CN109632315A - A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match - Google Patents
A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match Download PDFInfo
- Publication number
- CN109632315A CN109632315A CN201910028686.8A CN201910028686A CN109632315A CN 109632315 A CN109632315 A CN 109632315A CN 201910028686 A CN201910028686 A CN 201910028686A CN 109632315 A CN109632315 A CN 109632315A
- Authority
- CN
- China
- Prior art keywords
- diagnostic
- fault
- rule
- failure
- steam turbine
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M15/00—Testing of engines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The present invention relates to reliability maintenance field of engineering technology, especially a kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match.Existing judgement Turbo-generator Sets Faults method requires high and trouble to acquisition of information, and operability is poor.The present invention provides a kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match, including Failure Diagnostic Code library and Steam Turbine Fault Diagnosis module, Failure Diagnostic Code library includes that can be managed by way of definition, increase, deletion and modification to diagnostic rule data to the fault signature data and diagnostic rule data that diagnostic rule is managed, Failure Diagnostic Code library；Steam Turbine Fault Diagnosis module is directed to real time data or artificial input data, using the fault reasoning diagnostic method based on two-parameter rule match, triggers corresponding diagnostic rule work in Failure Diagnostic Code library and provides diagnosis.Two-parameter rule matching method is introduced fault diagnosis by the present invention, and fault diagnosis effect is good, accuracy is high.
Description
Technical field
The present invention relates to reliability maintenance field of engineering technology, especially a kind of steam turbine based on two-parameter rule match
Vibration fault reasoning diagnostic method.
Background technique
Steam turbine is the important equipment of power generation, and complicated structure causes vapour plus special work running environment synthesis
The failure rate of turbine is higher, some failures can seriously endanger steam turbine operation, therefore the fault diagnosis of Turbo-generator Set is very
It is important.With electric power industry development, the total installation of generating capacity of Turbo-generator Set and the installed capacity of unit unit have all been obtained rapidly
It improves, shaft system of unit also becomes increasingly complex, and the potential factor for inducing unit vibration is also increase accordingly.Vibration problem is in unit safety
Running to influence increasing, people increasingly pay close attention to unit vibration and stabilize the economy the influence of operation for production safety.
The failure form of expression of Steam Turbine is sufficiently complex, and fault diagnosis system gradually develops into guarantee unit equipment
The important means of safe and reliable operation, the diagnosis of existing Diagnosis for Turbogenerator Sets be partial to acquisition signal-based, point
Analysis and diagnosis, Fault Diagnosis Inference be concentrated mainly on neural network method using upper, to the knowledge based on empirical rule
It indicates and the research of inference method is less.It is the incomplete situation of information mostly in Vibration fault diagnosis of steam turbine generating reasoning process.
In the case, frequently with reasoning method under uncertainty.Having common reasoning method under uncertainty includes certain factor, mind
Through network method, Bayes method, evidence theory and possibility theory neural network etc..These methods in different fields all
It is widely used, but has limitation on solving the problems, such as uncertain inference, for example, the application method of rule is wanted
Ask, weight the limitation of reasoning problems, parameter of regularity to obtain aspect etc., acquisition of information requirement needed for specifically making inferences judgement
It is high and cumbersome, cause operability poor, fault diagnosis result accuracy is poor.
Summary of the invention
The technical problem to be solved in the present invention and the technical assignment of proposition are that existing reasoning method under uncertainty is overcome to judge
Turbo-generator Sets Faults limitation is big, high and cumbersome to acquisition of information requirement, and operability and diagnostic result accuracy are all
Poor, the present invention provides a kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match, passes through generation
The combination of formula rule and two-parameter method and the combination of fault signature obtain the higher fault diagnosis result of accuracy.
