CN117350377A - Knowledge graph driving-based equipment fault diagnosis method and device - Google Patents
Knowledge graph driving-based equipment fault diagnosis method and device Download PDFInfo
- Publication number
- CN117350377A CN117350377A CN202311228706.9A CN202311228706A CN117350377A CN 117350377 A CN117350377 A CN 117350377A CN 202311228706 A CN202311228706 A CN 202311228706A CN 117350377 A CN117350377 A CN 117350377A
- Authority
- CN
- China
- Prior art keywords
- fault
- equipment
- knowledge
- data
- graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000003745 diagnosis Methods 0.000 title claims abstract description 73
- 238000012423 maintenance Methods 0.000 claims abstract description 57
- 239000013598 vector Substances 0.000 claims abstract description 36
- 238000005516 engineering process Methods 0.000 claims abstract description 17
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000012544 monitoring process Methods 0.000 claims description 35
- 238000004458 analytical method Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 12
- 238000005259 measurement Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000005315 distribution function Methods 0.000 claims description 6
- 238000013178 mathematical model Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 239000000463 material Substances 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 4
- 230000005236 sound signal Effects 0.000 claims description 4
- 238000001228 spectrum Methods 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 238000012800 visualization Methods 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 230000008439 repair process Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 8
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 4
- 229910000831 Steel Inorganic materials 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention provides a method and a device for diagnosing equipment faults based on knowledge graph driving, which utilize the relation among equipment structures through the knowledge graph technology, obtain the entity and the relation related to the equipment faults through feature matching of fault data, deduce possible fault reasons through the relation among the equipment structures, and can carry out fault diagnosis and reasoning more simply and accurately. Comprising the following steps: acquiring and processing operation data of the equipment, and representing related ontology as entities and relations in the atlas by using an ontology modeling technology to obtain a knowledge atlas for equipment fault diagnosis; converting fault data of the sensor into feature vectors by utilizing a feature extraction technology, and obtaining entities and relations related to equipment faults through feature matching; carrying out fault cause reasoning through the relation among the equipment structures to obtain fault types and positions; and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
Description
Technical Field
The invention relates to the technical field of optical communication, in particular to a method and a device for diagnosing equipment faults based on knowledge graph driving.
Background
In the industrial production of steel flows, production lines are stopped each year due to various equipment faults, so that production efficiency is low, and huge economic loss is caused, and therefore an effective equipment state operation monitoring means is found, and an important role is played in safety and stability of production operation and cost reduction. The traditional equipment fault diagnosis method generally depends on expert experience and rule base, is difficult to cover all possible fault conditions, an equipment monitoring data pool is gradually huge, operation and maintenance personnel with abundant experience are required to conduct manual analysis for accurately judging fault characteristics, the requirement on quality of field maintenance personnel is high, and the efficiency is extremely low. The introduction of the knowledge graph brings priori knowledge and new research means for tracing and rapidly removing faults of equipment, the knowledge graph carries out deeper analysis on the faults of the equipment through automatic reasoning, uncertainty reasoning and other methods, and the rules and modes in the knowledge graph are found, so that the faults of the equipment are predicted and diagnosed better, and intelligent upgrading of equipment maintenance is promoted.
In the fault diagnosis method of the prior art, a plurality of fault diagnosis methods adopting a knowledge graph technology are adopted, different technical modes are adopted for different emphasis points in each method, the fault diagnosis method based on the knowledge graph is provided by Chinese patent with publication number of CN 114491037A, faults are judged by a method for determining the similarity between the characteristic vector of the current fault equipment and the characteristic vector of each historical fault equipment in the equipment fault knowledge graph, and the data calculation amount is large. The invention provides a novel knowledge-graph-driving-based equipment fault diagnosis method, and provides a novel thought which can carry out fault diagnosis and reasoning more simply and accurately.
Disclosure of Invention
In order to solve the technical problems in the background technology, the invention provides a device fault diagnosis method and device based on a knowledge graph drive, which utilizes the relationship among device structures through the knowledge graph technology, obtains the entity and relationship related to the device fault through feature matching of fault data, deduces possible fault reasons through the relationship among the device structures, and can carry out fault diagnosis and reasoning more simply and accurately.
