CN115857469A - Industrial equipment fault knowledge base construction method and device and fault diagnosis method and system - Google Patents

Abstract

The invention discloses a method and a device for constructing a fault knowledge base of industrial equipment, and a fault diagnosis method and a fault diagnosis system, wherein the method comprises the following steps: acquiring the incidence relation between each component in the tested equipment and the failure mode; abstracting a graph model of each failure mode to construct failure mode nodes, constructing a fault tree by taking abnormal fault events as nodes and constructing according to the sequence, constructing a basic graph model by taking all parts in the tested equipment as nodes, and connecting each node in the basic graph model and the fault tree with the corresponding failure mode nodes to construct a fault knowledge system; and storing the fault knowledge system by using a graph database according to a preset logic model to obtain a fault knowledge base of the tested equipment. The method has the advantages of simple implementation method, low cost, high data integrity, capability of fully reflecting the correlation among the equipment configuration, the internal structure relation and the failure mode, improvement on the efficiency and the precision of fault diagnosis and the like.

Description

Industrial equipment fault knowledge base construction method and device and fault diagnosis method and system

Technical Field

The invention relates to the technical field of fault diagnosis of industrial equipment, in particular to a method and a device for constructing a fault knowledge base of the industrial equipment, and a method and a system for fault diagnosis.

Background

For fault diagnosis of large-scale industrial equipment with a complex structure, a manual detection and troubleshooting mode is generally adopted or fault judgment is carried out based on simple operation data at present. For example, for the maintenance of a railway locomotive, state and fault diagnosis is mainly performed according to state and alarm information detected during the online operation of the locomotive, usually, the locomotive is temporarily butted after the locomotive arrives, the fault data of the current time of the day on the locomotive is downloaded, the field analysis is performed on the fault data of the locomotive, and the fault location fault point is checked and the solution is determined by combining manual experience and manual detection. However, the above fault diagnosis method depends on knowledge level and experience of analysts, has low diagnosis efficiency and intelligence, is easy to cause missed detection and false detection, has the problems of untimely fault processing, incapability of accurately finding out fault positions, incomplete fault processing and the like, and is difficult to realize fault positioning in time and accurately.

The intelligent fault diagnosis method can solve the problems that the traditional manual troubleshooting fault positioning method is low in efficiency and low in intelligent degree, in the intelligent fault diagnosis method in the prior art, a fault diagnosis model is generally constructed on the basis of data in different fault states by acquiring data of equipment in different fault states, and intelligent fault diagnosis can be achieved by inputting the acquired fault data into the fault diagnosis model after the fault data are acquired in real time. However, the above intelligent fault diagnosis method needs a large amount of data in different fault states to perform model training, and for large industrial equipment with a complex structure, a large amount of model training data is needed, and the model training time is long, and meanwhile, the fault diagnosis model can only represent the relationship between the fault state and the fault data, and cannot fully utilize the equipment configuration, the internal structure relationship, the information contained in the fault mode, and the like, so that the fault diagnosis precision is still low.

If a fault knowledge base can be constructed by utilizing the equipment configuration, the internal structure relationship, the information contained in the fault mode and the like, the intelligent fault diagnosis of the equipment fault can be realized by utilizing the fault knowledge base. In the prior art, a fault knowledge base is usually only formed by operation data of equipment during fault, and mutual correlation among the configuration, internal structure relationship, failure modes and the like of the equipment cannot be reflected, so that fault matching in a simple mode can be realized only by utilizing the fault knowledge base, fault occurrence sources and fault positions are difficult to accurately diagnose, and parts possibly influenced after the fault occurrence are difficult to position.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides a method and a device for constructing a fault knowledge base of industrial equipment, which have the advantages of simple implementation method, low cost and high data integrity, can fully reflect the correlation among equipment configuration, internal structure relationship and failure modes, and a fault diagnosis method and a fault diagnosis system which have high diagnosis precision and efficiency, can accurately diagnose the source of a fault and can position parts possibly influenced by the fault.

