Drainage equipment fault tracing method combined with knowledge graph

Abstract

A drainage equipment fault tracing method in combination with a knowledge graph relates to the technical field of drainage equipment fault intelligent analysis, and mainly comprises the following steps: step 1, carrying out data acquisition and preprocessing; step 2, performing knowledge graph construction; step 3, fault feature extraction is carried out; step 4, performing fault diagnosis and traceability; and step 5, result display and feedback are carried out. According to the method, rapid and accurate traceability of the drainage equipment fault is realized by combining a knowledge graph technology. A fault source can be quickly positioned, the maintenance efficiency is improved, the downtime of equipment is reduced, and the influence on the operation of a drainage system is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of drainage equipment fault diagnosis, and in particular relates to a drainage equipment fault tracing method combined with a knowledge graph. Background Art

[0002] With the acceleration of urbanization, the importance, scale and complexity of drainage systems are increasing. The normal operation of drainage equipment is crucial to the normal operation of cities and environmental protection. However, due to the long-term harsh working environment of drainage equipment, various faults are prone to occur. Traditional fault diagnosis methods often rely on manual experience and simple data analysis, which makes it difficult to accurately and quickly locate the root cause of the fault, resulting in low maintenance efficiency and increased costs. With the rapid development of artificial intelligence technology, knowledge graphs, which are constructed manually or by extracting knowledge from structured data sources, have higher accuracy and reliability, and have unique advantages and scope of application for fault tracing in the drainage field.

[0003] Therefore, the present invention combines knowledge graph technology to solve the shortcomings of the existing technology: most traditional drainage equipment fault diagnosis methods rely on experienced technicians to judge the status of the equipment by observing and analyzing the equipment operation data. This method is not only time-consuming and labor-intensive, but also due to the lack of systematic data support, the accuracy of the diagnosis results is often difficult to guarantee. In addition, traditional methods are difficult to deal with complex and changeable fault types, especially when facing chain failures, it is difficult to quickly locate the source of the fault. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a drainage equipment fault tracing method combined with a knowledge graph. By constructing a knowledge graph in the field of drainage equipment, intelligent analysis and rapid tracing of fault information can be achieved, thereby improving the fault detection rate and tracing efficiency.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A drainage equipment fault tracing method combined with knowledge graph, the steps are:

[0007] Step 1: Data collection and preprocessing;

[0008] Step 2: Build a knowledge graph;

[0009] Step 3: Extract fault features;

[0010] Step 4: Perform fault diagnosis and tracing;

[0011] Step 5: Display results and provide feedback.

[0012] Preferably, step 1, data collection and preprocessing:

[0013] Step 1.1, collecting the operation data of the drainage equipment, the operation data comes from the sensors and actuators of the drainage equipment;

[0014] Step 1.2: Clean, denoise and normalize the collected operation data to ensure the quality and availability of the operation data.

[0015] Preferably, the sensors include flow meters, pressure gauges and temperature sensors; the actuators include motors, water pumps and valves; the sensors are used to monitor the operating status and environmental parameters of the equipment in real time, and synchronize the historical maintenance record information of the equipment.

[0016] Preferably, step 2: constructing a knowledge graph:

[0017] Step 2.1, clarify the scope of drainage equipment failure, including various types of drainage equipment, possible failure types, failure symptoms and equipment components;

[0018] Step 2.2: Extract data from various types of actual drainage production as the basis for knowledge graph construction, and define entities related to drainage equipment failures, including equipment type, components, failure type, and failure symptoms;

[0019] Step 2.3, establish the relationship between entities, establish the "composed of" relationship between equipment and components; establish the "may cause" relationship between components and fault types, and establish the "manifested as" relationship between fault types and fault symptoms;

[0020] Step 2.4: Design the knowledge graph model, plan the overall structure and hierarchy of the knowledge graph, and determine the attributes of entities and relationships;

[0021] Step 2.5: Fill the collected data into the knowledge graph according to the defined entities, relationships and attributes.

[0022] Preferably, in step 2.3, the "manifested as" relationship includes the association between equipment components and fault types, and the corresponding relationship between fault symptoms and fault types;

[0023] In step 2.4, the attributes of the equipment entity include brand, model, and installation time.

[0024] The attributes of the fault type entity may include severity and repair difficulty.

[0025] Preferably, the filling requirements of step 2.5 include:

[0026] (1) The process involves the conversion, cleaning, and integration of data to ensure the accuracy and consistency of knowledge;

[0027] (2) Verify the populated knowledge graph to check the accuracy and completeness of the data;

[0028] (3) Continuously optimize and update the knowledge graph based on new failure cases and data.

