CN116771576A - Comprehensive fault diagnosis method for hydroelectric generating set - Google Patents
Comprehensive fault diagnosis method for hydroelectric generating set Download PDFInfo
- Publication number
- CN116771576A CN116771576A CN202310536536.4A CN202310536536A CN116771576A CN 116771576 A CN116771576 A CN 116771576A CN 202310536536 A CN202310536536 A CN 202310536536A CN 116771576 A CN116771576 A CN 116771576A
- Authority
- CN
- China
- Prior art keywords
- fault
- diagnosis
- configuration data
- mode
- data
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 171
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000004458 analytical method Methods 0.000 claims abstract description 50
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 36
- 230000004927 fusion Effects 0.000 claims abstract description 23
- 238000013499 data model Methods 0.000 claims abstract description 18
- 208000024891 symptom Diseases 0.000 claims description 42
- 238000004364 calculation method Methods 0.000 claims description 27
- 230000008569 process Effects 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 238000013145 classification model Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000011058 failure modes and effects analysis Methods 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 16
- 238000013461 design Methods 0.000 description 11
- 238000002405 diagnostic procedure Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000001364 causal effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000008439 repair process Effects 0.000 description 3
- 238000012216 screening Methods 0.000 description 3
- 238000013515 script Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03B—MACHINES OR ENGINES FOR LIQUIDS
- F03B11/00—Parts or details not provided for in, or of interest apart from, the preceding groups, e.g. wear-protection couplings, between turbine and generator
- F03B11/008—Measuring or testing arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Chemical & Material Sciences (AREA)
- Mechanical Engineering (AREA)
- Combustion & Propulsion (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Control Of Water Turbines (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
A comprehensive diagnosis method for the faults of a hydroelectric generating set comprises the following steps: step 1, preparing work before fault scanning diagnosis: acquiring power station configuration data, unit configuration data and equipment configuration data; step 2, starting fault diagnosis: diagnosing the power station configuration data, the unit configuration data and the equipment configuration data by utilizing a plurality of diagnosis modes; the diagnosis mode comprises a big data model algorithm, rule reasoning, a fault case matching mode and a fault tree analysis mode; step 3, obtaining the result of each diagnosis mode; step 4, performing fault diagnosis fusion decision on the diagnosis results of a plurality of diagnosis modes: step 5: and obtaining a final diagnosis result. The invention can diagnose the hydropower units of different types, and adopts a plurality of diagnosis modes to be combined, finally obtains the diagnosis result with larger probability, and improves the accuracy of diagnosis.
Description
Technical Field
The invention belongs to the technical field of hydroelectric generating set equipment fault diagnosis, and particularly relates to a comprehensive hydroelectric generating set fault diagnosis method.
Background
The hydraulic generator set has complex working ring, frequent working condition switching, the working state is influenced by hydraulic factors, mechanical factors and electrical factors, and the mutual coupling of the various factors is difficult to decompose, so that the difficulty of fault diagnosis of the hydraulic generator set is greatly improved. At present, the fault analysis and diagnosis of the hydroelectric generating set are mostly based on a clear fault mechanism and expert experience. The fault types with definite mechanisms are relatively few, and most of the faults are based on mechanism diagnosis under the influence of a single factor, and the diagnosis quantity and accuracy are insufficient to cover various operation conditions with complex working conditions; the diagnosis of expert experience is highly dependent on the level and experience of the expert, and it is difficult to perform a practical and efficient diagnostic analysis on different types of hydroelectric generating sets or faults that have not been experienced. Therefore, the accurate and effective fault diagnosis of the implementation of the hydroelectric generating set is always an industry pain point and a difficult problem.
Chinese patent document CN115685972a discloses a device fault diagnosis method based on FMECA and ontology, the adopted diagnosis method is: and combining with the analysis thought of FMECA, extracting the knowledge of the fault diagnosis field of the equipment, and taking the knowledge as the basis for constructing a fault diagnosis ontology. For the device to be diagnosed, the device structure and the functional flow are analyzed first. For general industrial equipment, the structure thereof can be divided into an overall equipment level, a system level, a component level, and a component level. Summarizing the equipment corresponding to each hierarchical structure in the diagnosed equipment, analyzing the workflow of the equipment according to the equipment structure, and summarizing the possible influence on the same-level equipment or high-level equipment when each hierarchical equipment fails. The diagnosis process only adopts a single diagnosis tool, the diagnosis accuracy is not high, and misdiagnosis is easy to occur.
Disclosure of Invention
In view of the technical problems existing in the background technology, the comprehensive diagnosis method for the hydroelectric generating set faults can diagnose the hydroelectric generating sets of different types, and a plurality of diagnosis modes are combined, so that a diagnosis result with high probability is finally obtained, and the accuracy of diagnosis is improved.
In order to solve the technical problems, the invention adopts the following technical scheme:
a comprehensive diagnosis method for the faults of a hydroelectric generating set comprises the following steps:
step 1, preparing work before fault scanning diagnosis: acquiring power station configuration data, unit configuration data, equipment configuration data, FMEA fault mode configuration data and diagnostic workflow configuration data;
step 2, starting fault diagnosis: diagnosing the power station configuration data, the unit configuration data and the equipment configuration data by utilizing a plurality of diagnosis modes; the diagnosis mode comprises a big data model algorithm, rule reasoning, a fault case matching mode and a fault tree analysis mode;
step 3, obtaining the result of each diagnosis mode;
step 4, performing fault diagnosis fusion decision on the diagnosis results of a plurality of diagnosis modes:
step 5: and obtaining a final diagnosis result.
Preferably, the substeps of step 1 are:
step 1.1, acquiring power station configuration data, wherein the power station configuration data comprises a power station code, a power station configuration data associated unit information table and a power station configuration data id;
step 1.2, the unit configuration data comprises a unit code, a unit configuration data association equipment table and a unit configuration data id;
step 1.3, the device configuration data comprises a device name, a device code and an introduced system device information table associated with the device configuration data;
and step 1.4, acquiring all fault mode information and equipment logic positions.
Preferably, in step 2, each diagnosis mode forms a diagnosis node, and the diagnosis node comprises a big data model algorithm node, a rule node, a fault case library node and a fault tree analysis mode node; during diagnosis, simultaneously entering a big data model algorithm node, a rule node, a fault case library node and a fault tree analysis mode node for diagnosis; and diagnosing the power station configuration data, the unit configuration data and the equipment configuration data in sequence.