In order to solve the technical problems, the technical scheme provided by the invention is as follows:

a method for constructing a fault knowledge base of industrial equipment comprises the following steps:

acquiring an incidence relation between each component and a failure mode;

abstracting a graph model of each failure mode to construct failure mode nodes, constructing abnormal failure events as nodes according to the sequence of the layers from top to bottom to form a failure tree, constructing a basic graph model by using each component in the tested equipment as the nodes according to the structural relationship among the components, and connecting each node in the basic graph model and the failure tree with the corresponding failure mode nodes according to the incidence relationship to construct a failure knowledge system of the tested equipment;

and storing the fault knowledge system by using a graph database according to a preset logic model to obtain a fault knowledge base of the tested equipment.

Further, a causal logic relationship exists between the parent node and the child node in the fault tree, wherein the next layer of fault event is a cause of the previous layer of fault event.

Further, the logic model defines node labels and attributes for abnormal fault events, failure modes and components, and defines relationships between nodes in a fault knowledge system, wherein the relationships between the nodes include relationships within the same class of nodes and relationships between different classes of nodes.

Further, the node label includes the type of the abnormal failure event, the failure mode, the component and the subsystem, and the attribute includes the abnormal failure event ID, the failure mode ID, the component ID and the name.

Further, the relationship between the nodes includes a fault reason, a relationship of belonging, an energy transfer relationship between components, a relationship of fault and failure mode, and a relationship of failure mode and components, the fault reason corresponds to a relationship from a parent node to a child node in a fault tree, the relationship of belonging corresponds to a relationship from a child component to a total component, and the node of the component type includes an energy transfer relationship between components and a relationship of subordinate of the child component to the total component.

Further, the graph database is a Neo4j graph database.

A fault diagnosis method for industrial equipment comprises the following steps:

constructing a fault knowledge base of the tested equipment according to the fault knowledge base construction method;

when an abnormal fault event is detected, inputting the currently detected abnormal fault event into the fault knowledge base, searching leaf nodes corresponding to the node of the current abnormal fault event from the fault tree, determining all possible failure modes according to all the searched abnormal fault events and the corresponding relation between the abnormal fault event and the failure modes, and judging all components related to the current abnormal fault event according to all the determined failure modes and the relation between the failure modes and the components.

An industrial equipment failure knowledge base construction device comprises:

the system comprises an acquisition module, a failure detection module and a failure detection module, wherein the acquisition module is used for acquiring the failure domain knowledge of the tested equipment, and the failure domain knowledge comprises the incidence relation between each component and a failure mode;

the knowledge system building module is used for abstracting a graph model of each failure mode to build and form failure mode nodes, taking abnormal failure events as nodes and building and forming a failure tree according to the sequence of the layers from top to bottom, taking all the components in the tested equipment as nodes, building a basic graph model according to the structural relationship among the components, and connecting the basic graph model and all the nodes in the failure tree with the corresponding failure mode nodes according to the incidence relationship to build and form a failure knowledge system of the tested equipment;

and the database storage module is used for storing the fault knowledge system by using a database according to a preset logic model to obtain a fault knowledge base of the tested equipment.

An industrial equipment failure knowledge base building comprises a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program to execute the method.

An industrial equipment fault diagnostic system comprising:

the fault knowledge base building device is used for building the fault knowledge base;

and the diagnosis device is used for inputting the currently detected abnormal fault event into the fault knowledge base of the fault knowledge base construction device when the abnormal fault event is detected, finding out leaf nodes corresponding to the node of the current abnormal fault event from the fault tree, determining all possible failure modes according to all found abnormal fault events and the corresponding relation between the abnormal fault event and the failure modes, and judging all components related to the current abnormal fault event according to all determined failure modes and the relation between the failure modes and the components.

Compared with the prior art, the invention has the advantages that:

1. the invention constructs a fault tree for nodes based on abnormal fault events, the nodes of the fault tree are respectively and correspondingly connected with failure mode nodes to establish the relation between a basic event and a failure mode, simultaneously, each component in the tested equipment is used as a node to construct a basic graph model, each node in the basic graph model is respectively and correspondingly connected with the failure mode nodes to form a fault knowledge system with an incidence relation network between the abnormal fault events, the fault tree, the failure mode and the components, and the fault domain knowledge and the equipment mechanism structure can be compatible in the same knowledge system, thereby realizing the organic combination and unified representation of the fault domain knowledge and the equipment representation method, simultaneously realizing the complex corresponding relation between the components and the abnormal events, fully representing the abnormal fault events caused by covering multiple factors or the problems of multiple components, and the situation that the fault problems crossing the system level have complex coupling cause-effect relation.