[0029] Preferably, step 3, extracting fault features:

[0030] Using data analysis algorithms, feature vectors that can reflect equipment failures are extracted from preprocessed data; feature vectors include statistical features, spectral features, or model-based features; combined with the constructed knowledge graph, data is monitored in real time to effectively identify fault information of different severity levels generated by drainage equipment during operation.

[0031] Preferably, step 4, fault diagnosis and tracing:

[0032] Step 4.1, design a drainage equipment fault intelligent diagnosis model based on knowledge graph; input the extracted drainage equipment fault feature vector into the reasoning model based on knowledge graph;

[0033] Step 4.2: The diagnostic model performs fault diagnosis and source tracing based on the relationships and rules in the knowledge graph to determine the possible fault type and source.

[0034] Step 4.3: Trace the cause of the fault forward or backward according to the tree-like causal sequence distribution.

[0035] 9. The drainage equipment fault tracing method combined with knowledge graph according to claim 1 is characterized by:

[0036] Step 5: Display results and provide feedback:

[0037] The fault diagnosis and tracing results are intuitively displayed to users in the form of charts and reports through the information system; user feedback on the diagnosis results is collected to optimize the knowledge graph and diagnostic model; and the accuracy of fault diagnosis and tracing is improved.

[0038] Preferably, in step 4.3, the fault diagnosis process is:

[0039] (1) Input fault phenomenon set P = {p1, p2, ... p n}; n is a positive integer;

[0040] (2) Traverse the fault phenomenon set P to see whether P∈G is satisfied; where G represents the set of all entity elements in the graph;

[0041] If this condition is met, the set of known entities used for reasoning in the next step is P d =P;

[0042] If the above conditions are not met, the set P is operated, and the entity elements in the set P that are not in the knowledge graph G are removed to obtain the entity set: P d ={p i |p i ∈G and|p i ∈P};

[0043] (3) The entity set is used as a known entity set for search, and the known entity set is based on the relationship path: Introduction and entity set P d The directly associated fault cause entity collection:

[0044] The search path for the directly associated cause can be expressed as:

[0045] (4) From the known entity set P d According to the relationship path Introduction and entity set P d A collection of directly related fault cause entities According to the relationship path R monitor (C2, C1) introduces a set of associated device structure entities: S = range (R monitor (C2,C1));

[0046] (5) Verify the associated structure entity set S and the fault phenomenon set P d Does R exist? hasPhe (C1, C3) relationship path, that is, relationship P d =range(R hasPhe (C1,C3)) exists;

[0047] (6) If the relationship exists, the associated device structure entity set S is based on the relationship R hasPhe (C1, C3) The path introduces the indirect correlation fault phenomenon set P i =range(R hasPhe (C1,C3)) is represented by the indirect association fault display entity set P i According to the relationship path Introduction and fault phenomenon set P d Indirectly related fault cause set:

[0048]

[0049] If not present, then the indirectly related fault cause set The search path for indirect association causes can be expressed as:

[0050]

[0051] (7) Take the fault cause entity set F0 = F d UF i , that is, F0 is the set of directly associated fault causes F d The set of indirectly related fault causes F i The union of the fault cause entity set F0 according to the relationship path R advice (C4, C5) introduces the maintenance suggestion entity set A0, then: A0 = range (R advice (C4,C5));

[0052] (8) Sort the entities in the fault cause set F0. The sorting method is as follows: first, put the directly related causes before the indirectly related causes; second, sort the entities in the directly related causes according to their correlation with the entity set P. d The entities in the list are sorted from most to least in terms of the number of directly associated paths; after sorting, the fault cause entity set F1 is obtained, and the maintenance suggestion entities are sorted in the same order to obtain the entity set A1;

[0053] (9) After formatting the fault cause entity set F1 and the maintenance suggestion entity set A1, the diagnosis result can be obtained.

[0054] A drainage equipment fault tracing system combined with a knowledge graph adopts the drainage equipment fault tracing method combined with a knowledge graph.

[0055] The present invention can achieve the following beneficial effects:

[0056] 1. Improve the accuracy of fault diagnosis: Combining the rich knowledge and relationships in the knowledge graph can more comprehensively and accurately analyze the cause of the fault and reduce misdiagnosis and missed diagnosis.

[0057] 2. Shorten troubleshooting time: quickly locate the source of the fault, improve maintenance efficiency, reduce equipment downtime, and reduce the impact on the operation of the drainage system.

[0058] 3. Reduce maintenance costs: Through accurate fault tracing, unnecessary parts replacement and repair work can be avoided, saving maintenance costs. Knowledge accumulation and inheritance: The knowledge graph can continuously accumulate and update the experience and knowledge of drainage equipment fault diagnosis, which is convenient for new employees to learn and refer to, and improve the overall operation and maintenance level. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

[0060] Figure 1 This is a knowledge verification framework diagram of the present invention;

[0061] Figure 2 This is a schematic diagram of the principle of the diagnostic method of the present invention;

[0062] Figure 3 This is a fault diagnosis flow chart of the present invention;

[0063] Figure 4 A flowchart of the sentence vector generation algorithm of the present invention;

[0064] Figure 5 This is the information system architecture diagram of the present invention. DETAILED DESCRIPTION

[0065] The preferred solution is Figures 1 to 5 As shown in the figure, a drainage equipment fault tracing method combined with knowledge graph, the steps are:

[0066] 1. Data collection and preprocessing:

[0067] Collecting the operation data of drainage equipment is the first step in fault diagnosis, which mainly comes from various sensors and actuators of drainage equipment. These devices include motors, water pumps, valves, flow meters, pressure gauges, temperature sensors, etc., which monitor the operation status and environmental parameters of the equipment in real time, and synchronize the historical maintenance records of the equipment.

[0068] The collected data are cleaned, denoised and normalized to ensure the quality and usability of the data.

[0069] In addition to collecting information directly from devices, you can also collect information from authoritative, reliable and verified data sources, such as professional databases, authoritative research reports, audited industry standards, etc.

[0070] 2. Knowledge graph construction:

[0071] Clarify the scope of drainage equipment failure-related fields, including various types of drainage equipment (such as pumps, valves, pipes, processing equipment, monitoring equipment, etc.), possible types of failures (such as siltation, leakage, mechanical damage, substandard water quality, etc.), failure symptoms (such as abnormal noise, abnormal vibration, abnormal pressure, abnormal flow, abnormal water quality, etc.) and equipment components, etc. Extract data from various actual drainage production as the basis for knowledge graph construction, and define entities related to drainage equipment failures, such as equipment type, components, failure type, failure symptoms, etc.

[0072] Establish relationships between entities, establish a "composition" relationship between equipment and components, a "may cause" relationship between components and fault types, and a "manifests as" relationship between fault types and fault symptoms, such as the association between equipment components and fault types, and the correspondence between fault symptoms and fault types.

[0073] Knowledge graph model design, planning the overall structure and hierarchy of the knowledge graph, determining the attributes of entities and relationships, such as the attributes of equipment entities can include brand, model, installation time, etc., and the attributes of fault type entities can include severity, repair difficulty, etc. Fill the collected data into the knowledge graph according to the defined entities, relationships, and attributes. The process involves data conversion, cleaning, and integration to ensure the accuracy and consistency of knowledge. Verify the populated knowledge graph to check the accuracy and completeness of the data. At the same time, continuously optimize and update the knowledge graph based on new fault cases and data.

[0074] 3. Fault feature extraction:

[0075] Using data analysis algorithms, including principal component analysis (PCA) and independent component analysis (ICA), feature vectors that can reflect equipment failures are extracted from preprocessed data. These features can be statistical features (such as mean, variance, peak, etc.), spectral features, or model-based features. Combined with the constructed knowledge graph, real-time monitoring data can effectively identify fault information of different severity levels generated by drainage equipment during operation.

[0076] 4. Fault diagnosis and tracing:

[0077] Design an intelligent diagnosis model for drainage equipment faults based on knowledge graph. Input the extracted drainage equipment fault feature vectors into the reasoning model based on knowledge graph.

[0078] The diagnostic model performs fault diagnosis and tracing based on the relationships and rules in the knowledge graph to determine the possible fault types and sources.

[0079] A large number of chain failure events caused by a certain basic fault are called chain failures. In a chain failure, the failures are hierarchical and distributed in a tree-like causal sequence. They are mainly composed of three parts: the root fault, the subordinate faults caused by the root fault, and a driving path from the root fault to a subordinate fault. This path can be used to trace the cause of the fault forward or reversely.

[0080] 5. Results display and feedback:

[0081] The information system can intuitively display the fault diagnosis and tracing results to users in the form of charts, reports, etc. The user feedback on the diagnosis results is collected to further optimize the knowledge graph and diagnosis model. The accuracy of fault diagnosis and tracing is improved.