In order to achieve the above purpose, the invention is realized by adopting the following technical scheme:
a device fault diagnosis method based on knowledge graph driving comprises the following steps:
step S1, acquiring and processing operation data of equipment: converting the original data into a data form suitable for knowledge graph construction and analysis;
step S2, constructing a knowledge graph aiming at equipment faults: representing the related ontology as an entity and a relation in the map by using an ontology modeling technology to obtain a knowledge map for equipment fault diagnosis;
step S3, converting the fault data of the sensor into feature vectors by utilizing a feature extraction technology, obtaining entities and relations related to equipment faults through feature matching, and matching the feature data with the entities and relations in the knowledge graph, wherein the steps comprise the following steps:
1) Converting fault data of a device into a feature vector f= (f) 1 ,f 2 ,…f n ) The knowledge graph is expressed as G (V, E), and the relationship between each entity in the graph is E i,j The method comprises the steps of carrying out a first treatment on the surface of the Wherein f 1 ,f 2 ,…f n Representing n fault data; v represents a collection of entities and concepts; e represents a relation set between the entity and the concept, i and j represent entity numbers;
2) Measuring the similarity between the feature vector and the entity relationship by adopting a similarity function, and finding the entity v with the largest correlation in the knowledge graph i The calculation method is as follows:
wherein similarity (f, v) i ) For the feature vector f and the entity v i Similarity of v i Representing entities, v i, k represents entity v i Is the kth eigenvalue, f k Is the kth fault data;
3) By entity v i Obtaining a corresponding equipment component entity and a fault type entity;
4) The characteristic vector and the characteristic relation e in the knowledge graph are combined i,j Matching is carried out, and the relation between the corresponding equipment component and the fault type is found;
s4, carrying out fault cause reasoning through the relation among the equipment structures to obtain a fault type and a fault position;
step S5, fault diagnosis and maintenance: and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
Further, the step S4 specifically includes:
1) Before observing the device data, it is assumed that the device has N possible fault states s 1 ,s 2 ,...s N When each fault state has a probability P (s 1 ),P(s 2 ),...P(s N ) Wherein
2) From M measured parameters, from x, according to the device status 1 ,x 2 ,...x M Description of fault state s i The device state uses an M-dimensional vector X i ＝[x i1 ,x i2 ,...x iM ]Describing, namely, a state equation of the equipment;
3) Obtaining probability distribution function f of each parameter under each fault state according to state equation of the equipment i,k (x k ) Xk is the kth measurement parameter;
4) Obtaining an actual value of the monitoring data according to the monitoring parameters and the monitoring data of the measuring points;
5) According to the Bayesian theorem, under the condition of giving the actual measured value of the equipment, the posterior probability when the equipment fails is obtained as follows:
wherein,is shown in the fault state S i Next, the j-th monitoring parameter->By combining f i,k (x k ) Is carried into a monitoring parameter model to be calculated, X i Is a state vector +.>Representing actual measurements of the monitored data; p (S) i ) Representing a fault state S i Probability of (2);
determining the fault type and position of the equipment according to the magnitude of the posterior probability, and determining the maximum fault state S i I.e. the type of fault of the device, whereas the location of the fault of the machine is determined by analysing the state vector X i Abnormal values of the respective parameters.
Further, in the reasoning process, the influence of the measurement error is considered, and the monitoring data is preprocessed by a Kalman filter method.
Further, a mathematical model of the device is constructed, wherein the mathematical model comprises information including the structure, the dynamics characteristics and the operation parameters of the machine; establishing a state equation of the machine through the information, and describing response and vibration characteristics of the machine under different fault states; and monitoring and analyzing the vibration, sound and temperature parameters of the machine to obtain the required monitoring data.
Further, the step S2 specifically includes the following steps:
1) Using ontology modeling technology, establishing an ontology concept set of the device, wherein the ontology concept set comprises the following steps: the method comprises the steps of a concept set of a device structure, a concept set of each measuring point for monitoring the running state of the device, a concept set of various phenomena expressed when the device breaks down, a concept set of a device fault cause and a concept set of maintenance suggestions;
2) Establishing an ontology relation of the equipment, and representing the relation among the ontology concepts of the equipment, wherein the method comprises the following steps: the equipment parts have the fault phenomenon, the sensors monitor the running state of the equipment, other phenomena are caused by the fault, the reason for the equipment to have the fault is found, and measures are taken for maintenance;
3) According to the established equipment fault diagnosis ontology representation, information extraction is carried out on IT side data through data configuration, the label of an entity is defined, an equipment triplet relation table is established, the relation is identified, and the equipment knowledge is represented as a triplet structure: the entity-relation-entity adopts a graph database to store data of the triplet knowledge, nodes in the graph database represent the entities, the relation among the knowledge is represented by the entity relation in the triplet, the relation among the knowledge is represented by the side with the direction, and the relation among the triplet is corresponding to the constructed equipment knowledge graph part relation and node visualization graph.