The present invention solves the technical solution that technical problem uses: a kind of Steam Turbine Vibration event based on two-parameter rule match
Hinder reasoning diagnostic method, including Failure Diagnostic Code library and Steam Turbine Fault Diagnosis module, the Failure Diagnostic Code library include
Can be to the fault signature data and diagnostic rule data that diagnostic rule is managed, and Failure Diagnostic Code library can be by fixed
Justice, the mode for increasing, deleting and modifying are managed diagnostic rule data；The Steam Turbine Fault Diagnosis module is for real-time
Data or artificial input data trigger Failure Diagnostic Code using the fault reasoning diagnostic method based on two-parameter rule match
Corresponding diagnostic rule work is in library to provide diagnosis.Two-parameter rule matching method is introduced Steam Turbine Vibration event by the present invention
In barrier diagnosis, and fault signature is combined to obtain more complete diagnostic rule；According to obtained real time data or manually
Input data, triggering diagnostic rule work provides diagnosis, higher for the accuracy of fault diagnosis.
As further improving and supplementing to above-mentioned technical proposal, the present invention uses following technical measures: described real-time
Data or the exception definition basic fault mode of artificial input data；It is examined using based on two-parameter production rule as failure
The expression-form for rule of breaking, the expression-form are to meet fault condition and then carry out respective operations, IF[fault condition] THEN[examines
Disconnected conclusion and operation], it is write based on two-parameter production rule: IF E1(i1,p1)AND E2(i2,p2)AND…AND En
(in,pn)
THEN H(λ)
The iiIt is sub- condition EiIntensity factor, indicate group condition EiIn the presence of corresponding evidence, to conclusion H at
Vertical adequacy degree of support；piIt is sub- condition EiDilution of precision, indicate group condition EiIt is right in the presence of corresponding evidence
The necessity degree of support that conclusion H is set up；iiAnd piThe value on [0,1]: i=1,2 ... n；λ is the threshold value of rule, in [0,1]
Upper value.iiAnd piValue by domain expert provide also can according to Related Computational Methods by calculate obtain；The threshold value of λ is by field
Expert provides.For the corresponding fault signature of same type Failure Diagnostic Code of steam turbine different parts, intensity and dilution of precision
According to the actual situation will be different, to handle the same type fault mode of different parts.
The fault signature data representation: each fault signature has unique ID number, each diagnostic rule
The ID of each diagnostic rule, is defined as the combination of its corresponding fault signature ID by corresponding multiple fault signatures, i.e. diagnosis rule
Then ID=fault signature 1ID+ fault signature 2ID+ fault signature 3ID+ ... advises each diagnosis as character string addition ID
Then all have unique ID, when adding diagnostic rule, the diagnostic rule data in Failure Diagnostic Code library are looked into first
It askes, sees with the presence or absence of this ID, be then added if it does not exist.There is the fault signature condition being overlapped herein for Different Rule collection
The case where, the present invention is integrated by the way of Composite Fault Feature, and different diagnostic rules is segmented in difference.
The step of fault reasoning diagnostic method based on two-parameter rule match, is as follows:
Step 1 calls in Failure Diagnostic Code library；
Step 2, according to real time data or artificial input data, obtain primary symptom, extract fault signature；
Step 3 carries out tentative diagnosis, obtains the combination by different faults feature according to the obtained fault set of diagnostic rule
It closes；
Step 4 evaluates adequacy between fault signature and diagnostic rule using intensity I and precision P and necessity is closed
System, computation model use parallel system model of reliability calculation, and calculation formula is as follows:
In above formula, provided that fault signature meet with diagnostic rule condition, then the c of the fault signaturei=1, i=1,
2 ... n, otherwise ci=0；The support that known fault feature sets up diagnostic rule conclusion is then indicated using synthetic reliability GC
Degree, calculation formula are as follows:
The failure collection is verified one by one, i.e., is made inferences according to fault signature, the synthesis for calculating each failure can
Reliability, and be ranked up from high to low based on synthetic reliability and provide diagnosis.Professional can be in the diagnosis knot provided
On the basis of fruit and each synthetic reliability information, then further judged.