[0082] Embodiment 1:

[0083] 1. Data collection and preprocessing:

[0084] Relying on the existing drainage facility monitoring equipment on the project, the equipment status and monitoring data are analyzed in real time. The reporting frequency of equipment monitoring data is set to 30 minutes. It is necessary to consider the instability of equipment reporting and abnormal reporting data, and clean, denoise and normalize the collected data to ensure the quality and availability of the data.

[0085] 2. Knowledge graph construction:

[0086] The most widely used Neo4j graph database is constructed to express knowledge in the form of triples of the structure <entity, relationship, entity>. Entities are represented by edges in the graph database, and the graph structure corresponds to the graph structure characteristics of the knowledge graph. The node data attributes are shown in Table 1:

[0087]

[0088] Table 1 Node data attributes

[0089] In the process of constructing the knowledge graph, different types of concepts are distinguished and represented by node labels in Neo4j, and the remaining attributes are represented by node attributes; ontology relationships are represented by edges in Neo4j, and the attributes of relationships are represented by edge attributes. The relationship data attributes are shown in Table 2:

[0090]

[0091] Table 2 Relationship data attributes

[0092] The id attribute defined in the above table has a globally unique value in the Neo4j graph database and can be used as a unique index. According to the definition in the above table, taking the construction of the sewage lift pump fault diagnosis knowledge graph as an example, the data after the ontology representation is formatted into the corresponding csv file, and finally the Neo4j command line tool is used to store the data persistently in the database to complete the construction of the main pump fault diagnosis knowledge graph, that is, to complete the construction of the data model layer in the system development. The statistics of the number of nodes and relationships in the sewage lift pump fault diagnosis knowledge graph are shown in Table 3:

[0093]

[0094] Table 3 Statistics of the number of nodes and relationships in the knowledge graph for sewage lift pump fault diagnosis

[0095] 3. Fault feature extraction:

[0096] Using principal component analysis (PCA) and independent component analysis (ICA), feature vectors that can reflect equipment failures are extracted from preprocessed data. These features can be statistical features (such as mean, variance, peak, etc.), spectral features, or model-based features. Combined with the constructed knowledge graph, real-time monitoring data can effectively identify fault information of different severity levels generated by drainage equipment during operation.

[0097] 4. Fault diagnosis and tracing:

[0098] The purpose of fault analysis is to find the most likely cause of the fault, based on the monitoring alarm information. In particular, for a fault caused by multiple reasons, it is represented in the knowledge graph as multiple faults or fault antecedent entities pointing to the same fault entity. When querying with the triggering relationship, there will be multiple traceability fault chains. Considering that the knowledge graph decision can only select the traceability direction through the single attribute data of the entity or relationship, it is impossible to combine and utilize multiple attributes in the fault chain, and the decision result is not accurate. Therefore, the knowledge graph is integrated with the fuzzy multi-attribute decision-making method to evaluate and decide on the fault chain corresponding to the alarm fault, so as to obtain scientific and accurate analysis results.

[0099] The detailed implementation steps are as follows:

[0100] (1) Input fault phenomenon set P = {p1, p2, ... p n}, n is a positive integer.

[0101] (2) Traverse the fault phenomenon set P to see if it satisfies (where G represents the set of all entity elements in the graph). If this condition is met, the set of known entities used for reasoning in the next step is P d =P; if the above conditions are not met, operate on the set P and remove the set P that is not in the knowledge graph P d ={p i |p i ∈G and|p i ∈P}, after filtering the entity elements in G, we get the entity set.

[0102] (3) The entity set is used as a known entity set for search, and the known entity set is used according to the relationship path.

[0103] Introduction and entity set P d The directly associated fault cause entity collection:

[0104] The search path for the directly related cause can be expressed as

[0105] (4) From the known entity set P dAccording to the relationship Path introduction and entity set P d Directly related faults Cause entity set R monitor (C2, C1), according to the relationship path, the set of associated device structure entities S = range (R monitor (C2,C1)).

[0106] (5) Verify the associated structure entity set S and the fault phenomenon set P d Does R exist? hasPhe (C1,C3) relationship path P d =range(R hasPhe (C1,C3)) That is, whether the relationship exists;

[0107] (6) If the relationship exists, the associated device structure entity set S is calculated according to the relationship path R hasPhe (C1, C3) introduces the indirect correlation fault phenomenon set Pi = range (R hasPhe (C1,C3)) is represented by the indirect association fault display entity set P i According to the relationship path Introduction and fault phenomenon set P d Indirectly related fault cause set: If not present, then the indirectly related fault cause set The search path for indirect association causes can be expressed as:

[0108] (7) Take the fault cause entity set F0 = F d UF i , that is, F0 is the set of directly associated fault causes F d The set of indirectly related fault causes F i The union of the fault cause entity set F0 according to the relationship path R advice (C4, C5) introduces the maintenance suggestion entity set A0, then: A0 = range (R advice (C4,C5)).

[0109] (8) Sort the entities in the fault cause set F0. The sorting method is as follows: first, put the directly related causes before the indirectly related causes; second, sort the entities in the directly related causes according to their correlation with the entity set P. d The entities in the list are sorted from most to least in terms of the number of directly associated paths. After sorting, the fault cause entity set F1 is obtained, and the maintenance suggestion entities are sorted in the same order to obtain the entity set A1;

[0110] (9) After formatting the fault cause entity set F1 and the maintenance suggestion entity set A1, the diagnosis result can be obtained.

[0111] 5. Results display and feedback:

[0112] The information system can intuitively display the fault diagnosis and tracing results to users in the form of charts, reports, etc. The user feedback on the diagnosis results is collected to further optimize the knowledge graph and diagnosis model.

[0113] This embodiment takes the failure of a drainage pump in a sewage treatment plant in Yuanjiang City, Hunan Province as an example, and collects the operating data of the four pump groups in the secondary lift pump room of the sewage treatment plant, including flow, pressure, current, vibration, etc. After preprocessing and feature extraction, a feature vector reflecting the pump failure is obtained. These feature vectors are input into the diagnostic model based on the knowledge graph. According to the relationship between the component structure, common fault types and symptoms of the pump in the knowledge graph, the model diagnoses that the pump may have a drop in flow caused by impeller wear, and accurately traces it to the impeller component. The maintenance personnel of the sewage treatment plant carried out repairs based on the diagnosis results and successfully restored the normal operation of the pump. The drainage equipment fault tracing technology of the knowledge graph has been applied and verified in the project.

[0114] The above embodiments are only preferred technical solutions of the present invention and should not be regarded as limiting the present invention. The protection scope of the present invention shall be the technical solutions recorded in the claims, including equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present invention.

Claims

1. A drainage equipment fault tracing method combined with knowledge graph, characterized in that The following steps are involved: Step 1: Data collection and preprocessing; Step 2: Build a knowledge graph; Step 3: Extract fault features; Step 4: Perform fault diagnosis and tracing; Step 5: Display results and provide feedback.

2. According to claim 1, a drainage equipment fault tracing method combined with a knowledge graph is characterized by: Step 1: Data collection and preprocessing: Step 1.1, collecting the operation data of the drainage equipment, the operation data comes from the sensors and actuators of the drainage equipment; Step 1.2: Clean, denoise and normalize the collected operation data to ensure the quality and availability of the operation data.

3. According to claim 2, a drainage equipment fault tracing method combined with a knowledge graph is characterized by: The sensors include flow meters, pressure gauges and temperature sensors; the actuators include motors, water pumps and valves; the sensors are used to monitor the operating status and environmental parameters of the equipment in real time, and synchronize the historical maintenance record information of the equipment.

4. According to claim 1, a drainage equipment fault tracing method combined with a knowledge graph is characterized by: Step 2: Build a knowledge graph: Step 2.1, clarify the scope of drainage equipment failure, including various types of drainage equipment, possible failure types, failure symptoms and equipment components; Step 2.2: Extract data from various types of actual drainage production as the basis for knowledge graph construction, and define entities related to drainage equipment failures, including equipment type, components, failure type, and failure symptoms; Step 2.3, establish the relationship between entities, establish the "composed of" relationship between equipment and components; establish the "may cause" relationship between components and fault types, and establish the "manifested as" relationship between fault types and fault symptoms; Step 2.4: Design the knowledge graph model, plan the overall structure and hierarchy of the knowledge graph, and determine the attributes of entities and relationships; Step 2.5: Fill the collected data into the knowledge graph according to the defined entities, relationships and attributes.

5. According to claim 1, a drainage equipment fault tracing method combined with a knowledge graph is characterized by: In step 2.3, the "manifested as" relationship includes the association between equipment components and fault types, and the correspondence between fault symptoms and fault types; In step 2.4, the attributes of the equipment entity include brand, model, and installation time. The attributes of the fault type entity may include severity and repair difficulty.

6. According to claim 4, a drainage equipment fault tracing method combined with knowledge graph is characterized by: The filling requirements for step 2.5 include: (1) The process involves the conversion, cleaning, and integration of data to ensure the accuracy and consistency of knowledge; (2) Verify the populated knowledge graph to check the accuracy and completeness of the data; (3) Continuously optimize and update the knowledge graph based on new failure cases and data.

7. According to claim 1, a drainage equipment fault tracing method combined with a knowledge graph is characterized by: Step 3: Extract fault features: Using data analysis algorithms, extracting feature vectors that can reflect equipment failures from preprocessed data; feature vectors include statistical features, spectral features, or model-based features; Combined with the constructed knowledge graph, real-time monitoring data can be used to effectively identify fault information of different severity levels generated by drainage equipment during operation.

8. The drainage equipment fault tracing method combined with knowledge graph according to claim 1 is characterized by: Step 4: Fault diagnosis and tracing: Step 4.1, design a drainage equipment fault intelligent diagnosis model based on knowledge graph; input the extracted drainage equipment fault feature vector into the reasoning model based on knowledge graph; Step 4.2: The diagnostic model performs fault diagnosis and source tracing based on the relationships and rules in the knowledge graph to determine the possible fault type and source. Step 4.3: Trace the cause of the fault forward or backward according to the tree-like causal sequence distribution.

9. The drainage equipment fault tracing method combined with knowledge graph according to claim 1 is characterized by: Step 5: Display results and provide feedback: The fault diagnosis and tracing results are intuitively displayed to users in the form of charts and reports through the information system; user feedback on the diagnosis results is collected to optimize the knowledge graph and diagnostic model; and the accuracy of fault diagnosis and tracing is improved.

10. The drainage equipment fault tracing method combined with knowledge graph according to claim 8 is characterized by: In step 4.3, the fault diagnosis process is as follows: (1) Input fault phenomenon set P = {p1, p2, ... p n }; n is a positive integer; (2) Traverse the fault phenomenon set P to see whether it satisfies P∈G; Where G represents the set of all entity elements in the graph; If this condition is met, the set of known entities used for reasoning in the next step is P d =P; If the above conditions are not met, the set P is operated, and the entity elements in the set P that are not in the knowledge graph G are removed to obtain the entity set: P d ={p i p i ∈G and p i ∈P}; (3) The entity set is used as a known entity set for search, and the known entity set is based on the relationship path: Introduction and entity set P d The directly associated fault cause entity collection: The search path for the directly associated cause can be expressed as: (4) From the known entity set P d According to the relationship path Introduction and entity set P d A collection of directly related fault cause entities According to the relationship path R monitor (C2, C1) introduces a set of associated device structure entities: S = range (R monitor (C2,C1)); (5) Verify the associated structure entity set S and the fault phenomenon set P d Is Rhas present? P he(C1,C3) relation path, that is, relation P d =range(R hasPhe (C1,C3)) exists; (6) If the relationship exists, the associated device structure entity set S is based on the relationship R hasPhe (C1,C3) system path Introducing the indirect correlation fault phenomenon set P i =range(R hasPhe (C1,C3)) is represented by the indirect association fault display entity set P i According to the relationship path Introduction and fault phenomenon set P d Indirectly related fault cause set: If not present, then the indirectly related fault cause set The search path for indirect association causes can be expressed as: (7) Take the fault cause entity set F0 = F d UF i , that is, F0 is the set of directly associated fault causes F d The set of indirectly related fault causes F i The union of the fault cause entity set F0 according to the relationship path R advice (C4, C5) introduces the maintenance suggestion entity set A0, then: A0 = range (R advice (C4,C5)); (8) Sort the entities in the fault cause set F0. The sorting method is as follows: first, put the directly related causes before the indirectly related causes; second, sort the entities in the directly related causes according to their correlation with the entity set P. d The entities in the list are sorted from most to least in terms of the number of directly associated paths; after sorting, the fault cause entity set F1 is obtained, and the maintenance suggestion entities are sorted in the same order to obtain the entity set A1; (9) After formatting the fault cause entity set F1 and the maintenance suggestion entity set A1, the diagnosis result can be obtained.

11. A drainage equipment fault tracing system combined with knowledge graph, characterized by: A drainage equipment fault tracing method combined with a knowledge graph according to any one of claims 1-10 is adopted.

Images