Further, the step S1 includes: the problems of data missing, abnormal value and repeated value existing in fault diagnosis are timely processed through data cleaning, feature extraction and data preprocessing, the frequency spectrum features, time domain features and wavelet transformation features of equipment are extracted from data such as vibration signals, sound signals and temperature signals, and feature data are normalized and standardized.
Further, the maintenance advice of step S5 specifically includes:
1) Related knowledge and rules in the knowledge graph are continuously updated by combining the actual condition of the machine operation and maintenance records, so that the knowledge graph can adapt to the actual operation condition of the machine;
2) According to the real-time running state of the equipment, a fault diagnosis result driven by a knowledge graph is obtained, a maintenance mode of the machine is determined, and related maintenance tools and maintenance materials are prepared for maintenance.
Further, the invention also provides a diagnosis system of the equipment fault diagnosis method based on the knowledge graph driving, which comprises the following steps:
and the data acquisition and processing module is used for: collecting sensor data, and preprocessing the collected data to obtain a data form suitable for knowledge graph construction and analysis;
knowledge graph construction module: representing the related ontology as an entity and a relation in the map by using an ontology modeling technology to obtain a knowledge map for equipment fault diagnosis;
and the fault characteristic data and entity and relation matching module in the knowledge graph: converting fault data of the sensor into feature vectors by utilizing a feature extraction technology, and obtaining entities and relations related to equipment faults through feature matching;
fault type and location analysis module: carrying out fault cause reasoning through the relation among the equipment structures to obtain fault types and positions;
fault diagnosis and repair advice module: and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
The invention also provides a device for realizing the equipment fault diagnosis method based on the knowledge graph driving, which comprises a processor and a memory;
wherein the processor is configured to execute the knowledge-graph-driven equipment fault diagnosis method;
the memory is for storing executable instructions of the processor.
The invention also provides a computer storage medium, on which a computer program is stored, the computer program being executed by a processor to implement the knowledge-graph-driven equipment fault diagnosis method.
Compared with the prior art, the invention has the beneficial effects that:
according to the equipment fault diagnosis method and device based on the knowledge graph driving, the relationships among the equipment structures are utilized through the knowledge graph technology, the entities and the relationships related to equipment faults are obtained through feature matching of fault data, and possible fault reasons are deduced through the relationships among the equipment structures, so that fault diagnosis and reasoning can be carried out more simply and accurately.
Drawings
FIG. 1 is a flow chart of a method for diagnosing equipment faults based on knowledge graph driving;
FIG. 2 is a schematic diagram of a knowledge graph used in the present invention;
FIG. 3 is a flow chart for constructing a knowledge graph according to an embodiment of the present invention;
FIG. 4 is a device ontology model according to an embodiment of the present invention;
fig. 5 is a knowledge graph example of a motor according to an embodiment of the present invention.
Detailed Description
The following is a further description of embodiments of the invention, taken in conjunction with the accompanying drawings:
fig. 1 is a schematic flow chart of a method for diagnosing a device fault based on knowledge graph driving according to an embodiment of the present invention, and fig. 2 is a schematic flow chart of a knowledge graph used in the present invention. Fig. 3 is a construction flow of a knowledge graph designed by the present invention, and a fault diagnosis method of the present application is described with reference to fig. 1, fig. 2, and fig. 3. The equipment fault diagnosis method based on knowledge graph driving comprises the following steps:
step S1, acquiring and processing operation data of equipment: and converting the original data into a data form suitable for knowledge graph construction and analysis.
Step S2, constructing a knowledge graph aiming at equipment faults: and expressing the related ontology as entities and relations in the atlas by using an ontology modeling technology to obtain a knowledge atlas for equipment fault diagnosis.
Step S3, converting the fault data of the sensor into feature vectors by utilizing a feature extraction technology, obtaining entities and relations related to equipment faults through feature matching, and matching the feature data with the entities and relations in the knowledge graph, wherein the steps comprise the following steps:
1) Converting fault data of a device into feature directionsQuantity f= (f 1 ,f 2 ,…f n ) The knowledge graph is expressed as G (V, E), and the relationship between each entity in the graph is E i,j The method comprises the steps of carrying out a first treatment on the surface of the Wherein f 1 ,f 2 ,…f n Representing n fault data; v represents a collection of entities and concepts; e represents a set of relationships between entities and concepts, i, j represents an entity number.
