A knowledge graph construction method for industrial anomaly detection and solution

Abstract

The present invention proposes a method for constructing an industrial anomaly detection and solution knowledge graph, including the following steps: obtaining enterprise equipment data; extracting entities and relationships between entities from the enterprise equipment data to construct an enterprise equipment knowledge graph; obtaining phenomenon data occurring during the production process of enterprise products, and establishing a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment graph and the phenomenon data; connecting the enterprise equipment knowledge graph and the logical relationship network through the fusion of nodes at the physical level and nodes at the logical level, establishing a data chain of phenomenon->cause->solution, and completing the enterprise equipment knowledge graph; and performing anomaly detection in industrial production and obtaining corresponding solutions based on the completed enterprise equipment knowledge graph.

Description

Technical Field

[0001] The present invention belongs to the technical field of big data processing. Background Art

[0002] Existing knowledge graph-based enterprise risk prediction methods are hampered by the fact that, during the production process, not only will various machine parameters change due to factors such as long-term operation, but this can also lead to changes in the appearance and even quality of the products produced. Furthermore, due to a lack of professional knowledge, the high dimensionality of enterprise equipment parameter data, and the large number of products produced, there is currently no effective method for real-time control of enterprise equipment commissioning. This is because regular or timely large-scale equipment commissioning during the production process consumes too much manpower and resources, and there is no early warning of potential equipment production risks within the enterprise itself or its affiliated companies. Consequently, the enterprise is unable to effectively manage its equipment, which can result in significant losses.

[0003] Since the knowledge graph-related technical system was proposed not long ago, the entire technical system has not yet been perfected. However, the predictive analysis and adaptive processing of various equipment parameters as enterprise production equipment changes or iterates with products have become increasingly important for corporate interests and national development. Companies including Google, Baidu, Alibaba, etc. are committed to providing better services to enterprises through knowledge graph technology. However, with the rapid growth of enterprise scale, some urgent problems in actual application have also been exposed. Among them, the demand for predictive analysis of enterprise equipment management is particularly significant. The industry is almost blank in using knowledge graphs to perform predictive analysis and processing of enterprise equipment management. Therefore, it is particularly urgent to build enterprise knowledge graphs and perform predictive analysis and adaptive processing of enterprise equipment management. Summary of the Invention

[0004] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, the purpose of the present invention is to propose a method for constructing an industrial anomaly detection and solution knowledge graph, which is used to solve the problems of high cost, low efficiency, high threshold and low timeliness in enterprise equipment management prediction analysis and processing, and to realize automatic query and self-diagnosis of the data chain of "phenomenon->cause->solution" that occurs during the generation process.

[0006] To achieve the above objectives, the first embodiment of the present invention proposes a method for constructing an industrial anomaly detection and solution knowledge graph, including:

[0007] Obtain enterprise device data;

[0008] Extracting entities and relationships between entities from the enterprise equipment data to construct an enterprise equipment knowledge graph;

[0009] Acquire phenomenon data that occurs during the production process of enterprise products, and establish a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data;

[0010] By integrating the nodes at the physical level with the nodes at the logical level, the enterprise equipment knowledge graph and the logical relationship network are connected, a data chain of phenomenon->cause->solution is established, and the enterprise equipment knowledge graph is completed;

[0011] Based on the completed enterprise equipment knowledge graph, anomaly detection in industrial production is performed and corresponding solutions are obtained.

[0012] In addition, the method for constructing an industrial anomaly detection and solution knowledge graph according to the above embodiment of the present invention may also have the following additional technical features:

[0013] Furthermore, in one embodiment of the present invention, obtaining parameter data of enterprise equipment further includes:

[0014] The parameter data is preprocessed, including cleaning, parsing and standardization.

[0015] Furthermore, in one embodiment of the present invention, entities and relationships between entities are extracted from the parameter data to construct an enterprise equipment knowledge graph, including:

[0016] Design nodes and relationships in the enterprise equipment knowledge graph; node design includes node labels <node-name> : <label-name>, node attributes { <property-name> : <property-value>}, the node supports multiple tags;

[0017] Relationship design includes relationship tags: CREATE( <node1-name> : <label1-name> )-

[0018] [( <relationship-name> : <relationship-label-name> )]->( <node2-name> : <label2-name>),

[0019] Relationship attributes: [( <relationship-name> : <relationship-label-name>); the relationship includes a unidirectional relationship and a bidirectional relationship.

[0020] Furthermore, in one embodiment of the present invention, the step of acquiring phenomenon data occurring during the production process of enterprise products and establishing a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data includes:

[0021] Obtaining device component parameter attribute data in the enterprise knowledge device graph through the Cy query language;

[0022] Obtaining equipment component parameter attribute data when phenomenon data is abnormal;

[0023] A logical relationship network between enterprise products and enterprise equipment is established based on abnormal phenomenon data and corresponding equipment component parameter attribute data.