Preferably, the result table in the step 3 comprises a big data model algorithm result table, a rule result table, a fault case base result table and a fault tree analysis mode result table.
Preferably, the substeps of step 4 are:
step 4.1, filtering the diagnosis results of all the diagnosis modes, and taking the result with the maximum probability of different fault modes of all the diagnosis modes;
and 4.2, carrying out fusion calculation processing on the filtered results of the diagnosis modes to obtain a final fault mode and probability thereof.
Preferably, the fusion calculation of step 4.2 is divided into two cases, the first: each tool result has only one failure mode; second kind: each tool result has multiple failure modes;
first kind: each diagnosis mode has only one fault mode, and the following table is adopted for carrying out:
table 1 fusion decision contrast table with only one failure mode for each diagnosis mode result
S is the fault probability of the final fusion result.
Preferably, the second: and if the results of each diagnosis mode have multiple fault modes, a DS evidence reasoning fusion decision model is needed to be adopted for calculation.
Preferably, the fault tree analysis method comprises the following steps:
1) Dividing the fault into three layers of fault top events, intermediate events and basic events, wherein the fault top events comprise a plurality of intermediate events, and the intermediate events comprise a plurality of basic events;
2) The fault events of all the levels are connected through logic gates to form a tree structure for fault analysis;
3) The fault events of each level correspond to corresponding fault symptoms.
Preferably, the rule reasoning process is:
s1, establishing rules according to empirical data of the relation between known fault modes and fault symptoms, and judging the occurrence probability of the fault modes based on input fault symptom data;
s2, establishing an analysis model for time sequence data of the equipment, and analyzing whether related fault symptoms occur or not
S3, packaging the equipment symptom data analyzed in the S2 into a JSON structure input by the rule interface, and requesting the rule REST interface established in the S1;
s4, the rule execution engine processes the REST interface request, inputs data to the rule, judges the occurrence probability of the fault mode and outputs the probability.
Preferably, the big data model algorithm is processed as follows: firstly, inputting a marked fault sample, then preprocessing data, and then training through a classification model; when the prediction is carried out, the data is subjected to the same preprocessing work, and then the result is output through the classification model.
The invention has the following beneficial effects:
the invention can diagnose the hydropower units of different types, and adopts a plurality of diagnosis modes to be combined, finally obtains the diagnosis result with larger probability, and improves the accuracy of diagnosis.
Drawings
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
FIG. 1 is a diagnostic flow chart of the present invention;
fig. 2 is a diagnostic process diagram of example 1 of the present invention.
Detailed Description
The preferable scheme is as shown in fig. 1, and the comprehensive fault diagnosis method for the hydroelectric generating set comprises the following steps:
step 1, preparing work before fault scanning diagnosis: acquiring power station configuration data, unit configuration data, equipment configuration data, FMEA fault mode configuration data and diagnostic workflow configuration data;
failure mode and impact analysis (FMEA) is used for carrying out standardized description on failure modes, failure symptoms, failure categories, failure reasons, characteristic parameters, severity, occurrence degree and fault related information characteristics of detectable degree of the whole unit level and the equipment level, and establishing diagnosis analysis of failure modes and impact analysis of the unit system according to the structure and functional connection relation among equipment systems or components;
step 1.1, acquiring power station configuration data, wherein the power station configuration data comprises a power station code, a power station configuration data associated unit information table and a power station configuration data id;
and (3) associating the power station configuration data with a unit information table: and all unit configuration data contained in a certain power station.
Step 1.2, the unit configuration data comprises a unit code, a unit configuration data association equipment table and a unit configuration data id;
a unit configuration data association device table: all equipment configuration data contained in a certain unit.
Step 1.3, the device configuration data comprises a device name, a device code and an introduced system device information table associated with the device configuration data;
an incoming system device information table of device configuration data associations: all sub-devices that a certain device contains.
And step 1.4, acquiring all fault mode information and equipment logic positions.
Step 2, starting fault diagnosis: diagnosing the power station configuration data, the unit configuration data and the equipment configuration data by utilizing a plurality of diagnosis modes; the diagnosis mode comprises a big data model algorithm, rule reasoning, a fault case matching mode and a fault tree analysis mode; the diagnosis mode adopted by the invention is the existing diagnosis mode, and the function introduction is as follows:
diagnostic mode 1, fault tree analysis mode (FTA) described above: is a quantitative analysis method which can help engineers to determine the reliability and safety index of equipment and to formulate proper precautions. Fault tree analysis is often used in combination with other analysis methods such as FMEA (fault mode and impact analysis) to improve the accuracy and reliability of the analysis. The fault tree analysis mode comprises the following steps: dividing the fault into three layers of fault top events, intermediate events and basic events, wherein the fault top events comprise a plurality of intermediate events, and the intermediate events comprise a plurality of basic events; the fault events of all the levels are connected through logic gates to form a tree structure for fault analysis; the fault events of each level correspond to corresponding fault symptoms. The specific analysis process is as follows:
s1, creating an FTA logic model tree according to possible accidents of the system or information provided by the accidents or based on FMEA;
s2: instantiating the created logical model tree to generate an FTA instantiated model tree
And S3, the FTA instantiation model completes analysis and deduction of the whole fault tree through a self fault diagnosis algorithm and by combining an algorithm model, diagnosis rules, scripts and measurement point data, so that the fault occurrence probability is obtained.
The step S1 specifically comprises the following substeps:
s11: according to possible accidents or accidents of the system, a certain system fault with the greatest influence is selected as a top event. And binds the logical devices and fault diagnostic tools (algorithm models, diagnostic rules, scripts, and points) for fault tree top events.
And S12, decomposing the cause of the system fault step by step into intermediate events, and binding logic equipment and fault diagnosis tools (algorithm models, diagnosis rules, scripts and measuring points) for the fault tree top event.
S13: events that cannot or need to be resolved are taken as base events, i.e., bottom event locations. And sets a fault diagnosis tool and default occurrence probability for the bottom event.
S14: the fault tree nodes are connected by using a logical relation operation method (such as logical AND, logical OR) so as to finish a complete drawing of the logical fault tree.
S15: after the fault tree is drawn, the logic fault is created through integrity check.
The step S2 specifically includes the following substeps:
s21: the logic model is converted into an instantiation model by selecting an instantiation factory and equipment.