2) Measuring the similarity between the feature vector and the entity relationship by adopting a similarity function, and finding the entity v with the largest correlation in the knowledge graph i The calculation method is as follows:
wherein similarity (f, v) i ) For the feature vector f and the entity v i Similarity of v i Representing entities, v i, k represents entity v i Is the kth eigenvalue, f k Is the kth fault data.
3) By entity v i And obtaining the corresponding equipment component entity and fault type entity.
4) The characteristic vector and the characteristic relation e in the knowledge graph are combined i,j Matching is carried out, and the relation between the corresponding equipment component and the fault type is found.
And S4, carrying out fault cause reasoning through the relation among the equipment structures to obtain the fault type and the position.
Step S5, fault diagnosis and maintenance: and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
Further, the step S4 specifically includes:
1) Before observing the device data, it is assumed that the device has N possible fault states s 1 ,s 2 ,...s N When each fault state has a probability P (s 1 ),P(s 2 ),...P(s N ) Wherein
2) From M measured parameters, from x, according to the device status 1 ,x 2 ,...x M Description of fault state s i The device state uses an M-dimensional vector X i ＝[x i1 ,x i2 ,...x iM ]Describing, namely, a state equation of the equipment;
3) Obtaining probability distribution function f of each parameter under each fault state according to state equation of the equipment i,k (x k ) Xk is the kth measurement parameter;
4) Obtaining an actual value of the monitoring data according to the monitoring parameters and the monitoring data of the measuring points;
5) According to the Bayesian theorem, under the condition of giving the actual measured value of the equipment, the posterior probability when the equipment fails is obtained as follows:
wherein,is shown in the fault state S i Next, the j-th monitoring parameter->By combining f i,k (x k ) Is carried into a monitoring parameter model to be calculated, X i Is a state vector +.>Representing actual measurements of the monitored data; p (S) i ) Representing a fault state S i Probability of (2);
determining the fault type and position of the equipment according to the magnitude of the posterior probability, and determining the maximum fault state S i I.e. the type of fault of the device, whereas the location of the fault of the machine is determined by analysing the state vector X i Abnormal values of the respective parameters.
Further, in the reasoning process, the influence of the measurement error is considered, and the monitoring data is preprocessed by a Kalman filter method.
Further, a mathematical model of the device is constructed, wherein the mathematical model comprises information including the structure, the dynamics characteristics and the operation parameters of the machine; establishing a state equation of the machine through the information, and describing response and vibration characteristics of the machine under different fault states; and monitoring and analyzing the vibration, sound and temperature parameters of the machine to obtain the required monitoring data.
Further, the step S2 specifically includes the following steps:
1) Using ontology modeling technology, establishing an ontology concept set of the device, wherein the ontology concept set comprises the following steps: the method comprises the steps of a concept set of a device structure, a concept set of each measuring point for monitoring the running state of the device, a concept set of various phenomena expressed when the device breaks down, a concept set of a device fault cause and a concept set of maintenance suggestions;
2) Establishing an ontology relation of the equipment, and representing the relation among the ontology concepts of the equipment, wherein the method comprises the following steps: the equipment parts have the fault phenomenon, the sensors monitor the running state of the equipment, other phenomena are caused by the fault, the reason for the equipment to have the fault is found, and measures are taken for maintenance;
3) According to the established equipment fault diagnosis ontology representation, information extraction is carried out on IT side data through data configuration, the label of an entity is defined, an equipment triplet relation table is established, the relation is identified, and the equipment knowledge is represented as a triplet structure: the entity-relation-entity adopts a graph database to store data of the triplet knowledge, nodes in the graph database represent the entities, the relation among the knowledge is represented by the entity relation in the triplet, the relation among the knowledge is represented by the side with the direction, and the relation among the triplet is corresponding to the constructed equipment knowledge graph part relation and node visualization graph.
Further, the step S1 includes: the problems of data missing, abnormal value and repeated value existing in fault diagnosis are timely processed through data cleaning, feature extraction and data preprocessing, the frequency spectrum features, time domain features and wavelet transformation features of equipment are extracted from data such as vibration signals, sound signals and temperature signals, and feature data are normalized and standardized.
Further, the maintenance advice of step S5 specifically includes:
1) Related knowledge and rules in the knowledge graph are continuously updated by combining the actual condition of the machine operation and maintenance records, so that the knowledge graph can adapt to the actual operation condition of the machine;
2) According to the real-time running state of the equipment, a fault diagnosis result driven by a knowledge graph is obtained, a maintenance mode of the machine is determined, and related maintenance tools and maintenance materials are prepared for maintenance.