2. According to the method, the graph database is used as a physical model for realizing the knowledge base, so that the relevance between abnormal fault events, failure modes and components can be completely represented, the relevant failure modes and the components can be quickly obtained based on the abnormal fault events, the graph retrieval performance and the expansibility advantages of the graph database can be exerted, and the fault field knowledge base suitable for complex industrial equipment is constructed;

3. the invention further determines the corresponding abnormal fault event after the tested device generates the abnormal state triggering alarm, and inputs the abnormal fault event into the established fault knowledge base, namely, the initial source of the fault and the associated parts possibly influenced by the prediction can be analyzed quickly and accurately by utilizing the associated relation among the fault tree, the failure mode and the parts established by the abnormal fault event.

Drawings

Fig. 1 is a schematic flow chart illustrating a principle of implementing fault diagnosis based on a fault knowledge base of an industrial device according to the present embodiment.

Fig. 2 is a schematic flow chart of an implementation of the method for building the fault knowledge base of the industrial equipment according to the embodiment.

Fig. 3 is a schematic diagram of the principle of constructing a fault tree based on abnormal fault events according to the present embodiment.

Fig. 4 is a schematic diagram of the effect of modeling when the present invention is applied to a lubrication system in a specific application embodiment.

FIG. 5 is a schematic diagram of nodes and relationships of a power subsystem stored in a graph database in an embodiment of the invention.

Fig. 6 is a schematic flow chart of implementation of fault diagnosis of the industrial equipment according to the embodiment.

Detailed Description

The invention is further described below with reference to the drawings and specific preferred embodiments of the description, without thereby limiting the scope of protection of the invention.

As shown in fig. 1, an abnormal fault event and a fault diagnosis analysis process may form a complete loop, for example, when a module (component for short) representing a subsystem or a component in equipment in a large-scale industrial equipment (for example, a locomotive) has a fault, it is shown that a certain failure mode related to the component affects a corresponding function, an abnormal value may be found by calculating an index in a component health state evaluation system through monitored operation data, and an abnormal event alarm is triggered, so that a fault analysis and diagnosis link is started, and fault analysis, positioning and prediction are performed based on a constructed fault knowledge base.

As shown in fig. 2, the steps of the method for constructing the knowledge base of the industrial equipment failure in the embodiment include:

s01, acquiring fault domain knowledge of the tested equipment, wherein the fault domain knowledge comprises an incidence relation between each component and a failure mode;

s02, abstracting a graph model of each failure mode to construct failure mode nodes, constructing and forming a failure tree by taking abnormal failure events as nodes and according to the sequence of layers from top to bottom, constructing a basic graph model by taking all parts in the tested equipment as nodes and according to the structural relationship among the parts, and respectively connecting each node in the basic graph model and the failure tree with the corresponding failure mode nodes according to the incidence relationship to construct and form a failure knowledge system of the tested equipment;

and S03, storing a fault knowledge system by using a graph database according to a preset logic model to obtain a fault knowledge base of the tested equipment.

In the embodiment, structured and hierarchical equipment configuration and internal structure are used as core bases of fault knowledge data storage management, fault domain knowledge is abstracted into a topological structure on an equipment system structure block diagram and abstracted into an independent graph structure which is linked with the system structure block diagram, a fault tree is constructed based on uniformly defined abnormal fault events, each node in the fault tree corresponds to one abnormal fault event, the nodes of the fault tree are respectively and correspondingly connected with a failure mode node to link the basic events with the failure modes, meanwhile, each component in the tested equipment is used as a node to construct a basic graph model, each node in the basic graph model is respectively and correspondingly connected with the failure mode node to establish the association relationship between each component and the failure modes, a fault knowledge system with the association relationship network between the abnormal fault events, the fault tree, the failure modes and the components is formed, the fault domain knowledge and the equipment mechanism structure can be compatible in the same knowledge system, so that the characterization methods of the fault domain knowledge and the equipment are organically combined and uniformly characterized, and the complex correspondence between the components and the fault events can be realized, the coverage factors or the problem of multiple fault events and the problem of the fault domain knowledge and the problem of the fault knowledge which is caused by the fault knowledge and the problem of the cross-over-level coupling system exist.

In the embodiment, a storage mode of a graph database is further combined, the graph database is used as a physical model for realizing the knowledge base, the relevance between an abnormal fault event and a failure mode and components can be completely represented, the relevant failure mode and the components can be quickly obtained based on the abnormal fault event, the graph retrieval performance and the expansibility advantages of the graph database can be simultaneously exerted, and the failure domain knowledge base suitable for complex industrial equipment is constructed.

The failure mode of the embodiment can be obtained by analyzing the related records of the component-failure mode analysis table and the record information of the component potential failure mode analysis table, and the failure mode and the association relationship between the potential failure mode and the component can be determined based on the analysis table. The component-failure mode relation table is relatively clear and formalized failure domain knowledge in the system, after the relation between failure modes and components is determined by utilizing the relation table, a failure knowledge base is constructed and formed according to the method, and after an abnormal failure event occurs, the relation can be connected to the specific components of the equipment through the relation of the abnormal failure event-failure tree-failure mode-component relation, so that intelligent failure diagnosis and analysis are realized. The potential failure mode analysis table may be an item that is evaluated as important or risky.

Because one component may correspond to multiple failure modes, if a failure mode is used as an attribute on a node of the basic graph model, the convenience of query is affected when one node attribute contains multiple failure modes, and therefore, the failure modes are abstracted into the graph model and then are used as independent nodes, and then are connected to the basic graph model through the corresponding relationship between the component and the failure modes. In a specific application embodiment, a function-component-failure mode data relationship model is generated through the component-failure mode relationship table and the potential failure mode table, and then a failure mode graph model is abstracted and used as an independent node to be connected with each component node and a fault tree node in the basic graph model.

In the embodiment, the graph model is used for representing the concept models of the tested device structure, the fault tree and the failure mode, and the fault tree is used as the graph structure independent of the basic graph model because no direct corresponding relation exists between the nodes and the components of the fault tree. In the embodiment, for the construction of the fault tree, firstly, an evaluation index system of the health state of the device under test components is established, and on the basis, abnormal fault events are uniformly defined according to the abnormal state, and are used as basic component events (i.e., nodes) of the fault tree, and after system level analysis from top to bottom, the abnormal fault events are integrated to build the fault tree, as shown in fig. 3, a causal logical relationship exists between a parent node and a child node in the fault tree, wherein the fault event at the lower layer is a cause of the fault event at the upper layer. Since the exception event is defined such that consistency with the failure mode is preserved, the nodes of the fault tree can establish corresponding edges with the failure mode nodes.

Taking the lubrication system applied to locomotive equipment in a specific application embodiment as an example, as shown in fig. 4, the left side is a lubrication system component structure diagram, a basic diagram structure is constructed and formed according to the component structure diagram, the upper part of the right side is a constructed fault tree, the lower part is constructed failure mode nodes, each failure mode node is correspondingly connected with each node in the fault tree, each failure mode node is also correspondingly connected with each component in the lubrication system, that is, a one-to-many relationship exists between the basic diagram model representing the component structure relationship and the failure mode nodes, and a many-to-many relationship exists between the failure mode nodes and the nodes of the fault tree. It should be noted that the nodes of the failure modes in fig. 4 are distributed scattered because the currently owned failure modes are independent, and if the failure modes are applied to other systems, the nodes of the failure modes may be organized by the association or hierarchical relationship between the failure modes.

In the logic model of this embodiment, node labels and attributes are defined for abnormal fault events, failure modes, and components, and relationships between nodes in a fault knowledge system are defined, where the relationships between nodes include relationships within the same class of nodes and relationships between different classes of nodes. The node tag specifically includes a type of an abnormal fault event, a failure mode, a component, and a subsystem, and the attribute specifically includes an abnormal fault event ID, a failure mode ID, a component ID, and a name. The relationship among the nodes comprises a fault reason, an belonging relationship, an energy transfer relationship among the components, a fault and failure mode related relationship and a failure mode and component related relationship, the fault reason corresponds to the relationship from a parent node to a child node in a fault tree, the belonging relationship corresponds to the relationship from the child component to a general component, and the node of the component type comprises the energy transfer relationship among the components and the subordinate relationship from the child component to the general component. Taking the lubrication system applied to locomotive equipment in a specific application example as an example, the configuration of the logic model is shown in table 1.

The logical model of the failure knowledge base needs to define the types and attributes of the nodes and edges of the graph model, in this embodiment, a Neo4j graph database is specifically used for storing a failure knowledge system, and in the Neo4j graph database, the types of the nodes are defined by using tags, so that one node can be endowed with a plurality of tags. Both nodes and relationships have attributes and can be queried by attribute values. Specifically, firstly, respectively defining a label for a fault event, a failure mode and a system component so as to distinguish the three types of nodes; also defined is a label representing a subsystem (such as the lubrication system described above), then a component node belonging to the subsystem possesses both labels representing the component and the lubrication system. In this embodiment, the subsystem is used as a tag instead of an attribute value of a node, on one hand, redundancy caused by storing subsystem information through a node attribute can be avoided, on the other hand, a hierarchical relationship of a system structure can be embodied through multiple tags, and a required node can be conveniently and quickly searched in different subsystems, for example, a component can be searched in a case where the subsystem to which the node belongs is determined, a node set having the tag can be searched through the tag of the subsystem, and the attribute value of the node set serving as the attribute value of the subsystem can be searched in all nodes as the node of the subsystem and meeting the search condition.

TABLE 1 locomotive lubrication System logic model

The embodiment defines a corresponding Id for each type of node specifically in the attribute, and may set a unique identifier to the node, forming a primary key similar to that in a relational database. While the Neo4j graph database provides each node with an automatically generated Id, neo4j reuses the automatic ids of deleted nodes, and this embodiment can form unique identifiers for nodes while avoiding Id reuse of deleted nodes by defining corresponding ids for each class of nodes. Names in attributes are common attributes for all nodes and relationships. Besides the above attribute information, other attribute information may be configured according to actual requirements. If the information of the fault measure exists, the attribute can be added into the node, so that the inquiry node can obtain suggestions for dealing with the fault.

The relationship of the nodes in the logic model in the embodiment includes the relationship inside the same type of nodes and the relationship between different types of nodes, wherein the nodes of the component type have both the energy transfer relationship between the components and the dependency relationship from the sub-components to the total component. If the magnitude of energy transferred between the parts is to be recorded, attributes representing magnitude can be added to the flow relationship; if there is more detailed information in the faulty tree, such as the causal correlation degree between different fault events, the attribute representing the influence or probability can be added in the replay relationship; if there is probability information between the component and the failure mode, an attribute representing the probability may be further added to the relationship between the two, so that a qualitative or quantitative analysis result is provided at the time of failure retrieval.

After the logical model is determined according to the steps, data abstracted from the conceptual model are further stored in a graph database according to a mode defined by the logical model. As shown in fig. 5, the nodes and relationships of the power subsystem stored in the graph database are shown, wherein the failure event nodes on the left side form a failure tree, the failure mode nodes are in the middle, the component nodes are on the right side, and a mesh structure is formed among the failure, failure mode and component nodes.

In the embodiment, the failure domain knowledge is stored in the graph database to form a failure knowledge base, structured data such as non-key attribute information, an equipment historical model and monitoring data of the failure domain knowledge are stored in the relational database, a connection is established through a node identifier (such as a node Id and a failure signature), and the failure knowledge base can be applied to a failure diagnosis and analysis platform based on the failure knowledge base so as to perform failure diagnosis and analysis by combining monitoring data of equipment, and further can realize application based on the knowledge base, such as failure retrieval, analysis, prediction, display and the like.

As shown in fig. 6, the steps of the method for diagnosing the fault of the industrial equipment in the embodiment include:

constructing a fault knowledge base of the tested equipment according to the fault knowledge base construction method;

when an abnormal fault event is detected, inputting the currently detected abnormal fault event into a fault knowledge base, searching leaf nodes corresponding to the nodes of the current abnormal fault event from a fault tree, determining all possible failure modes according to all the searched abnormal fault events and the corresponding relation between the abnormal fault event and the failure modes, and judging all components related to the current abnormal fault event according to all the determined failure modes and the relation between the failure modes and the components.

In the embodiment, after the abnormal state of the device to be tested triggers an alarm, the corresponding abnormal fault event is determined, and the abnormal fault event is input into the established fault knowledge base, so that the initial source of the fault and the associated components possibly influenced by prediction can be quickly and accurately analyzed by utilizing the association relation among the fault tree, the failure mode and the components established by the abnormal fault event.

Because the fault tree in the fault knowledge base is composed of abnormal events in a top-down mode, leaf nodes of the fault tree are the most basic fault events, causal logical relations exist between parent nodes and child nodes of the fault tree, the bottom layer fault event is the cause of the upper layer fault event, and the fault source searching problem is converted into the problem that the leaf nodes under the nodes are searched by the nodes (input abnormal events/fault identifiers) of a given fault tree, namely the problem of graph traversal searching is solved. Specifically, after an abnormal fault event of the system is input in the fault knowledge base, according to the structure of the fault knowledge base, leaf nodes under the node of the abnormal event are first searched in the fault tree, the node of the fault tree is linked with the failure mode through a representative basic event, the fault root of the abnormal event is found through the fault tree, and then a possibly-affected component is presumed by combining the corresponding relationship between the failure mode and the component.

The device for constructing the industrial equipment fault knowledge base comprises:

the acquisition module is used for acquiring the fault domain knowledge of the tested equipment, wherein the fault domain knowledge comprises the incidence relation between each component and the failure mode;

the knowledge system building module is used for abstracting a graph model of each failure mode to build and form failure mode nodes, taking abnormal failure events as nodes and building and forming a failure tree according to the sequence of the layers from top to bottom, taking all the components in the tested equipment as nodes, building a basic graph model according to the structural relationship among the components, and respectively connecting all the nodes in the basic graph model and the failure tree with the corresponding failure mode nodes according to the incidence relationship to build and form a failure knowledge system of the tested equipment;

and the graph database storage module is used for storing the fault knowledge system by using a graph database according to a preset logic model to obtain a fault knowledge base of the tested equipment.

The device for constructing the fault knowledge base of the industrial equipment corresponds to the method for constructing the fault knowledge base of the industrial equipment one by one, and the details are not repeated herein.

In another embodiment, the fault knowledge base of the industrial equipment can be constructed by the following steps: comprising a processor and a memory for storing a computer program, the processor being adapted to execute the computer program to perform the method as described above.

The industrial equipment fault diagnosis system comprises:

the fault knowledge base building device is used;

and the diagnosis device is used for inputting the currently detected abnormal fault event into the fault knowledge base of the fault knowledge base construction device when the abnormal fault event is detected, finding out leaf nodes corresponding to the node of the current abnormal fault event from the fault tree, determining all possible failure modes according to all the found abnormal fault events and the corresponding relation between the abnormal fault event and the failure modes, and judging all parts related to the current abnormal fault event according to all the determined failure modes and the relation between the failure modes and the parts.

The industrial equipment fault diagnosis system of the embodiment corresponds to the industrial equipment fault diagnosis method, and the industrial equipment fault diagnosis system and the industrial equipment fault diagnosis method have the same principle and effect.

The foregoing is considered as illustrative of the preferred embodiments of the invention and is not to be construed as limiting the invention in any way. Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention should fall within the protection scope of the technical scheme of the present invention, unless the technical spirit of the present invention departs from the content of the technical scheme of the present invention.

Claims (10)

1. A method for constructing a fault knowledge base of industrial equipment is characterized by comprising the following steps:

acquiring the incidence relation between each component and the failure mode;

abstracting a graph model of each failure mode to construct failure mode nodes, constructing and forming a failure tree by taking abnormal failure events as nodes and according to the sequence of the layers from top to bottom, constructing a basic graph model by taking all parts in the tested equipment as nodes and according to the structural relationship among the parts, and respectively connecting the basic graph model and all nodes in the failure tree with the corresponding failure mode nodes according to the incidence relationship to construct and form a failure knowledge system of the tested equipment;

and storing the fault knowledge system by using a graph database according to a preset logic model to obtain a fault knowledge base of the tested equipment.

2. The method for building the industrial equipment fault knowledge base according to claim 1, wherein a causal logic relationship exists between a parent node and a child node in the fault tree, and a next-layer fault event is a cause of a previous-layer fault event.

3. The method for building the industrial equipment fault knowledge base according to claim 1, wherein node labels and attributes are defined for abnormal fault events, failure modes and components in the logic model, and relationships between nodes in a fault knowledge system are defined, and the relationships between the nodes comprise relationships inside the same type of nodes and relationships between different types of nodes.

4. The method for building the industrial equipment fault knowledge base according to claim 3, wherein the node tags comprise types of abnormal fault events, failure modes, components and subsystems, and the attributes comprise an abnormal fault event ID, a failure mode ID, a component ID and a name.

5. The method for building the industrial equipment fault knowledge base according to claim 3, wherein the relationship among the nodes comprises a fault reason, an belonging relationship, an energy transfer relationship among components, a fault and failure mode related relationship and a failure mode and component related relationship, the fault reason corresponds to a relationship from a parent node to a child node in a fault tree, the belonging relationship corresponds to a relationship from a child component to a main component, and the node of the component type comprises the energy transfer relationship among the components and a subordinate relationship from the child component to the main component.

6. The method for constructing the knowledge base of the faults of the industrial equipment as claimed in any one of claims 1 to 5, wherein the graph database is a Neo4j graph database.

7. A fault diagnosis method for industrial equipment is characterized by comprising the following steps:

constructing a fault knowledge base of the tested equipment according to the fault knowledge base construction method of any one of claims 1 to 6;

when an abnormal fault event is detected, inputting the currently detected abnormal fault event into the fault knowledge base, searching leaf nodes corresponding to the node of the current abnormal fault event from the fault tree, determining all possible failure modes according to all the searched abnormal fault events and the corresponding relation between the abnormal fault event and the failure modes, and judging all components related to the current abnormal fault event according to all the determined failure modes and the relation between the failure modes and the components.

8. An industrial equipment failure knowledge base construction device is characterized by comprising:

the system comprises an acquisition module, a failure detection module and a failure detection module, wherein the acquisition module is used for acquiring the failure domain knowledge of the tested equipment, and the failure domain knowledge comprises the incidence relation between each component and a failure mode;

the knowledge system building module is used for abstracting a graph model of each failure mode to build and form failure mode nodes, taking abnormal failure events as nodes and building and forming a failure tree according to the sequence of the layers from top to bottom, taking all the components in the tested equipment as nodes, building a basic graph model according to the structural relationship among the components, and connecting the basic graph model and all the nodes in the failure tree with the corresponding failure mode nodes according to the incidence relationship to build and form a failure knowledge system of the tested equipment;

and the database storage module is used for storing the fault knowledge system by using a database according to a preset logic model to obtain a fault knowledge base of the tested equipment.

9. An industrial equipment failure knowledge base construction apparatus comprising a processor and a memory, the memory being configured to store a computer program, wherein the processor is configured to execute the computer program to perform the method according to any one of claims 1 to 7.

10. An industrial equipment fault diagnostic system, comprising:

the failure knowledge base constructing apparatus according to claim 8 or 9;

and the diagnosis device is used for inputting the currently detected abnormal fault event into the fault knowledge base of the fault knowledge base construction device when the abnormal fault event is detected, finding out leaf nodes corresponding to the node of the current abnormal fault event from the fault tree, determining all possible failure modes according to all found abnormal fault events and the corresponding relation between the abnormal fault event and the failure modes, and judging all components related to the current abnormal fault event according to all determined failure modes and the relation between the failure modes and the components.

Images