S22: in the process of instantiation, the logic equipment, the measuring point and the like bound by the node are converted into the physical equipment and the physical measuring point.
S22: converting a logical algorithm model into a physical algorithm model
S23: if the fault tree node exists as a reference subtree, then the subtree needs to be converted to an instantiation tree as well.
S24: and converting the logical fault tree model name into an instantiation model name, and completing the instantiation creation of the fault tree.
The step S3 specifically includes the following substeps:
s31: the instantiated fault tree can be formally put into use, and can be called through a result interface or automatically executed through a periodic task.
S32: and acquiring a diagnosis algorithm bound with the fault tree node, calling a corresponding diagnosis interface according to different types of diagnosis algorithms, acquiring a diagnosis result, and setting a node alarm result and probability.
S33: if the failed node binds the failed subtree, S32 is repeatedly performed.
S34: substituting the fault probability obtained in the step S32 into a fault tree, and obtaining the fault occurrence probability of default setting if the node is not bound to the fault diagnosis algorithm. According to the FTA fault diagnosis algorithm, such as a minimum cut set, a minimum path set and the like, analysis and deduction are carried out to obtain whether a fault tree top event has faults and the probability of the faults.
The diagnosis mode 2 is characterized in that the rule reasoning process is as follows:
s1: rules are established in the rule configuration tool according to empirical data of known relationships between failure modes and failure symptoms, and the probability of failure modes can be judged based on the input failure symptom data. And after the rule is issued, providing REST interface service.
The rule configuration process consists of two parts, namely data design and logic design.
In data design, input data, output data, and intermediate calculation result data in a fault diagnosis rule need to be defined, including names and types of data variables. Multiple associated data may be packaged as complex variable types as desired. The defined data variables may be used in rule logic. The business parameters in the rule logic can be defined as constants, so that the readability and the maintainability are improved.
In logic design, a series of rule blocks comprising data processing and judgment logic are constructed by dragging various rule elements, including condition judgment, data assignment, data processing functions, basic four-rule operations, etc., according to the processing flow of input data.
When the rule is issued, which is the input data of the rule is specified from the data variables in the data design, which is the data which needs to be output after the rule execution is completed, the REST interface after the rule is issued receives the input data in the JSON format, and the result in the JSON format is returned.
S2: establishing an analysis model for time sequence data of equipment to analyze whether related fault symptoms occur or not
S3: and according to the input requirement appointed during the rule release, packaging the equipment symptom data analyzed by the S2 into a JSON structure input by the rule interface, and requesting the rule REST interface established by the S1.
S4: the rule configuration back-end rule execution engine processes the REST interface request, inputs data to the rule, judges the occurrence probability of the fault mode according to the designed logic and outputs the fault mode.
The rule execution engine compiles each rule block into a rule tree, runs the rule trees in turn according to the sequence of each rule block in the rule, and updates the related data variables. And finally outputs specified output data including failure modes and occurrence probabilities thereof.
Aiming at a certain fault or problem, the related information is arranged, generalized and classified, so that a visual knowledge structure diagram is formed. Such charts often include content of the cause, manifestation, method of troubleshooting, solutions, and the like, to help one better understand and solve a particular fault or problem.
Diagnostic mode 3, fault case matching mode: the information of the fault cases is stored and organized by a fault case library, and when a new fault occurs, the system automatically compares and matches the data in the case library to find the most similar fault case and extract useful information therefrom to infer. And inquiring time sequence data of corresponding measuring points according to the information of time periods, equipment, units and the like required to be diagnosed. The diagnostic procedure is as follows:
s1: and carrying out symptom calculation according to the time sequence data of the measuring points, wherein each symptom corresponds to different calculation logic. And judging which symptoms and the credibility of the symptoms occur in the period.
S2: grouping according to the calculated symptoms according to the corresponding fault modes, and carrying out normalization processing on the calculated symptom credibility according to fixed weights to obtain the credibility of each fault mode.
S3: and calculating the similarity of the occurrence symptom and the corresponding symptom of each case in the case library through a plurality of similarity algorithms.
S4: and finally, combining the reliability and the similarity of the fault modes to obtain the reliability of each fault mode, namely the probability of fault occurrence.
Diagnostic mode 4, the diagnostic procedure of the big data model algorithm is as follows:
s1, inquiring time sequence data of corresponding measuring points based on a KKS example equipment tree according to a time period and unit equipment to be diagnosed.
S2, performing table conversion processing (pivot, converting id, time, v format data into id1, id2, id3 … … format) on the time sequence data
S3, performing front value repair on the active power measuring point and the unit water head measuring point due to the fact that time misalignment occurs after the meter is turned (data exist at part of measuring points at the same moment and data do not exist at part of measuring points)
S4, stability screening is carried out on the data according to the active power measuring point, wherein the specific logic is as follows
S4.1, setting a judging window t and a fluctuation threshold M, and judging whether a unit stably operates in the window with the duration of t, namely whether the fluctuation amplitude of active power (the maximum value-minimum value in the window) is smaller than M
S4.2, sliding the window backwards, and repeatedly judging logic 1
S5, performing front value repair on all the empty data, wherein the reason that the front value repair is performed in two steps is that the authenticity of the data needs to be ensured
S6, screening measuring point data exceeding the measuring range of the sensor
S7, screening the threshold value of the target measuring point, and marking if the target measuring point exceeds the threshold value set by the user
And S8, storing all marked data into a data table, wherein the table is used for data interaction between the model development part and the fault diagnosis module.
In addition, the diagnosis modes 1-4 are quantitative diagnosis, and when the quantitative diagnosis cannot be performed, the diagnosis is performed by adopting a knowledge graph diagnosis mode, and the diagnosis process is as follows:
s1: according to the hydroelectric generating set and the related knowledge of the faults, a knowledge and relation model for diagnosis is designed, knowledge nodes comprise a system, a subsystem, a set, a fault case library, an FMEA (failure and impact analysis) library, an FTA (fault analysis tree) library, early warning information, measuring points and the like, and the association relation among the knowledge is combined to form a conceptual model of a hydroelectric fault diagnosis knowledge map;
(1) Knowledge maps are used to describe various entities or concepts and their relationships that exist in the real world, which ultimately constitute a huge semantic network graph. Wherein, the nodes represent entities or concepts, and the edges are composed of attributes or relationships.