The invention also provides a diagnosis system of the equipment fault diagnosis method based on the knowledge graph driving, which comprises the following steps:
and the data acquisition and processing module is used for: collecting sensor data, and preprocessing the collected data to obtain a data form suitable for knowledge graph construction and analysis;
knowledge graph construction module: representing the related ontology as an entity and a relation in the map by using an ontology modeling technology to obtain a knowledge map for equipment fault diagnosis;
and the fault characteristic data and entity and relation matching module in the knowledge graph: converting fault data of the sensor into feature vectors by utilizing a feature extraction technology, and obtaining entities and relations related to equipment faults through feature matching;
fault type and location analysis module: carrying out fault cause reasoning through the relation among the equipment structures to obtain fault types and positions;
fault diagnosis and repair advice module: and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
Embodiment one:
taking a steel mill motor as an example, constructing a motor fault diagnosis knowledge graph, describing test case design through motor winding short-circuit faults, and verifying the feasibility of the method.
Step S1, acquiring and processing operation data of equipment;
converting the original data into a data form suitable for knowledge graph construction and analysis;
the method comprises the steps of preprocessing the problems of data missing, abnormal value, repeated value and the like in motor fault diagnosis, extracting corresponding frequency spectrum characteristics, time domain characteristics and wavelet transformation characteristics from data such as vibration signals, sound signals and temperature signals, and carrying out normalization, standardization and the like on characteristic data.
S2, constructing a knowledge graph aiming at motor faults;
utilizing an ontology modeling technology, establishing an ontology concept set of the motor, wherein the ontology concept set comprises the following steps: the method comprises the steps of a concept set of a motor structure, a concept set of each measuring point for monitoring the running state of the motor, a concept set of various phenomena expressed when the motor fails, a concept set of motor failure reasons and a concept set of maintenance suggestions;
establishing the body relation of the motor, and representing the relation among the body concepts of the motor, wherein the method comprises the following steps: the motor has a fault reason, the sensor monitors the running state of the motor, other phenomena are caused by the fault, the motor has a fault phenomenon, and measures are taken for maintenance;
according to the established motor fault diagnosis knowledge body representation, information extraction is carried out on IT side data through data configuration, a label of an entity is defined, a motor triplet relation table is established, the relation is identified, motor knowledge is represented as a triplet structure, a graphic database Neo4j is adopted to carry out data storage on the triplet knowledge, nodes in the graphic database represent the entity, entity relations in the triplet correspond, sides with directions represent the relation among the knowledge, the relation in the triplet corresponds, and a constructed motor knowledge map partial relation and node visualization diagram is shown in fig. 4;
the motor faults mainly occur on the stator, the rotor and the bearing parts, the sensor monitors the states of the parts at each monitoring point, the motor heating, noise abnormality and other fault phenomena are monitored through the measuring point, and the fault phenomena of the motor are caused by abrasion, insulation and other fault reasons, as shown in fig. 5.
Step S3, matching the characteristic data with the entity and the relation in the knowledge graph;
converting fault data of the motor into a feature vector f= (f) 1 ,f 2 ,…f n ) The knowledge-graph can be expressed as G (V, E)
Wherein V represents a collection of entities and concepts; e represents a set of relationships between entities and concepts, v i Representing entities e i,j The relation, i, j represents the entity number. The fault data of the motor comprise data such as current, vibration and temperature of the motor.
Measuring the similarity between the feature vector and the entity relationship by adopting a similarity function, and finding the entity v with the largest correlation in the knowledge graph i The calculation method is as follows:
wherein v is i,k Representing entity v i Is the kth eigenvalue of (c);
by entity v i Obtaining a motor component entity and a fault type entity corresponding to the motor component entity;
the characteristic vector and the characteristic relation e in the knowledge graph are combined i,j Matching is carried out, and the relation between the motor component and the fault type corresponding to the matching is found.
S4, carrying out fault cause reasoning through the relation among motor structures to obtain fault types and positions;
the motor winding short-circuit fault can be judged to occur only when the symptoms are simultaneously occurred;
by the characteristics of response, vibration and the like of the motor in different fault states, the fault cause reasoning monitors and analyzes parameters of vibration, sound, temperature and current of the motor by adopting an inference method, and deduces the existing fault type and fault position according to specific discrimination rules, and the method specifically comprises the following steps:
before motor data is observed, it is assumed that the motor has N possible fault states s 1 ,s 2 ,...s N When each fault state has a probability P (s 1 ),P(s 2 ),...P(s N ) Wherein
From M measured parameters, from x, in dependence on the state of the motor 1 ,x 2 ,...x M Description of fault state s i The motor state can use an M-dimensional vector X i ＝[x i1 ,x i2 ,...x iM ]To describe it;
obtaining probability distribution function f of each parameter under each fault state according to the state equation of the motor ik (x k )；
According to the monitoring parameters y of six measuring points 1 ,y 2 ,...y 6 And monitoring data to obtain actual values of 48 measurement parameters, wherein the calculation formula is as follows:
wherein,representing measurement error, y j Representing the value of the j-th parameter.