[0024] Furthermore, in one embodiment of the present invention, the integration of physical-level nodes and logical-level nodes realizes the connection between the enterprise equipment knowledge graph and the edit relationship network, establishes a data chain of phenomenon->cause->solution, and completes the enterprise equipment knowledge graph, including:

[0025] Design the nodes and relationships in the enterprise equipment knowledge graph after completion;

[0026] Node design includes node labels and node attributes, among which,

[0027] The node labels include logical level: cause, phenomenon, solution, and physical level: product, equipment, unit, component, carrier, and fixture;

[0028] The node attributes include: cause, phenomenon, and solution;

[0029] Relationship design includes causes, phenomena and solutions, among which

[0030] The reasons include: hardware abnormality, software abnormality, configuration abnormality, and environmental abnormality;

[0031] The phenomenon includes: fault code, fault description, and product defect;

[0032] The solution includes: actions and parameter adjustments.

[0033] Furthermore, in one embodiment of the present invention, the performing of anomaly detection in industrial production and obtaining corresponding solutions based on the completed enterprise equipment knowledge graph includes:

[0034] According to the changes in product properties, we explore solutions to change the property values ​​of equipment components to ensure the normal operation of enterprise equipment.

[0035] To achieve the above objectives, the second embodiment of the present invention proposes an industrial anomaly detection and solution knowledge graph construction device, including the following modules:

[0036] Acquisition module, used to obtain enterprise equipment data;

[0037] A first building module is used to extract entities and relationships between entities from the enterprise device data to build an enterprise device knowledge graph;

[0038] The second building module acquires phenomenon data that occurs during the production process of enterprise products, and establishes a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data;

[0039] A fusion module is used to connect the enterprise equipment knowledge graph and the logical relationship network by fusing the nodes at the physical layer with the nodes at the logical layer, establish a data chain of phenomenon->cause->solution, and complete the enterprise equipment knowledge graph;

[0040] A detection module is used to detect anomalies in industrial production and obtain corresponding solutions based on the completed enterprise equipment knowledge graph.

[0041] To achieve the above-mentioned purpose, the third aspect of the present invention proposes a computer device, which is characterized in that it includes a memory, a processor, and a computer program stored on the memory and runnable on the processor. When the processor executes the computer program, it implements the above-mentioned method for constructing an industrial anomaly detection and solution knowledge graph.

[0042] To achieve the above-mentioned purpose, the fourth aspect of the present invention proposes a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, it implements the above-mentioned industrial anomaly detection and solution knowledge graph construction method.

[0043] The industrial anomaly detection and solution knowledge graph construction method proposed in this embodiment connects the enterprise equipment knowledge graph and logical relationship network by integrating nodes at the physical layer with nodes at the logical layer. This establishes a data chain from phenomenon to cause to solution, thus completing the enterprise equipment knowledge graph. After completing the corresponding knowledge graph network, the system can promptly query the cause based on any attribute changes in the product. Based on the cause, it can explore solutions to modify the attribute values ​​of equipment components to ensure the normal operation of the enterprise equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0045] Figure 1 A flowchart of a method for constructing an industrial anomaly detection and solution knowledge graph provided by an embodiment of the present invention.

[0046] Figure 2 An industrial knowledge graph relationship diagram provided by an embodiment of the present invention.

[0047] Figure 3 A schematic flow chart of an industrial anomaly detection and solution knowledge graph construction device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0049] The following describes the industrial anomaly detection and solution knowledge graph construction method according to an embodiment of the present invention with reference to the accompanying drawings.

[0050] Figure 1 A flowchart of a method for constructing an industrial anomaly detection and solution knowledge graph provided by an embodiment of the present invention.

[0051] like Figure 1 As shown in FIG, the method for constructing a knowledge graph for industrial anomaly detection and solutions includes the following steps:

[0052] S101: Obtain enterprise device data;

[0053] S102: Extracting entities and relationships between entities from the enterprise equipment data to construct an enterprise equipment knowledge graph;

[0054] S103: Acquire phenomenon data that occurs during the production process of enterprise products, and establish a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data;

[0055] S104: Connecting the enterprise equipment knowledge graph and the logical relationship network by integrating the nodes at the physical layer with the nodes at the logical layer, establishing a data chain of phenomenon->cause->solution, and completing the enterprise equipment knowledge graph;

[0056] S105: Perform anomaly detection in industrial production based on the completed enterprise equipment knowledge graph and obtain corresponding solutions.

[0057] Furthermore, in one embodiment of the present invention, obtaining parameter data of enterprise equipment further includes:

[0058] The parameter data is preprocessed, including cleaning, parsing and standardization.

[0059] Furthermore, in one embodiment of the present invention, entities and relationships between entities are extracted from the parameter data to construct an enterprise equipment knowledge graph, including:

[0060] Design nodes and relationships in the enterprise equipment knowledge graph; node design includes node labels <node-name> : <label-name>, node attributes { <property-name> : <property-value>}, the node supports multiple tags;

[0061] Relationship design includes relationship tags: CREATE( <node1-name> : <label1-name> )-

[0062] [( <relationship-name> : <relationship-label-name> )]->( <node2-name> : <label2-name>),

[0063] Relationship attributes: [( <relationship-name> : <relationship-label-name>); the relationship includes a unidirectional relationship and a bidirectional relationship.

[0064] For example: node label create(emp:Employee),

[0065] Node attributes: create(dep:Department{dno:10,dname:"cs",…}),

[0066] Relationship tag: CREATE(p1:Profile1)-[r1:LIKES]->(p2:Profile2).

[0067] Furthermore, in one embodiment of the present invention, the step of acquiring phenomenon data occurring during the production process of enterprise products and establishing a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data includes:

[0068] Obtaining device component parameter attribute data in the enterprise knowledge device graph through the Cy query language;

[0069] Obtaining equipment component parameter attribute data when phenomenon data is abnormal;

[0070] A logical relationship network between enterprise products and enterprise equipment is established based on abnormal phenomenon data and corresponding equipment component parameter attribute data.

[0071] Furthermore, in one embodiment of the present invention, the integration of physical-level nodes and logical-level nodes realizes the connection between the enterprise equipment knowledge graph and the edit relationship network, establishes a data chain of phenomenon->cause->solution, and completes the enterprise equipment knowledge graph, including:

[0072] Design the nodes and relationships in the enterprise equipment knowledge graph after completion;

[0073] Node design includes node labels and node attributes, among which,

[0074] The node labels include logical level: cause, phenomenon, solution, and physical level: product, equipment, unit, component, carrier, and fixture;

[0075] The node attributes include: cause, phenomenon, and solution;

[0076] Relationship design includes causes, phenomena and solutions, among which

[0077] The reasons include: hardware abnormality, software abnormality, configuration abnormality, and environmental abnormality;

[0078] The phenomenon includes: fault code, fault description, and product defect;

[0079] The solution includes: actions and parameter adjustments.

[0080] like Figure 2 The following is a relationship diagram of the industrial knowledge graph.

[0081] Furthermore, in one embodiment of the present invention, the performing of anomaly detection in industrial production and obtaining corresponding solutions based on the completed enterprise equipment knowledge graph includes:

[0082] According to the changes in product properties, we explore solutions to change the property values ​​of equipment components to ensure the normal operation of enterprise equipment.

[0083] The industrial anomaly detection and solution knowledge graph construction method proposed in this embodiment connects the enterprise equipment knowledge graph and logical relationship network by integrating nodes at the physical layer with nodes at the logical layer. This establishes a data chain from phenomenon to cause to solution, thus completing the enterprise equipment knowledge graph. After completing the corresponding knowledge graph network, the system can promptly query the cause based on any attribute changes in the product. Based on the cause, it can explore solutions to modify the attribute values ​​of equipment components to ensure the normal operation of the enterprise equipment.

[0084] In order to implement the above embodiments, the present invention also proposes an industrial anomaly detection and solution knowledge graph construction device.

[0085] Figure 3 A schematic diagram of the structure of an industrial anomaly detection and solution knowledge graph construction device provided by an embodiment of the present invention.

[0086] like Figure 3 As shown, the industrial anomaly detection and solution knowledge graph construction device includes: an acquisition module 100, a first construction module 200, a second construction module 300, a fusion module 400, and a detection module 500, wherein:

[0087] Acquisition module, used to obtain enterprise equipment data;

[0088] A first building module is used to extract entities and relationships between entities from the enterprise device data to build an enterprise device knowledge graph;

[0089] The second building module acquires phenomenon data that occurs during the production process of enterprise products, and establishes a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data;

[0090] A fusion module is used to connect the enterprise equipment knowledge graph and the logical relationship network by fusing the nodes at the physical layer with the nodes at the logical layer, establish a data chain of phenomenon->cause->solution, and complete the enterprise equipment knowledge graph;

[0091] A detection module is used to detect anomalies in industrial production and obtain corresponding solutions based on the completed enterprise equipment knowledge graph.

[0092] To achieve the above-mentioned purpose, the third aspect of the present invention proposes a computer device, which is characterized in that it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the industrial anomaly detection and solution knowledge graph construction method as described above.