(2) Nodes (or ontologies). Representing a concept content, an entity, an attribute or an event. Entities such as a main transformer, a business person, etc., or abstract concepts such as big data platforms, knowledge maps, generator sets, etc.
(3) Edges (or arcs). The node correspondence "relationship" indicates that some kind of relationship exists between nodes. The labels of the edges indicate the type of relationship or relationships between entities, such as generic relationships, constitutive relationships, etc.
In the fault diagnosis knowledge graph of the hydroelectric generating set, the nodes comprise a system, components, equipment, a fault case library, an FMEA (failure and impact analysis) library, an FTA (fault analysis tree) library, an early warning library, a measuring point library and the like, and the relationships of generic relationships, sequence relationships, causal relationships, constructional relationships and the like among the nodes jointly form a conceptual model (FaultDiagnosisKG Conception Model) of the fault diagnosis knowledge graph.
S2: and (5) knowledge classification. The link is important, and prepares for later knowledge service, knowledge calculation, knowledge reasoning, knowledge visualization and the like, and divides the nodes into a system class, a device class, a fault class, a method class, a data class, a user class and the like; the relationship class is divided into generic relationship, composition relationship, aggregation relationship, instance relationship, sequence relationship, association relationship, dependency relationship, causal relationship, relatives and the like;
s3: refining the data structure of the knowledge node, i.e. combing the knowledge node as a target to be relevant attributes facilitating fault analysis, such as a unit (equipment KKS code, equipment specification, commissioning time, equipment type, manufacturer, etc.), fault cases (case codes, case names, case brief descriptions, case source types, case related pictures, etc.), faults (fault codes, fault modes, fault phenomenon descriptions, fault levels, fault effects, etc.), fault handling (processing measures, processing steps, processing effects), fault characteristics (feature data, graphic characteristics, fault data samples, fault recovered data samples), etc.;
s4: using a knowledge graph construction tool, constructing a concept knowledge model into a graph database Neo4j, and converting a concept layer knowledge template into triples (subjects, predicates, objects), such as (units, association, faults), (water turbines, generic relations, units), (fault reasons, causal relations, faults), (fault characteristics, association, fault symptoms), (faults, relatives, defect recorders), (early warning, association, units) and the like;
s5: knowledge extraction preparation. Performing the mapping configuration for the structured knowledge nodes, configuring knowledge source configuration, knowledge labeling and other works for semi-structured and unstructured data, and performing the configuration of knowledge extraction task creation, task execution parameters (such as task execution frequency, execution coverage range and knowledge coverage mode) and the like;
s6: knowledge extraction and updating. The knowledge extraction task is executed regularly or regularly, instantiation data are stored in a graph database according to a conceptual model of a prefabricated hydropower fault diagnosis knowledge graph, knowledge integration is carried out by auxiliary manual means when entity definition inconsistency and the like occur, and finally the hydropower fault diagnosis knowledge graph is formed;
s7: knowledge calculation and reasoning. And carrying out knowledge calculation or knowledge reasoning on the formed hydropower fault diagnosis knowledge graph according to the fault diagnosis reasoning rule and the related model to obtain a fault diagnosis result and a fault analysis process record.
S8: knowledge visualization. And displaying the fault diagnosis result and the analysis process to a user in a proper form.
Each diagnosis mode forms a diagnosis node, and the diagnosis node comprises a big data model algorithm node, a rule node, a fault case library node and a fault tree analysis mode node; during diagnosis, simultaneously entering a big data model algorithm node, a rule node, a fault case library node and a fault tree analysis mode node for diagnosis; and diagnosing the power station configuration data, the unit configuration data and the equipment configuration data in sequence.
The specific diagnosis process is as follows:
table 2 diagnostic process table
Step 3, obtaining the result of each diagnosis mode; the result table comprises a big data model algorithm result table, a rule result table, a fault case base result table and a fault tree analysis mode result table.
Step 4, performing fault diagnosis fusion decision on the diagnosis results of a plurality of diagnosis modes:
step 4.1, filtering the diagnosis results of all the diagnosis modes, and taking the result with the maximum probability of different fault modes of all the diagnosis modes;
and 4.2, carrying out fusion calculation processing on the filtered results of the diagnosis modes to obtain a final fault mode and probability thereof.
First kind: each tool result has only one failure mode; second kind: each tool result has multiple failure modes;
step 4.2.1, first: each diagnosis mode has only one fault mode, and the following table is adopted for carrying out:
table 1 fusion decision contrast table with only one failure mode for each diagnosis mode result
Note that: if a tool is generating a fault result or the probability of occurrence of the fault is 0, the calculation is not participated.
S is the fault probability of the final fusion result;
step 4.2.1, second: and if the results of each diagnosis mode have multiple fault modes, a DS evidence reasoning fusion decision model is needed to be adopted for calculation. DS evidence reasoning-related literature is: the Chinese patent document CN110166437B discloses a method for selecting a moving target defense optimal strategy based on DS evidence reasoning.
Step 5: and obtaining a final diagnosis result.
The final diagnosis results comprise rule diagnosis results, case matching results, fault tree analysis mode matching results, model matching results and knowledge reasoning results, and the final diagnosis results are finally integrated and converged.
Specifically, detailed result information of various diagnostic results is checked at a fault diagnosis detail interface. The method comprises the steps of rule diagnosis results, case matching results, fault tree analysis mode matching results, model matching results and knowledge reasoning results, integrating and converging various diagnosis results finally, synthesizing various model diagnosis probabilities and historical confidence, and giving fault diagnosis decision results through fusion decisions.