According to the Bayesian theorem, under the condition of giving the actual measured value of the motor, the posterior probability when the motor fails is obtained as follows:
wherein,is shown in the fault state S i Next, the probability distribution function of the jth monitored parameter may be determined by combining f i,k (x k ) Is carried into the monitoring parameter model to calculate +.>Representing the actual measured value;
determining the fault type and position of the motor according to the magnitude of the posterior probability, and determining the maximum fault state S i I.e. the type of failure of the motor, while the location of the failure of the machine is determined by analysing the state vector X i Abnormal values of the respective parameters in the database;
in the reasoning process, the influence of measurement errors is considered, and a Kalman filter method is often used for preprocessing the monitoring data.
S5, fault diagnosis and maintenance;
and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
Wherein, the maintenance suggestion specifically includes:
related knowledge and rules in the knowledge graph are continuously updated by combining the actual condition of the machine operation and maintenance records, so that the knowledge graph can adapt to the actual operation condition of the machine;
according to the real-time running state of the equipment, the fault diagnosis result driven by the knowledge graph is obtained, the maintenance mode of the machine is determined, and related maintenance tools, maintenance materials and the like are prepared for maintenance.
The invention also provides a device for realizing the equipment fault diagnosis method based on the knowledge graph driving, which comprises a processor and a memory; wherein the processor is configured to execute the knowledge-graph-driven equipment fault diagnosis method; the memory is for storing executable instructions of the processor.
The invention also provides a computer storage medium, on which a computer program is stored, the computer program being executed by a processor to implement the knowledge-graph-driven equipment fault diagnosis method.
In summary, the invention discloses a method and a device for diagnosing equipment faults based on knowledge graph driving, which are characterized in that related knowledge graph bodies are expressed as entities and relations in graphs through ontology modeling technology to obtain knowledge graphs of equipment fault diagnosis, data are processed to obtain data suitable for knowledge graph construction and analysis, sensor data are converted into feature vectors, entities and relations related to equipment faults are obtained through feature matching, fault cause reasoning is carried out on the relation between equipment structures to obtain fault types and positions, and according to related knowledge and rules in the knowledge graphs, diagnosis results and maintenance methods are analyzed to obtain maintenance suggestions of equipment faults.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The solutions in the embodiments of the present application may be implemented in various computer languages, for example, object-oriented programming language Java, and an transliterated scripting language JavaScript, etc.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the spirit or scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (10)
1. The equipment fault diagnosis method based on the knowledge graph driving is characterized by comprising the following steps of:
step S1, acquiring and processing operation data of equipment: converting the original data into a data form suitable for knowledge graph construction and analysis;
step S2, constructing a knowledge graph aiming at equipment faults: representing the related ontology as an entity and a relation in the map by using an ontology modeling technology to obtain a knowledge map for equipment fault diagnosis;
s3, converting the equipment fault data detected by the sensor into feature vectors by utilizing a feature extraction technology, and obtaining entities and relations related to the equipment faults through feature matching; matching the feature vector with the entities and relationships in the knowledge-graph includes:
1) Converting fault data of a device into a feature vector f= (f) 1 ,f 2 ,…f n ) The knowledge graph is expressed as G (V, E), and the relationship between each entity in the graph is E i,j The method comprises the steps of carrying out a first treatment on the surface of the Wherein f 1 ,f 2 ,…f n Representing n fault data; v represents a collection of entities and ontology concepts; e represents a relation set between the entity and the ontology concept, i and j represent entity numbers;
2) Measuring the similarity between the feature vector and the entity relationship by adopting a similarity function, and finding the entity v with the largest correlation in the knowledge graph i The calculation method is as follows:
wherein similarity (f, v) i ) For the feature vector f and the entity v i Similarity of v i Representing entities, v i,k Representing entity v i Is the kth eigenvalue, f k Is the kth fault data;
3) By entity v i Obtaining a corresponding equipment component entity and a fault type entity;
4) The characteristic vector and the characteristic relation e in the knowledge graph are combined i,j Matching is carried out, and the relation between the corresponding equipment component and the fault type is found;
s4, carrying out fault cause reasoning through the relation among the equipment structures to obtain a fault type and a fault position;
step S5, fault diagnosis and maintenance: and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