[0093] To achieve the above-mentioned purpose, the fourth aspect of the present invention proposes a computer-readable storage medium on which a computer program is stored, characterized in that when the computer program is executed by a processor, it implements the industrial anomaly detection and solution knowledge graph construction method as described above.

[0094] In the description of this specification, reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0095] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature identified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0096] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limiting the present invention. A person skilled in the art may change, modify, replace, and modify the above embodiments within the scope of the present invention. < / relationship-name> < / node2-name> < / relationship-label-name> < / relationship-name> < / label1-name> < / node1-name> < / property-name> < / node-name> < / relationship-name> < / node2-name> < / relationship-label-name> < / relationship-name> < / label1-name> < / node1-name> < / property-name> < / node-name>

Claims

1. A method for constructing a knowledge graph for industrial anomaly detection and solutions, characterized in that: The following steps are involved: Obtain enterprise device data; Extracting entities and relationships between entities from the enterprise equipment data to construct an enterprise equipment knowledge graph; Acquire phenomenon data that occurs during the production process of enterprise products, and establish a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data; By integrating the nodes at the physical level with the nodes at the logical level, the enterprise equipment knowledge graph and the logical relationship network are connected, a data chain of phenomenon->cause->solution is established, and the enterprise equipment knowledge graph is completed; Detect anomalies in industrial production and obtain corresponding solutions based on the completed enterprise equipment knowledge graph; The step of acquiring phenomenon data occurring during the production process of enterprise products and establishing a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data includes: Obtaining device component parameter attribute data in the enterprise knowledge device graph through the Cy query language; Obtaining equipment component parameter attribute data when phenomenon data is abnormal; Establish a logical relationship network between enterprise products and enterprise equipment based on abnormal phenomenon data and corresponding equipment component parameter attribute data; The integration of the physical layer nodes and the logical layer nodes realizes the connection of the enterprise equipment knowledge graph and the edit relationship network, establishes a data chain of phenomenon->cause->solution, and completes the enterprise equipment knowledge graph, including: Design the nodes and relationships in the enterprise equipment knowledge graph after completion; Node design includes node labels and node attributes, among which, The node labels include logical level: cause, phenomenon, solution, and physical level: product, equipment, unit, component, carrier, and fixture; The node attributes include: cause, phenomenon, and solution; Relationship design includes causes, phenomena and solutions, among which The reasons include: hardware abnormality, software abnormality, configuration abnormality, and environmental abnormality; The phenomenon includes: fault code, fault description, and product defect; The solution includes: actions and parameter adjustments.

2. The method according to claim 1, characterized in that The obtaining of parameter data of enterprise equipment further includes: The parameter data is preprocessed, including cleaning, parsing and standardization.

3. The method according to claim 1, characterized in that Extracting entities and relationships between entities from the parameter data to construct an enterprise equipment knowledge graph, including: Design nodes and relationships in the enterprise equipment knowledge graph; node design includes node labels <node-name> : <label-name>, node attributes { <property-name> : <property-value>}, the node supports multiple tags; < / property-name> < / node-name> Relationship design includes relationship tags: CREATE ( <node1-name> : <label1-name> )-< / label1-name> < / node1-name> [( <relationship-name> : <relationship-label-name> )]->( <node2-name> : <label2-name> )，< / label2-name> < / node2-name> < / relationship-label-name> < / relationship-name> Relationship attributes: [( <relationship-name> : <relationship-label-name> ); the relationship includes a unidirectional relationship and a bidirectional relationship. < / relationship-name> 4. The method according to claim 1, wherein The method of detecting anomalies in industrial production and obtaining corresponding solutions based on the completed enterprise equipment knowledge graph includes: According to the changes in product properties, we explore solutions to change the property values ​​of equipment components to ensure the normal operation of enterprise equipment.

5. An industrial anomaly detection and solution knowledge graph construction device, characterized in that: The device implements the method according to claim 1, and the device includes the following modules: Acquisition module, used to obtain enterprise equipment data; A first building module is used to extract entities and relationships between entities from the enterprise device data to build an enterprise device knowledge graph; The second building module acquires phenomenon data that occurs during the production process of enterprise products, and establishes a logical relationship network between enterprise products and enterprise equipment based on the enterprise knowledge equipment map and the phenomenon data; A fusion module is used to connect the enterprise equipment knowledge graph and the logical relationship network by fusing the nodes at the physical layer with the nodes at the logical layer, establish a data chain of phenomenon->cause->solution, and complete the enterprise equipment knowledge graph; A detection module is used to detect anomalies in industrial production and obtain corresponding solutions based on the completed enterprise equipment knowledge graph.

6. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for constructing an industrial anomaly detection and solution knowledge graph as described in any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for constructing an industrial anomaly detection and solution knowledge graph as described in any one of claims 1 to 4 is implemented.

Images