the node to fail, and an initial weight of each failure factor; correcting the initial weight of each failure factor to obtain a target weight of each failure factor; constructing an attribution analysis file of a failure mode of the production line according to the at least one failure factor corresponding to each node and the target weight of each failure factor, and performing failure analysis on the production line through the attribution analysis file.

2. The method of claim 1, wherein, The correction of the initial weight of each failure factor to obtain a target weight of each failure factor comprises: determining a correction coefficient corresponding to each failure factor respectively; multiplying the initial weight of each failure factor by the corresponding correction coefficient to obtain the target weight of each failure factor.

3. The method of claim 1, wherein, The correction of the initial weight of each failure factor to obtain a target weight of each failure factor further comprises: normalizing the initial weight of each failure factor to obtain a normalized weight of each failure factor; correcting the normalized weight of each failure factor to obtain a corrected weight of each failure factor; normalizing the corrected weight of each failure factor to obtain the target weight of each failure factor.

4. The method of claim 1, wherein, The construction of the product structure tree of the production line comprises: collecting sub-production line architecture information of each edge node; each edge node corresponds to production control of a sub-production line of the production line; constructing an initial product structure tree of the production line according to the sub-production line architecture information of each edge node; mounting a corresponding business node on the initial product structure tree according to the relationship between the business information of the production line and the production line architecture information to obtain the product structure tree of the production line.

5. The method of claim 4, wherein, The construction of the initial product structure tree of the production line according to the sub-production line architecture information of each edge node comprises: defining the attributes and capabilities of the product structure tree of each level in an automated modeling language, and describing the component relationship of the product structure tree of each level through a graph database, and constructing the initial product structure tree of the production line according to the sub-production line architecture information of each edge node.

6. The method of claim 1, wherein, The failure analysis on the production line through the attribution analysis file comprises: in the case that a target node of the production line fails, searching the attribution analysis file to determine each failure factor that causes the target node to fail and the target weight of each failure factor; according to the target weight of each failure factor, checking each failure factor to determine a target failure factor that causes the target node to fail.

7. The method of claim 6, wherein, The checking of each failure factor according to the target weight of each failure factor comprises: determining a priority of checking each failure factor according to the target weight of each failure factor; the priority and the target weight are in a positive correlation relationship. The various failure factors are investigated in order from high to low priority.

8. A failure analysis apparatus characterized by comprising: The device comprises: a construction module configured to construct a product structure tree of the production line, the product structure tree having multiple levels, each level comprising at least one node of the production line; a determination module configured to determine, for each node in the product structure tree, at least one failure factor that can cause the node to fail, and an initial weight of each failure factor; a correction module configured to correct the initial weights of the various failure factors to obtain target weights of the various failure factors; an analysis module configured to construct, according to the at least one failure factor corresponding to each node and the target weight corresponding to each failure factor, an attribution analysis file about a failure mode of the production line, and to perform failure analysis on the production line through the attribution analysis file. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the failure analysis method of any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the failure analysis method of any one of claims 1 to 7.

11. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the failure analysis method of any one of claims 1 to 7.

Images