2. The method for diagnosing a fault in a device based on knowledge-graph driving of claim 1, wherein the step S4 specifically includes:
1) Before observing the device data, it is assumed that the device has N possible fault states s 1 ,s 2 ,...s N When each fault state has a probability P (s 1 ),P(s 2 ),...P(s N ) Wherein
2) From M measured parameters, from x, according to the device status 1 ,x 2 ,...x M Description of fault state s i The device state uses an M-dimensional vector X i ＝[x i1 ,x i2 ,...x iM ]Describing, namely, a state equation of the equipment;
3) Obtaining probability distribution function f of each parameter under each fault state according to state equation of the equipment i,k (x k )，x k For the kth measured parameter, f i,k (x k ) A probability distribution function for the kth parameter in the ith fault state;
4) Obtaining an actual value of the monitoring data according to the monitoring parameters and the monitoring data of the measuring points;
5) According to the Bayesian theorem, under the condition of giving the actual measured value of the equipment, the posterior probability when the equipment fails is obtained as follows:
wherein,is shown in the fault state S i Next, the j-th monitoring parameter->By combining f i,k (x k ) Carrying out calculation in a monitoring parameter model; x is X i Is a state vector +.>Representing actual measurements of the monitored data; p (S) i ) Representing a fault state S i Probability of (2);
determining the fault type and position of the equipment according to the magnitude of the posterior probability, and determining the maximum fault state S i I.e. the type of fault of the device, whereas the location of the fault of the machine is determined by analysing the state vector X i Abnormal values of the respective parameters.
3. The knowledge-graph-driven equipment fault diagnosis method according to claim 2, wherein the monitoring data is preprocessed by a kalman filter method in consideration of the influence of measurement errors in the reasoning process.
4. The knowledge-graph-driven equipment fault diagnosis method according to claim 2, wherein a mathematical model of the equipment is constructed, wherein the mathematical model comprises information including the structure, dynamics and operation parameters of the machinery; establishing a state equation of the machine through the information, and describing response and vibration characteristics of the machine under different fault states; and monitoring and analyzing the vibration, sound and temperature parameters of the machine to obtain the required monitoring data.
5. The method for diagnosing a fault in a device based on knowledge-graph driving of claim 1, wherein the step S2 specifically comprises the following steps:
1) Using ontology modeling technology, establishing an ontology concept set of the device, wherein the ontology concept set comprises the following steps: the method comprises the steps of a concept set of a device structure, a concept set of each measuring point for monitoring the running state of the device, a concept set of various phenomena expressed when the device breaks down, a concept set of a device fault cause and a concept set of maintenance suggestions;
2) Establishing an ontology relation of the equipment, and representing the relation among the ontology concepts of the equipment, wherein the method comprises the following steps: the equipment parts have the fault phenomenon, the sensors monitor the running state of the equipment, other phenomena are caused by the fault, the reason for the equipment to have the fault is found, and measures are taken for maintenance;
3) According to the established equipment fault diagnosis ontology representation, information extraction is carried out on IT side data through data configuration, the label of an entity is defined, an equipment triplet relation table is established, the relation is identified, and the equipment knowledge is represented as a triplet structure: the entity-relation-entity adopts a graph database to store data of the triplet knowledge, nodes in the graph database represent the entities, the relation among the knowledge is represented by the entity relation in the triplet, the relation among the knowledge is represented by the side with the direction, and the relation among the triplet is corresponding to the constructed equipment knowledge graph part relation and node visualization graph.
6. The knowledge-graph-driving-based equipment fault diagnosis method according to claim 1, wherein the step S1 comprises: the problems of data missing, abnormal value and repeated value existing in fault diagnosis are timely processed through data cleaning, feature extraction and data preprocessing, the frequency spectrum features, time domain features and wavelet transformation features of equipment are extracted from vibration signal, sound signal and temperature signal data, and normalization and standardization processing are carried out on feature data.
7. The knowledge-graph-driven equipment fault diagnosis method according to claim 1, wherein,
the maintenance proposal of the step S5 specifically includes:
1) Related knowledge and rules in the knowledge graph are continuously updated by combining the actual condition of the machine operation and maintenance records, so that the knowledge graph can adapt to the actual operation condition of the machine;
2) According to the real-time running state of the equipment, a fault diagnosis result driven by a knowledge graph is obtained, a maintenance mode of the machine is determined, and related maintenance tools and maintenance materials are prepared for maintenance.