Detailed description of the invention
Fig. 1: 5X and 6X vibrates variation tendency before and after climbing.
Vibration variation before and after Fig. 2: 5X and 6X climbs.
Anterioposterior curve occurs for the vibration of Fig. 3: China Tech TongAn TDM record.
Specific embodiment
The present invention is described further with specific embodiment for explanation with reference to the accompanying drawing.
A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match, including Failure Diagnostic Code library
With Steam Turbine Fault Diagnosis module, the Failure Diagnostic Code library includes can be to the fault signature data that diagnostic rule is managed
With diagnostic rule data, and Failure Diagnostic Code library can be by way of definition, increase, deletion and modification to diagnostic rule number
According to being managed；The Steam Turbine Fault Diagnosis module is directed to real time data or artificial input data, using based on two-parameter rule
Then matched fault reasoning diagnostic method triggers corresponding diagnostic rule in Failure Diagnostic Code library and works to provide diagnosis.
Two-parameter rule matching method is introduced into Turbine Vibrationfault Diagnosis by the present invention, and is combined fault signature to obtain
More complete diagnostic rule；According to obtained real time data or artificial input data, triggers diagnostic rule work and provide diagnosis knot
By higher for the accuracy of fault diagnosis.
As further improving and supplementing to above-mentioned technical proposal, the present invention uses following technical measures: described real-time
Data or the exception definition basic fault mode of artificial input data；It is examined using based on two-parameter production rule as failure
The expression-form for rule of breaking, the expression-form are to meet fault condition and then carry out respective operations, IF[fault condition] THEN[examines
Disconnected conclusion and operation], it is write based on two-parameter production rule: IF E1(i1,p1)AND E2(i2,p2)AND…AND En
(in,pn)
THEN H(λ)
The iiIt is sub- condition EiIntensity factor, indicate group condition EiIn the presence of corresponding evidence, to conclusion H at
Vertical adequacy degree of support；piIt is sub- condition EiDilution of precision, indicate group condition EiIt is right in the presence of corresponding evidence
The necessity degree of support that conclusion H is set up；iiAnd piThe value on [0,1]: i=1,2 ... n；λ is the threshold value of rule, in [0,1]
Upper value.iiAnd piValue by domain expert provide also can according to Related Computational Methods by calculate obtain；The threshold value of λ is by field
Expert provides.For the corresponding fault signature of same type Failure Diagnostic Code of steam turbine different parts, intensity and dilution of precision
According to the actual situation will be different, to handle the same type fault mode of different parts.
The fault signature data representation: each fault signature has unique ID number, each diagnostic rule
The ID of each diagnostic rule, is defined as the combination of its corresponding fault signature ID by corresponding multiple fault signatures, i.e. diagnosis rule
Then ID=fault signature 1ID+ fault signature 2ID+ fault signature 3ID+ ... advises each diagnosis as character string addition ID
Then all have unique ID, when adding diagnostic rule, the diagnostic rule data in Failure Diagnostic Code library are looked into first
It askes, sees with the presence or absence of this ID, be then added if it does not exist.There is the fault signature item being overlapped for Different Rule collection in the present invention
The case where part, is integrated by the way of Composite Fault Feature, and different diagnostic rules is segmented in difference.
The step of fault reasoning diagnostic method based on two-parameter rule match, is as follows:
Step 1 calls in Failure Diagnostic Code library；
Step 2, according to real time data or artificial input data, obtain primary symptom, extract fault signature；
Step 3 carries out tentative diagnosis, obtains the combination by different faults feature according to the obtained fault set of diagnostic rule
It closes；
Step 4 evaluates adequacy between fault signature and diagnostic rule using intensity I and precision P and necessity is closed
System, computation model use parallel system model of reliability calculation, and calculation formula is as follows:
In above formula, provided that fault signature meet with diagnostic rule condition, then the c of the fault signaturei=1, i=1,
2 ... n, otherwise ci=0；The support that known fault feature sets up diagnostic rule conclusion is then indicated using synthetic reliability GC
Degree, calculation formula are as follows:
The failure collection is verified one by one, i.e., is made inferences according to fault signature, the synthesis for calculating each failure can
Reliability, and be ranked up from high to low based on synthetic reliability and provide diagnosis.Further, professional can provide
Diagnostic result and each synthetic reliability information on the basis of, then make a decision.
Fault case one is No. 7 arbors vibration data of certain power plant.12:25 points, No. 7 machine red switch of certain power plant to rated speed, machine
Each axis vibration of group is respectively less than 65 μm.After reaching rated speed, the axis vibration of low pressure (LP) cylinder two sides No. 5 and No. 6 starts to climb, 13:06 points of vibrations
It climbs to 99 μm, opens beginning vibration for 13:39 points and begin to decline, wherein it is as shown in Figure 1 to vibrate variation tendency by 5X.Before and after 5X and 6X climbs
It is as shown in Figure 2 to vibrate trend chart.From 12:25 assign to 13:09 split axle vibration 5X and 6X vibration climb during, vibrate with one times
Based on frequency, amplitude slowly increases 60 μm, and phase also correspondingly increases, and 5X and 6X vibration diverse vector are respectively 104 ∠, 200 He
95 ∠ 18, diverse vector is based on reversed component.13:09 assigns to 13:39 points, and vibration amplitude remains unchanged substantially, and phase slightly has
Increase.13:39 separately begins, and vibration is begun to decline, and amplitude is reduced, and phase is slightly reduced.
Fault case two is certain unit vibration monitoring data.Fig. 3 is that the vibration of the China Tech TongAn TDM record of Installation in Plant is bent
Line, four curves are that 5 watts of axis X that shake to peak-to-peak value, 5 watts of axis X that shake to a frequency multiplication, 5 watts of axis shake x to phase respectively from top to bottom in figure
Position, unit load.It can be seen that from Fig. 3-1,5X vibrates the largest of about 115 μm of peak-to-peak value, and vibration, which increases, the process slowly climbed,
About 1 hour of time-to-climb, and before current vibration is climbed, actually there are the lesser process of climbing of amplitude three times, width
Value variation is smaller.
Specific fault reasoning diagnosis process is as follows:
Step 1: calling in diagnosis rule base；
Step 2: according to real time data or artificial input data, obtaining primary symptom, extract fault signature；For failure case
Example one extracts following vibration performance:
1) 5X and 6X vibration climbs process based on a harmonic, vibration passband value increase when, the amplitude of a frequency multiplication and
Phase increases；
2) amplitude is close during 5X and 6X vibration is climbed, opposite in phase；From table 2-1 vibration variation delta 2. -1. find out,
During vibration is climbed, 5X and 6X vibration have changed 104 ∠ of ∠ 200 and 95 18 respectively, and vibration amplitude variable quantity is close, phase
182 degree of variation difference illustrates that two side axle vibrational of low pressure (LP) cylinder is past phase negative side during the vibration is climbed based on reversed component
To variation；
3) vibration amplitude is close during 5X and 6X vibration is reduced, opposite in phase；In Fig. 2 vibrate variation delta 4. -3., vibration
Dynamic amplitude is close, opposite in phase；
For fault case two, following fault signature is extracted:
1) vibration variation is based on a frequency multiplication, and one frequency multiplication increases when vibration increases；
2) when vibration increases, vibration phase increases；
3) occurred repeatedly to vibrate the process of climbing, preceding amplitude of climbing several times is smaller, and vibration is climbed larger for the last time, several times
Vibration amplitude is identical with phase change trend during climbing；
4) vibration increases unobvious with unit load correlation, and when vibration is climbed, unit load has sometimes in high load capacity
When be in underload, without corresponding relationship；
Step 3: carrying out tentative diagnosis, obtain the combination by different faults feature according to the obtained fault set of diagnostic rule
It closes；According to 5X and 6X vibration performance, 5X and 6X vibration are no matter during climbing or during decline, the vibration of 5X and 6X
It is all to change toward opposite direction, variation amplitude is close, thinks that the low pressure (LP) cylinder rotor oscillation process of climbing is due to low pressure (LP) cylinder
Rotor and cylinder touch mill；
Step 4: the adequacy and necessity evaluated between fault signature and diagnostic rule using intensity I and precision P are closed
System, computational physics model use parallel system model of reliability calculation, and calculation formula is as follows:
Then, the degree of support that known fault feature sets up diagnostic rule conclusion is indicated using synthetic reliability GC,
Its calculation formula is using as follows:
Possible failure collection is verified one by one, i.e., is made inferences according to fault signature, the synthesis of each failure is calculated
Confidence level simultaneously sorts, and provides diagnosis.
The diagnostic rule content for calculating acquisition fault case one according to above formula is as follows:
The diagnostic rule content for calculating acquisition fault case two according to above formula is as follows:
It was accordingly found that the failure cause of two cases is the same, however the fault signature found in practice is but and different
Sample is integrated by the way of Composite Fault Feature when laying down a regulation under such circumstances.Diagnosis rule after being merged
Then content is as follows:
The present invention can carry out Turbine Vibrationfault Diagnosis, and be able to achieve the management of diagnostic rule.
Claims (4)
1. a kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match, including Failure Diagnostic Code library and
Steam Turbine Fault Diagnosis module, the Failure Diagnostic Code library include can to fault signature data that diagnostic rule is managed and
Diagnostic rule data, and Failure Diagnostic Code library can be by way of definition, increase, deletion and modification to diagnostic rule data
It is managed；The Steam Turbine Fault Diagnosis module is directed to real time data or artificial input data, using based on two-parameter rule
Matched fault reasoning diagnostic method triggers corresponding diagnostic rule in Failure Diagnostic Code library and works to provide diagnosis.
2. the Steam Turbine Vibration fault reasoning diagnostic method according to claim 1 based on two-parameter rule match, special
Sign is the real time data or the exception definition basic fault mode of artificial input data；Using based on two-parameter production rule
Then as the expression-form of Failure Diagnostic Code, the expression-form is to meet fault condition then to carry out respective operations, IF[failure
Condition] THEN[diagnosis and operation], it is write based on two-parameter production rule:
IF E1(i1, p1)AND E2(i2, p2)AND...AND En(in, pn)
THEN H(λ)
The iiIt is sub- condition EiIntensity factor, indicate group condition EiIn the presence of corresponding evidence, conclusion H is set up
Adequacy degree of support；piIt is sub- condition EiDilution of precision, indicate group condition EiIn the presence of corresponding evidence, to conclusion H
The necessity degree of support of establishment；iiAnd piThe value on [0,1]: i=1,2 ... n；λ is the threshold value of rule, is taken on [0,1]
Value.
3. the Steam Turbine Vibration fault reasoning diagnostic method according to claim 2 based on two-parameter rule match, special
Sign is the fault signature data representation: each fault signature has unique ID number, each diagnostic rule is corresponding
The ID of each diagnostic rule is defined as the combination of its corresponding fault signature ID, i.e. diagnostic rule ID by multiple fault signatures
ID is made each diagnostic rule as character string addition by=fault signature 1ID+ fault signature 2ID+ fault signature 3ID+ ...
Have unique ID, when adding diagnostic rule, the diagnostic rule data in Failure Diagnostic Code library is inquired first, are seen
With the presence or absence of this ID, then it is added if it does not exist.
4. the Steam Turbine Vibration fault reasoning diagnostic method according to claim 2 based on two-parameter rule match, special
The step of sign is the fault reasoning diagnostic method based on two-parameter rule match is as follows:
Step 1 calls in Failure Diagnostic Code library；
Step 2, according to real time data or artificial input data, obtain primary symptom, extract fault signature；
Step 3 carries out tentative diagnosis, obtains the combination by different faults feature according to the obtained failure collection of diagnostic rule；
Step 4 evaluates adequacy between fault signature and diagnostic rule and necessary sexual intercourse, meter using intensity I and precision P
It calculates model and uses parallel system model of reliability calculation, calculation formula is as follows:
In above formula, provided that fault signature meet with diagnostic rule condition, then the c of the fault signaturei=1, i=1,2 ... n,
Otherwise ci=0；The degree of support that known fault feature sets up diagnostic rule conclusion is then indicated using synthetic reliability GC,
Its calculation formula is as follows:
The failure collection is verified one by one, i.e., is made inferences according to fault signature, the synthetic reliability of each failure is calculated,
And it is ranked up from high to low based on synthetic reliability and provides diagnosis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910028686.8A CN109632315A (en) | 2019-01-11 | 2019-01-11 | A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910028686.8A CN109632315A (en) | 2019-01-11 | 2019-01-11 | A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109632315A true CN109632315A (en) | 2019-04-16 |
Family
ID=66061875
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910028686.8A Withdrawn CN109632315A (en) | 2019-01-11 | 2019-01-11 | A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109632315A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110162016A (en) * | 2019-05-30 | 2019-08-23 | 华北电力大学 | A kind of fault modeling method of gas turbine pneumatic actuator |
| CN110794815A (en) * | 2019-11-29 | 2020-02-14 | 哈尔滨汽轮机厂有限责任公司 | Remote intelligent operation and maintenance service system for steam turbine |
| CN111177485A (en) * | 2019-12-16 | 2020-05-19 | 中建材智慧工业科技有限公司 | Method, device and medium for predicting device failure based on parameter rule matching |
| CN112541155A (en) * | 2020-09-30 | 2021-03-23 | 上海电力大学 | Steam turbine rotor unbalance detection method based on confidence rule base |
| CN113685235A (en) * | 2021-08-23 | 2021-11-23 | 华能威海发电有限责任公司 | Fault determination method and system for steam turbine generator unit |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003344150A (en) * | 2002-05-24 | 2003-12-03 | Kawasaki Heavy Ind Ltd | Condition diagnosing method and apparatus therefor by analyzing shaft vibration of rotating machine |
| CN1514209A (en) * | 2003-08-01 | 2004-07-21 | 重庆大学 | Rotary machine failure intelligent diagnosis method and device |
| US20050171736A1 (en) * | 2004-02-02 | 2005-08-04 | United Technologies Corporation | Health monitoring and diagnostic/prognostic system for an ORC plant |
| CN104503434A (en) * | 2014-12-01 | 2015-04-08 | 北京航天试验技术研究所 | Fault diagnosis method based on active fault symptom pushing |
| CN108627345A (en) * | 2018-05-11 | 2018-10-09 | 浙江师范大学 | A kind of diagnostic method and system of turbine system grade failure |
-
2019
- 2019-01-11 CN CN201910028686.8A patent/CN109632315A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003344150A (en) * | 2002-05-24 | 2003-12-03 | Kawasaki Heavy Ind Ltd | Condition diagnosing method and apparatus therefor by analyzing shaft vibration of rotating machine |
| CN1514209A (en) * | 2003-08-01 | 2004-07-21 | 重庆大学 | Rotary machine failure intelligent diagnosis method and device |
| US20050171736A1 (en) * | 2004-02-02 | 2005-08-04 | United Technologies Corporation | Health monitoring and diagnostic/prognostic system for an ORC plant |
| CN104503434A (en) * | 2014-12-01 | 2015-04-08 | 北京航天试验技术研究所 | Fault diagnosis method based on active fault symptom pushing |
| CN108627345A (en) * | 2018-05-11 | 2018-10-09 | 浙江师范大学 | A kind of diagnostic method and system of turbine system grade failure |
Non-Patent Citations (2)
| Title |
|---|
| 宋光雄 等: "基于双参数方法的诊断规则不确定性问题研究", 《中国机械工程》 * |
| 田增华: "汽轮机组振动故障诊断及专家系统研究", 《万方》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110162016A (en) * | 2019-05-30 | 2019-08-23 | 华北电力大学 | A kind of fault modeling method of gas turbine pneumatic actuator |
| CN110162016B (en) * | 2019-05-30 | 2020-10-16 | 华北电力大学 | Fault modeling method for pneumatic actuator of gas turbine |
| CN110794815A (en) * | 2019-11-29 | 2020-02-14 | 哈尔滨汽轮机厂有限责任公司 | Remote intelligent operation and maintenance service system for steam turbine |
| CN111177485A (en) * | 2019-12-16 | 2020-05-19 | 中建材智慧工业科技有限公司 | Method, device and medium for predicting device failure based on parameter rule matching |
| CN112541155A (en) * | 2020-09-30 | 2021-03-23 | 上海电力大学 | Steam turbine rotor unbalance detection method based on confidence rule base |
| CN113685235A (en) * | 2021-08-23 | 2021-11-23 | 华能威海发电有限责任公司 | Fault determination method and system for steam turbine generator unit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109632315A (en) | A kind of Steam Turbine Vibration fault reasoning diagnostic method based on two-parameter rule match | |
| Yang et al. | VIBEX: an expert system for vibration fault diagnosis of rotating machinery using decision tree and decision table | |
| Yang et al. | Integration of ART-Kohonen neural network and case-based reasoning for intelligent fault diagnosis | |
| Geeganage et al. | Application of energy-based power system features for dynamic security assessment | |
| CN102288283B (en) | Method and device for monitoring vibration faults of turbo generator set | |
| CN106610617B (en) | A kind of accumulation of energy unit trouble analysis system and method | |
| Verma et al. | Wavelet based fault detection and diagnosis using online mcsa of stator winding faults due to insulation failure in industrial induction machine | |
| Wang | Application of extension theory to vibration fault diagnosis of generator sets | |
| Wang et al. | Quantitative risk assessment through hybrid causal logic approach | |
| CN110617981B (en) | Fault diagnosis method for phase modulator | |
| CN106596110B (en) | The automatic analyzing and diagnosing method of turbine-generator units waterpower imbalance fault based on online data | |
| Hara et al. | Fault detection of hydroelectric generators using isolation forest | |
| Deng et al. | Rule-based fault diagnosis expert system for wind turbine | |
| Todd et al. | A semiautomatic approach to deriving turbine generator diagnostic knowledge | |
| Jiarula et al. | Fault mode prediction based on decision tree | |
| Naveena et al. | Fault diagnosis of VSI fed induction motor drive using fuzzy logic approach | |
| Al-Momani et al. | Prediction of Transient Stability Using Wide Area Measurements Based on ANN | |
| Yang et al. | Reinforcing fuzzy rule-based diagnosis of turbomachines with case-based reasoning | |
| CN107066662A (en) | The diagnostic method of turbine-generator units quality imbalance fault based on online data | |
| Al-Gubri et al. | Performance analysis of machine learning algorithms for power system dynamic security assessment | |
| Marra et al. | Data Augmentation for Vibration Signals using System Identification Techniques | |
| Giri et al. | Optimized Fault Detection and Classification for 3-Phase Electric Drives–An AI/ML Approach | |
| Bayar et al. | PCA-ANN based algorithm for the determination of asymmetrical network failures of network-connected induction generators | |
| Yang et al. | Fuzzy diagnosis of turbomachines | |
| Saleem | Enhancement of power system transient stability using machine learning and MPC |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190416 |