1) Rule result display:
(1) detail value description: based on the fault modes, taking the maximum probability calculated by each fault mode as a result, and taking multiple modes if multiple fault modes exist
(2) Diagnosis is based on the description: which fault mode is hit, list the symptom name and occurrence
Based on the symptoms: assuming that a mass imbalance hits a combination of a certain sign in 110, the confidence of each sign and the confidence of the combination are listed
Based on the characteristic parameters: assuming that the fault mode is connected with the calculation of a group of characteristic parameters, the calculation result and the calculation method of the group of characteristic parameters are displayed
2) And (3) case result display:
(1) detail value description: based on the fault modes, taking the maximum probability calculated by each fault mode as a result, and taking multiple modes if multiple fault modes exist
(2) Diagnosis is based on the description: is divided into two kinds of mechanism symptoms and characteristic parameters
Based on the symptoms: the symptom ranges participated in the calculation are listed, which occur and do not occur, and the occurrence of the fault mode is deduced based on symptom combination
Based on the characteristic parameters: characteristic parameters participating in calculation at this time are listed
3) And (3) displaying results of a fault tree analysis mode:
(1) detail value description: based on the fault tree analysis mode tree, taking the hit fault mode in each fault tree analysis mode tree, and checking the whole path hit in the fault mode through the fault influence
(2) Diagnosis is based on the description: in the fault tree analysis mode tree, binding algorithm/rule fault mode nodes and whether the fault mode nodes occur or not to list
4) Knowledge context result display:
(1) detail value description: forming a complete correlation map of the reasoning based on the symptoms or the fault modes;
(2) diagnosis is based on the description:
based on the symptoms: if the calculation has symptoms, the relevant symptoms are obtained from a symptom result table, and the corresponding fault mode, fault reason and treatment measures are hit based on symptom reasoning
Based on the characteristic parameters: if no symptom is found in the calculation, the method is started by a rule ID calculated by the characteristic parameters, a corresponding fault mode is found, and fault cause processing measures related to fault mode reasoning and the like are adopted
5) And (3) showing an algorithm calculation result:
(1) detail value description: based on the fault modes, taking the maximum probability calculated by each fault mode as a result, and taking multiple modes if multiple fault modes exist
(2) Diagnosis is based on the description: clicking to view the corresponding basis based on each operator step
6) And (3) fusion calculation result display:
based on the calculation results of fault cases, fault tree analysis modes, big data model algorithms, rules and knowledge venation graphs, merging according to fault modes, comprehensively displaying (the display strategy can be adjusted) the data of 3 fault modes with the highest probability in each tool, and simultaneously integrating and outputting information of associated fault symptoms, fault categories, fault reasons, characteristic parameters, severity, occurrence degree, detectable degree and the like, so that scientific basis is provided for judging the equipment state by service personnel.
Example 1:
as shown in fig. 2, the full flow of fault diagnosis implementation is described in detail by taking transformer fault identification as an example.
Step one, acquiring configuration data:
the fmea configuration implementation is as follows:
1.1, establishing an FMEA standard template, and selecting equipment to be configured for newly adding FMEA data to the standard;
1.2, configuring corresponding fault modes, priorities and other information for (multi-level) equipment according to different levels of a system, a subsystem, equipment, a component and the like, and realizing the association relation between the equipment and the fault modes;
1.3, automatically acquiring characteristic parameters and fault symptoms aiming at all configured fault modes, or manually configuring the characteristic parameters corresponding to the fault modes;
1.4 configuring corresponding risk assessment information for all fault modes, wherein the risk assessment information comprises information such as severity (S), occurrence degree (O), detection degree (D) and the like, and finally forming overview information of FMEA;
step two, performing diagnosis in a plurality of diagnosis modes (or by using a plurality of diagnosis tools);
2. the implementation mode of the transformer fault case matching is as follows:
2.1 manual creation of transformer fault cases: selecting a logic device, instantiating the device and the fault mode, and a fault time period.
2.2 through the fault mode, all the characteristic parameters related to the fault mode of the transformer can be queried through FMEA, the characteristic parameters are displayed in a symptom attribute field and data characteristics, and the numerical values of the characteristic parameters can be all brought out through the instantiation equipment and the fault time period.
And 2.3, according to the fault mode, the fault cause, the processing measures, the detection method and other contents of the fault mode can be queried through the FMEA, and the fault cause is automatically carried out.
2.4 by instantiating the device and the failure time period, failure data samples can be generated and correlated. Information content such as alarms can also be associated.
2.5 case preservation and Release
2.6, matching according to the similarity calculation according to the numerical value of the characteristic parameter, wherein the matching algorithm uses a gray correlation matching algorithm.
2.7, the process of case matching is to extract all fault modes under the logic device so as to calculate the characteristic parameters related to the fault modes according to the time period, then match the values of the characteristic parameters in the cases through a matching algorithm, and calculate the similarity ordering of all cases.
2.8 case matching results will present data feature parameters and case feature parameters, and similarity ranking for all cases.
3. The transformer rule (rule reasoning) judging implementation mode is as follows:
3.1 rule construction
(1) Rule data design
a) An input data object is created, and 6 measuring points related to a three-ratio method are selected.
b) A data object is created for output by the intermediate data set, comprising the last variable to be output, and a save variable for intermediate data during data processing.
c) Some traffic parameters are set to constant as needed. The purpose of setting the constant: making the data readable enhances maintainability of the data.
(2) Rule logic design
a) And finishing logic design of the three-ratio method, and performing rule arrangement on the execution process of the three-ratio method according to sequential execution.
b) The general rule design process includes:
and (3) data processing: and finishing processing of the data, such as extracting attribute values from the object structure, and processing the variables by using the functions to obtain target variables.
And (3) condition logic judgment: business logic is performed in accordance with the if … then … else structure.
The process of the three ratio rule is generally as follows: taking a measuring point value; judging whether the starting condition of the three-ratio method is met; calculating a three-ratio value; for each ratio item; coding according to different ratios; and according to the code combination of the three ratio items, giving out data such as fault reasons and the like.
(3) And (3) rule rapid test: the input data is provided for testing, and the physical measuring points can be also connected for execution to check whether the input result meets the expectations.
3.2 regular scheduling:
(1) rule instantiation: an instance is created specifying the device to which the rule applies.
(2) Rule tasks: rules requiring periodic execution may create timed tasks that contain multiple rule instances.
4. The transformer knowledge graph judgment implementation mode is as follows:
constructing a knowledge system framework, classifying knowledge according to requirements, and determining the type and attribute information of each classification; meanwhile, the relationships among the knowledge are predefined, various relationships are defined, and maintenance and management are carried out.
4.1 knowledge representation
The knowledge representation determines the yield target of the graph construction, namely a semantic description framework, a Schema, an entity naming and an ID system of the knowledge graph;
knowledge representation standard languages include RDFS and OWL, where RDFS provides a simple description of classes and attributes, thereby providing a lexical modeling language for RDF data. More rich definitions require fir to OWL ontology description language.
RDF (triples) are used as a basic data model, and its basic logical structure contains three parts, namely a subject (individual: instance of class), a predicate (attribute) and an object (individual or instance of data type).
Knowledge acquisition: at present, fault mode information takes an FMEA library as a main source, knowledge extraction of structured data is oriented, and a logic table is usually mapped to RDF data through triple mapping, wherein the triple mapping is a rule capable of mapping each row in the logic table into a plurality of RDF triples. The triplet rules mainly consist of two parts: one subject map and multiple predicate-object maps. The second is that the predicate-object map includes a predicate map and an object map.
4.2 construction of the atlas
(1) Schema definition:
the Schema defines the entities, attributes and relationships in the knowledge graph clearly, and defines the feasible scope of the knowledge graph, namely the Schema defining the knowledge graph is equivalent to constructing the ontology of the knowledge graph.
Based on the requirements of fault diagnosis application, user requirements are taken as starting points, data statistics is taken as evidence, and different topics are built to meet the user requirements.
(2) And (3) body map construction:
under the fault diagnosis application scene, based on the Schema definition, the body map construction is carried out on the fault mode, the symptom and the equipment, and the map comprises entities, attributes and the relationship between the entities or between the entities and the attributes. And constructing the map of the defined ontology.
4.3 knowledge fusion:
the aim of knowledge fusion is to establish the connection between the bodies, so that the knowledge graphs between the bodies can be communicated with each other, and the mutual operation between the bodies is realized.
And calculating and judging the relation between the ontology and the ontology through an atlas algorithm and a rule, and automatically generating a new relation atlas.
5. Transformer fault recognition algorithm model configuration:
the marked faulty samples are first input, followed by preprocessing of the data, such as format conversion, time alignment, and removal of null data. And then trained through classification models (e.g., naive bayes, support vector machines, etc.).
When the prediction is carried out, the data is subjected to the same preprocessing work, and then the result is output through the classification model.
The training data is derived from samples accumulated during the running period of the unit, contains a large amount of fault and normal data, and establishes a sample library after summarization.
5.2 the fault data is subjected to label processing, namely, fault type identification such as high-temperature discharge, low-temperature discharge, partial discharge and the like is added to each piece of data. At model run time, fault codes such as "30", "46", etc., which can be identified by the program are replaced. When the calculation output is completed, the fault description is converted back again.
6. The transformer FTA implementation is as follows:
6.1 creating a transformer logic FTA, configuring FTA nodes with logic devices and fault modes, and associating fault diagnosis methods with configurable nodes, such as configuring three-ratio rules on high temperature overheat (above 700 c) nodes.
6.2, after the logic FTA configuration is completed, the integrity detection is carried out, and the detection is completed, stored and released.
6.3 logical FTA is instantiated onto specific plants and units, such as Ge Zhou dam plant 2B transformers.
6.4, the logic FTA is instantiated to generate an FTA instance, and the FTA instance can be edited, stored and released according to different stations and units.
6.5 diagnosis of FTA: advanced application-fault diagnosis sends request parameter KKS code, fault time period. The FTA matches out the corresponding FTA example according to KKS, and starts the diagnostic method that all nodes are correlated on the FTA example, the diagnostic result is returned after the diagnostic method is operated, the diagnostic result is reflected and marked as red, green on the FTA node. And the occurrence probability of the top event is calculated according to the faults fed back by each node.
Step three: in the workflow, a failure mode and associated diagnostic procedures and diagnostic tools are configured.
1. The creation flow is as follows:
the process is created, the model name, the model key and the model description are input, and note that the model key cannot be a pure number and is needed to be used when the interface is called later to play a key. After entering the determination, if the preservation is successful, the blank flow is created.
2. And (3) flow design:
(1) start flow design
The case, rule, FTA, knowledge venation graph, flow node events of the algorithm model are dragged into the design panel, and the different elements need to be connected with arrows (sequence flow).
For HTTP tasks, the listener is first executed because the rest service generally determines whether success or failure is written in the backend Code, rather than relying solely on the Status Code of HTTP. The execution listener here is therefore to let the workflow know whether this HTTP service call was really successful by configuring some variables. If the don't care call is successful, no configuration is needed here.
(2) Entering the + number of the event, and adding an event;
events: select end
Class: the fixed value com.dhcc.flotable.common.Listener.httptaskexecuitelter;
entering the plus number of the Name, and adding a field;
name: fixed value success
The expression: the responseRes is fixed, the remainder being configured according to how much the field of the actual interface is equal to successful. For example, the return of the actual interface is { status code:0, data: [ XXXXX ] }, where status code is represented as 0, then it is configured here as $ { responseres status code= =0 };
entering the plus number of the Name, and adding a field again;
name: fixed value errMsg;
the expression: the responseRes is fixed, the rest is configured according to the error information field of the actual interface. For example, the return of the actual interface is { status code: -77777, error message:' error has occurred }, then it is configured here as $ { responseres. Error message };
after the listener configuration is performed, other attributes of the HTTP service task need to be configured:
(3) the request method comprises the following steps: selecting GET or POST
Request URL: the { requestHost } is a fixed value, and the following address is filled in according to the actual situation;
request header: the second row configures Content-Type application/json if the interface needs to be imported into the body, and configures Content-Type application/x-www-form-url coded if the interface does not need to be imported;
(4) request body: the fixed value $ { requestBody }, note that the incoming variable requestBody and corresponding value are required in the interface parameters. This value is passed in as a parameter when the HTTP service is invoked;
request timeout time: a fixed value 100000 (or modified according to the actual situation);
response variable name: a fixed value responseRes;
save the response as JSON: a fixed value true;
step four: fusion decision making and result display: in the advanced fault diagnosis application, the starting and ending time of information scanning is selected, and all units of the whole station or a designated single unit device is selected according to the requirement to carry out fault scanning diagnosis. The process management tool can simultaneously start various diagnostic tools in the background, and return diagnostic results to the front end interface display of the fault diagnosis advanced application respectively. Detailed result information of various diagnosis results can be checked on a fault diagnosis interface.
The above embodiments are only preferred embodiments of the present invention, and should not be construed as limiting the present invention, and the scope of the present invention should be defined by the claims, including the equivalents of the technical features in the claims. I.e., equivalent replacement modifications within the scope of this invention are also within the scope of the invention.
Claims (10)
1. A comprehensive diagnosis method for a hydroelectric generating set fault is characterized by comprising the following steps:
step 1, preparing work before fault scanning diagnosis: acquiring power station configuration data, unit configuration data, equipment configuration data, FMEA fault mode configuration data and diagnostic workflow configuration data;
step 2, starting fault diagnosis: diagnosing the power station configuration data, the unit configuration data and the equipment configuration data by utilizing a plurality of diagnosis modes; the diagnosis mode comprises a big data model algorithm, rule reasoning, a fault case matching mode and a fault tree analysis mode;
step 3, obtaining the result of each diagnosis mode;
step 4, performing fault diagnosis fusion decision on the diagnosis results of a plurality of diagnosis modes:
step 5: and obtaining a final diagnosis result.
2. The hydro-power generating unit fault integrated diagnosis method according to claim 1, wherein: the substeps of step 1 are:
step 1.1, acquiring power station configuration data, wherein the power station configuration data comprises a power station code, a power station configuration data associated unit information table and a power station configuration data id;
step 1.2, the unit configuration data comprises a unit code, a unit configuration data association equipment table and a unit configuration data id;
step 1.3, the device configuration data comprises a device name, a device code and an introduced system device information table associated with the device configuration data;
and step 1.4, acquiring all fault mode information and equipment logic positions.
3. The hydro-power generating unit fault integrated diagnosis method according to claim 1, wherein: in the step 2, each diagnosis mode forms a diagnosis node, and the diagnosis node comprises a big data model algorithm node, a rule node, a fault case library node and a fault tree analysis mode node; during diagnosis, simultaneously entering a big data model algorithm node, a rule node, a fault case library node and a fault tree analysis mode node for diagnosis; and diagnosing the power station configuration data, the unit configuration data and the equipment configuration data in sequence.
4. The hydro-power generating unit fault integrated diagnosis method according to claim 3, wherein: the result table in the step 3 comprises a big data model algorithm result table, a rule result table, a fault case base result table and a fault tree analysis mode result table.
5. The method for comprehensively diagnosing faults of a hydroelectric generating set according to claim 4, wherein the method comprises the following steps of: the substep of step 4 is:
step 4.1, filtering the diagnosis results of all the diagnosis modes, and taking the result with the maximum probability of different fault modes of all the diagnosis modes;
and 4.2, carrying out fusion calculation processing on the filtered results of the diagnosis modes to obtain a final fault mode and probability thereof.
6. The method for comprehensively diagnosing faults of a hydroelectric generating set according to claim 5, wherein the method comprises the following steps of: the fusion calculation of step 4.2 is divided into two cases, the first: each tool result has only one failure mode; second kind: each tool result has multiple failure modes;
first kind: each diagnosis mode has only one fault mode, and the following table is adopted for carrying out:
table 1 fusion decision contrast table with only one failure mode for each diagnosis mode result
S is the fault probability of the final fusion result.
7. The hydro-power generating unit fault integrated diagnosis method according to claim 6, wherein:
second kind: and if the results of each diagnosis mode have multiple fault modes, a DS evidence reasoning fusion decision model is needed to be adopted for calculation.
8. The hydro-power generating unit fault integrated diagnosis method according to claim 1, wherein: the fault tree analysis mode comprises the following steps:
1) Dividing the fault into three layers of fault top events, intermediate events and basic events, wherein the fault top events comprise a plurality of intermediate events, and the intermediate events comprise a plurality of basic events;
2) The fault events of all the levels are connected through logic gates to form a tree structure for fault analysis;
3) The fault events of each level correspond to corresponding fault symptoms.
9. The hydro-power generating unit fault integrated diagnosis method according to claim 1, wherein: the rule reasoning process is as follows:
s1, establishing rules according to empirical data of the relation between known fault modes and fault symptoms, and judging the occurrence probability of the fault modes based on input fault symptom data;
s2, establishing an analysis model for time sequence data of the equipment, and analyzing whether related fault symptoms occur or not
S3, packaging the equipment symptom data analyzed in the S2 into a JSON structure input by the rule interface, and requesting the rule REST interface established in the S1;
s4, the rule execution engine processes the REST interface request, inputs data to the rule, judges the occurrence probability of the fault mode and outputs the probability.
10. The hydro-power generating unit fault integrated diagnosis method according to claim 1, wherein: the big data model algorithm comprises the following steps: firstly, inputting a marked fault sample, then preprocessing data, and then training through a classification model; when the prediction is carried out, the data is subjected to the same preprocessing work, and then the result is output through the classification model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2023098701A JP7442001B1 (en) | 2022-11-17 | 2023-06-15 | Comprehensive failure diagnosis method for hydroelectric power generation units |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211440560.XA CN115718472A (en) | 2022-11-17 | 2022-11-17 | Fault scanning and diagnosing method for hydroelectric generating set |
| CN202211440560X | 2022-11-17 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116771576A true CN116771576A (en) | 2023-09-19 |
Family
ID=85255446
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211440560.XA Pending CN115718472A (en) | 2022-11-17 | 2022-11-17 | Fault scanning and diagnosing method for hydroelectric generating set |
| CN202310536536.4A Pending CN116771576A (en) | 2022-11-17 | 2023-05-12 | Comprehensive fault diagnosis method for hydroelectric generating set |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211440560.XA Pending CN115718472A (en) | 2022-11-17 | 2022-11-17 | Fault scanning and diagnosing method for hydroelectric generating set |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP7442001B1 (en) |
| CN (2) | CN115718472A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116992958A (en) * | 2023-09-27 | 2023-11-03 | 中国长江电力股份有限公司 | Method for automatically generating FTA real-time dynamic tree based on fault knowledge base |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115718472A (en) * | 2022-11-17 | 2023-02-28 | 中国长江电力股份有限公司 | Fault scanning and diagnosing method for hydroelectric generating set |
| CN117666546B (en) * | 2024-01-31 | 2024-05-03 | 中核武汉核电运行技术股份有限公司 | Distributed control system fault diagnosis method and device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944416A (en) * | 2012-12-06 | 2013-02-27 | 南京匹瑞电气科技有限公司 | Multi-sensor signal fusion technology-based fault diagnosis method for wind turbine blades |
| CN104133981A (en) * | 2014-06-25 | 2014-11-05 | 国家电网公司 | Photovoltaic power station fault diagnosis method based on fuzzy production rule knowledge base |
| CN109117353A (en) * | 2018-08-20 | 2019-01-01 | 中国石油大学(北京) | The fusion method and device of fault diagnosis result |
| CN110006645A (en) * | 2019-05-10 | 2019-07-12 | 北京航空航天大学 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method of multi-source fusion |
| CN110206676A (en) * | 2019-05-27 | 2019-09-06 | 浙江大学 | The hydraulic turbine remotely monitors, diagnosis prediction and intelligent maintenance regulator control system |
| CN110287510A (en) * | 2019-04-25 | 2019-09-27 | 四川中鼎智能技术有限公司 | Diagnosis control method, system, storage medium and terminal based on hydraulic turbine main valve aperture data correlation |
| US20210041863A1 (en) * | 2018-04-24 | 2021-02-11 | Hitachi, Ltd. | Fault Diagnosis Device, Fault Diagnosis Method and Machine to Which Fault Diagnosis Device Is Applied |
| CN115718472A (en) * | 2022-11-17 | 2023-02-28 | 中国长江电力股份有限公司 | Fault scanning and diagnosing method for hydroelectric generating set |
| CN115828173A (en) * | 2022-11-17 | 2023-03-21 | 华能射阳新能源发电有限公司 | Fan fault diagnosis method and system based on fault knowledge and causal reasoning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3068091B1 (en) | 1999-06-02 | 2000-07-24 | 核燃料サイクル開発機構 | Fault diagnosis device |
| JP6725253B2 (en) | 2016-01-29 | 2020-07-15 | 日立Ｇｅニュークリア・エナジー株式会社 | Plant monitoring device and program |
| JP6989398B2 (en) | 2017-03-24 | 2022-01-05 | 株式会社東芝 | Failure diagnostic device, failure diagnosis method, and failure diagnosis program |
-
2022
- 2022-11-17 CN CN202211440560.XA patent/CN115718472A/en active Pending
-
2023
- 2023-05-12 CN CN202310536536.4A patent/CN116771576A/en active Pending
- 2023-06-15 JP JP2023098701A patent/JP7442001B1/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944416A (en) * | 2012-12-06 | 2013-02-27 | 南京匹瑞电气科技有限公司 | Multi-sensor signal fusion technology-based fault diagnosis method for wind turbine blades |
| CN104133981A (en) * | 2014-06-25 | 2014-11-05 | 国家电网公司 | Photovoltaic power station fault diagnosis method based on fuzzy production rule knowledge base |
| US20210041863A1 (en) * | 2018-04-24 | 2021-02-11 | Hitachi, Ltd. | Fault Diagnosis Device, Fault Diagnosis Method and Machine to Which Fault Diagnosis Device Is Applied |
| CN109117353A (en) * | 2018-08-20 | 2019-01-01 | 中国石油大学(北京) | The fusion method and device of fault diagnosis result |
| CN110287510A (en) * | 2019-04-25 | 2019-09-27 | 四川中鼎智能技术有限公司 | Diagnosis control method, system, storage medium and terminal based on hydraulic turbine main valve aperture data correlation |
| CN110006645A (en) * | 2019-05-10 | 2019-07-12 | 北京航空航天大学 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method of multi-source fusion |
| CN110206676A (en) * | 2019-05-27 | 2019-09-06 | 浙江大学 | The hydraulic turbine remotely monitors, diagnosis prediction and intelligent maintenance regulator control system |
| CN115718472A (en) * | 2022-11-17 | 2023-02-28 | 中国长江电力股份有限公司 | Fault scanning and diagnosing method for hydroelectric generating set |
| CN115828173A (en) * | 2022-11-17 | 2023-03-21 | 华能射阳新能源发电有限公司 | Fan fault diagnosis method and system based on fault knowledge and causal reasoning |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116992958A (en) * | 2023-09-27 | 2023-11-03 | 中国长江电力股份有限公司 | Method for automatically generating FTA real-time dynamic tree based on fault knowledge base |
| CN116992958B (en) * | 2023-09-27 | 2024-03-12 | 中国长江电力股份有限公司 | Method for automatically generating FTA real-time dynamic tree based on fault knowledge base |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115718472A (en) | 2023-02-28 |
| JP7442001B1 (en) | 2024-03-01 |
| JP2024073353A (en) | 2024-05-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Literature review and prospect of the development and application of FMEA in manufacturing industry | |
| CN111985561B (en) | Fault diagnosis method and system for intelligent electric meter and electronic device | |
| CN110008288B (en) | Construction method and application of knowledge map library for network fault analysis | |
| CN114579875B (en) | Equipment fault diagnosis and maintenance knowledge recommendation system based on knowledge graph | |
| CN116771576A (en) | Comprehensive fault diagnosis method for hydroelectric generating set | |
| CN107272646B (en) | Press fault diagnosis system based on expert system | |
| CN101989087B (en) | On-line real-time failure monitoring and diagnosing system device for industrial processing of residual oil | |
| US10878324B2 (en) | Problem analysis and priority determination based on fuzzy expert systems | |
| CN106844194A (en) | A kind of construction method of multi-level software fault diagnosis expert system | |
| CN110296124B (en) | Remote fault diagnosis system and method based on expert system | |
| CN109785180A (en) | A kind of scene perception system and method towards the twin workshop of number | |
| CN106533754A (en) | Fault diagnosis method and expert system for college teaching servers | |
| CN113361663B (en) | Artificial intelligence-based power grid event diagnosis autonomous learning method and system | |
| CN111581739A (en) | Helicopter intelligent fault diagnosis method based on fault tree and case reasoning | |
| CN116932523B (en) | Platform for integrating and supervising third party environment detection mechanism | |
| CN112084220B (en) | Abnormality diagnosis method and device for electric energy metering device and readable storage medium | |
| CN113627784A (en) | Enterprise asset management intelligent decision-making system based on industrial internet | |
| CN110007662A (en) | The full aircraft health evaluating of self -support technical identification system and management ground Inference Platform | |
| CN117313845A (en) | Power grid fault detection auxiliary method and system based on knowledge graph | |
| CN112905845B (en) | Multi-source unstructured data cleaning method for discrete intelligent manufacturing application | |
| CN115526531A (en) | Wheat quality safety early warning method and system | |
| Yu et al. | Aircraft fault diagnosis system research based on the combination of CBR and FTA | |
| CN110275875B (en) | Method and apparatus for providing an instantiated industrial semantic model for an industrial infrastructure | |
| US20220342743A1 (en) | Recommendation system, configuration method therefor, and recommendation method | |
| CN114265891A (en) | Intelligent workshop system and method based on multi-source data fusion and storage medium |
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 |