8. The diagnosis system of a knowledge-graph-driven equipment fault diagnosis method according to claim 1, comprising:
and the data acquisition and processing module is used for: collecting sensor data, and preprocessing the collected data to obtain a data form suitable for knowledge graph construction and analysis;
knowledge graph construction module: representing the related ontology as an entity and a relation in the map by using an ontology modeling technology to obtain a knowledge map for equipment fault diagnosis;
and the fault characteristic data and entity and relation matching module in the knowledge graph: converting fault data of the sensor into feature vectors by utilizing a feature extraction technology, and obtaining entities and relations related to equipment faults through feature matching;
fault type and location analysis module: carrying out fault cause reasoning through the relation among the equipment structures to obtain fault types and positions;
fault diagnosis and repair advice module: and analyzing the diagnosis result and the maintenance method according to the related knowledge and rules in the knowledge graph to obtain maintenance suggestions of equipment faults.
9. An apparatus for implementing the knowledge-graph-based apparatus fault diagnosis method of any one of claims 1 to 7, comprising a processor and a memory;
wherein the processor is configured to perform a knowledge-graph-driven device fault diagnosis method according to any one of claims 1 to 7;
the memory is for storing executable instructions of the processor.
10. A computer storable medium having stored thereon a computer program, the computer program being executable by a processor to implement a knowledge-graph driven device fault diagnosis method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311228706.9A CN117350377A (en) | 2023-09-22 | 2023-09-22 | Knowledge graph driving-based equipment fault diagnosis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311228706.9A CN117350377A (en) | 2023-09-22 | 2023-09-22 | Knowledge graph driving-based equipment fault diagnosis method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117350377A true CN117350377A (en) | 2024-01-05 |
Family
ID=89358505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311228706.9A Pending CN117350377A (en) | 2023-09-22 | 2023-09-22 | Knowledge graph driving-based equipment fault diagnosis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117350377A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118014564A (en) * | 2024-04-10 | 2024-05-10 | 山东和兑智能科技有限公司 | Power equipment fault diagnosis system and method based on data driving |
-
2023
- 2023-09-22 CN CN202311228706.9A patent/CN117350377A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118014564A (en) * | 2024-04-10 | 2024-05-10 | 山东和兑智能科技有限公司 | Power equipment fault diagnosis system and method based on data driving |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111947928B (en) | Multi-source information fusion bearing fault prediction system and method | |
| CN109146246B (en) | Fault detection method based on automatic encoder and Bayesian network | |
| CN112418277B (en) | Method, system, medium and equipment for predicting residual life of rotating machine parts | |
| CN113128561A (en) | Machine tool bearing fault diagnosis method | |
| CN112179691B (en) | Mechanical equipment running state abnormity detection system and method based on counterstudy strategy | |
| CN107797537A (en) | A kind of prognostic and health management method applied to automatic production line | |
| CN108896299A (en) | A kind of gearbox fault detection method | |
| CN112414694B (en) | Equipment multistage abnormal state identification method and device based on multivariate state estimation technology | |
| CN117350377A (en) | Knowledge graph driving-based equipment fault diagnosis method and device | |
| EP3712728A1 (en) | Apparatus for predicting equipment damage | |
| CN115437358A (en) | Intelligent state monitoring and fault diagnosis system and fault diagnosis method for industrial robot | |
| CN114429152A (en) | Rolling bearing fault diagnosis method based on dynamic index antagonism self-adaption | |
| CN110969185A (en) | Equipment abnormal state detection method based on data reconstruction | |
| CN108825482B (en) | Fault detection method and system for axial plunger pump of airplane | |
| CN113313365A (en) | Degradation early warning method and device for primary air fan | |
| CN117312972A (en) | Method for identifying health state of scraper conveyor speed reducer | |
| CN112380782A (en) | Rotating equipment fault prediction method based on mixed indexes and neural network | |
| CN112308391A (en) | Real-time monitoring and anomaly detection method for equipment state based on neural network | |
| CN117370879A (en) | Real-time online fault diagnosis method and system for wind turbine gearbox | |
| US11339763B2 (en) | Method for windmill farm monitoring | |
| CN110956308A (en) | Chemical equipment health state prediction method based on time sequence | |
| CN116502043A (en) | Finish rolling motor state analysis method based on isolated forest algorithm | |
| CN116383750A (en) | Rolling bearing early-stage abnormality detection method based on windowed differential health index | |
| Cempel | Decomposition of the symptom observation matrix and grey forecasting in vibration condition monitoring of machines | |
| CN114330413A (en) | Fault type identification and positioning method for traction motor bearing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |