CN102928231A - Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory - Google Patents
Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory Download PDFInfo
- Publication number
- CN102928231A CN102928231A CN2012104511399A CN201210451139A CN102928231A CN 102928231 A CN102928231 A CN 102928231A CN 2012104511399 A CN2012104511399 A CN 2012104511399A CN 201210451139 A CN201210451139 A CN 201210451139A CN 102928231 A CN102928231 A CN 102928231A
- Authority
- CN
- China
- Prior art keywords
- fault
- diagnostic
- equipment
- fault diagnosis
- diagnosis
- 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
Abstract
The invention relates to an equipment fault diagnosis method based on a D-S (Dempster-Shafer) evidence theory. The equipment fault diagnosis method comprises the steps as follows: firstly establishing a fault feature set; then, carrying out independent fault diagnosis on equipment by using various diagnosis methods; and finally, integrating the diagnosis results of the methods based on the D-S evidence theory to obtain a final fault diagnosis result. The method provides a strategy integrating the diagnosis results of the methods based on the existing diagnosis method, and has the characteristic of easiness in implementation while the accuracy of the diagnosis results is improved. According to the method, a diagnosis module can be independently written and embedded into equipment fault diagnosis software without adjusting the software structure. The method can be widely applied to equipment fault diagnosis in the fields of electric power, machines, ships and the like in addition to application to turbo generator sets, and an accurate, high-efficiency and low-cost method is provided for equipment fault diagnosis. The purposes of improving equipment running reliability and reducing equipment repair expense can be achieved through accurate judgment of equipment faults.
Description
Technical field
The present invention relates to a kind of equipment fault diagnosis, particularly a kind of equipment fault diagnosis method based on the D-S evidence theory.
Background technology
Since 20th century, along with the high speed development of science and technology, commercial production etc., the availability of plant equipment, reliability, the problem of security become increasingly conspicuous, and have promoted the research of people to technology for mechanical fault diagnosis.
Very early time, diagnostic techniques mainly depend on the status information that this field simple dependence sense organ of individual expert obtains equipment, by virtue of experience judge.Diagnosis Technique truly is a new technology that just grows up after the 1950's, and it is along with computing machine is processed at industrial signal, application aspect the Industry Control and development and improvement progressively.At the initial stage of Diagnosis Technique development, the correlation parameter when fault occurs the expert is recorded, and the mode that is combined into the diagnostic knowledge table is come analysis of failure.Development along with diagnostic techniques, expert system, fuzzy theory, gray theory and neural network theory, progressively go to hold the fault diagnosis rule from shallow to deep, the automatization level of diagnostic techniques is improved constantly, accuracy and the degree of confidence of diagnostic result also improve constantly.
The Data fusion technique that develops rapidly at present is a kind of Intelligent Diagnosis Technology, it has the advantage of utilizing the redundancy that comprises between each data source and complementary information, can improve accuracy and the robustness of the system decision-making, thereby provide an effective approach for the raising of equipment fault diagnosis rate.
Steam turbine is one group of visual plant in the industrial circle, especially in power industry.Yet the complicacy of Steam Turbine structure and system, singularity and running environment, Steam Turbine have failure rate higher, the shortcomings such as the dangerous height of fault.Therefore, the fault diagnosis of Steam Turbine is the importance that the fault diagnostic techniques is used always.
Yet the mapping relations between steam turbine failure symptom and the fault are complicated, fault gradual and sudden, and general data digging method can not satisfy the fault of diagnosis steam turbine.The change of many conditions, too high or too low such as vapor (steam) temperature, all will bring adverse effect to safety in production.Current traditional method for diagnosing faults is just finished by some simple data criterions, to come with this traditional method for diagnosing faults the deficiency of turbine system with regard to showing of diagnosis of complex, therefore need a kind of new method to come steam turbine is carried out reliable fault diagnosis.
Summary of the invention
The mapping relations that the present invention be directed between steam turbine failure symptom and the fault are complicated, existing diagnostic method limitation can't the efficient diagnosis fault problem, a kind of equipment fault diagnosis method based on the D-S evidence theory has been proposed, utilize the D-S evidence theory can process the ability of uncertain information, now having on the basis of diagnostic method, propose a kind of method of comprehensive various diagnostic results, and drawn thus diagnosis.Can improve the accuracy rate of equipment fault diagnosis.
Technical scheme of the present invention is: a kind of equipment fault diagnosis method based on the D-S evidence theory comprises following concrete steps:
1) foundation of fault signature collection: equipment is carried out fault diagnosis, at first will be by the analysis of fault type and failure symptom, the fault signature collection of apparatus for establishing;
2) fault diagnosis of distinct methods: after having set up the failure symptom collection, adopt different diagnostic methods that equipment is carried out independently fault diagnosis, the diagnostic method that can adopt comprises the diagnostic method of various neural networks, based on the method for gray theory, based on the method for rough set and support vector machine;
3) judge based on the diagnostic result of D-S evidence theory: after using distinct methods equipment failure is diagnosed, the diagnostic result of each method is carried out comprehensively, step is as follows:
A: according to the result of different diagnostic methods, the basic probability assignment function of structure D-S evidence theory, building method as the following formula shown in:
Wherein
iExpression the
iPlant diagnostic method,
jThe
jPlant fault type,
C i (j)Expression uses the
iKind of diagnostic method is for the
jPlant the diagnostic result that fault type is made,
R (i)The reliability coefficient that represents i kind diagnostic method,
R (i)Can obtain by calculating independent accuracy when carrying out fault diagnosis with the method;
B: after having obtained the basic probability assignment function of different diagnostic methods for the different faults type, the diagnostic result that application D-S fusion rule obtains the whole bag of tricks carries out comprehensively;
C: after calculating the comprehensive diagnos result of various diagnostic methods, the criterion during according to concrete use can obtain the net result of fault diagnosis.
The criterion that in the described step 3) diagnostic result of each method is carried out among the comprehensive step C can be:
The output valve of fault is the maximal value in all fault output valves; The output valve of fault is greater than predefined threshold value; The difference range of regulation and other each fault output valve.
Beneficial effect of the present invention is: the equipment fault diagnosis method that the present invention is based on the D-S evidence theory, on the basis of existing diagnostic method, propose the strategy of comprehensive the whole bag of tricks diagnostic result, when improving the diagnostic result accuracy, had the simple distinguishing feature of realization.Can write separately diagnostic module according to the method, embed existing equipment fault diagnosis software, and do not need software configuration is adjusted.The method also can be widely used in the middle of the equipment fault diagnosis in each field such as electric power, machinery, boats and ships, for equipment fault diagnosis provides cheaply method of a kind of precise and high efficiency except being applied to Turbo-generator Set.By the Accurate Diagnosis to equipment failure, can reach the purpose of raising equipment operational reliability, reduction maintenance of equipment expense.
Embodiment
The present invention mainly take based on the Fault Diagnosis Strategy of D-S evidence theory as core.The reason that Turbo-generator Sets Faults occurs is many-sided, is subject to the impact of various factors as the reliability of the sensor of transfer data information, and all there is certain limitation in various diagnostic method, and this just causes diagnostic result not accurate enough.In order to improve the accuracy of diagnostic result, can improve from signal extraction mode, method for diagnosing faults equal angles.
One, the foundation of fault signature collection:
Fault diagnosis is to come the judgment device state by the relation between research fault and the sign (characteristic element).For a concrete fault type, as long as the problem of care has two aspects: one is that this fault by which physical parameter shows; Another is the strong and weak situation of the relation of each physical parameter.Only have those and fault type in close relations, could be used for fault diagnosis to the sensitive reliably physical parameter of fault.At field of diagnosis about equipment fault, these are sensitive to fault, reliable and stable physical parameter becomes Failure Characteristic Parameter, becomes again fault characteristic signals.So equipment is carried out fault diagnosis, at first will be by the analysis of fault type and failure symptom, the fault signature collection of apparatus for establishing.
Two, the fault diagnosis of distinct methods:
After having set up the failure symptom collection, can adopt different diagnostic methods to carry out fault diagnosis.In order to adopt the D-S evidence theory that the diagnostic result of the whole bag of tricks is merged, need to guarantee that the whole bag of tricks is independently equipment to be diagnosed.The diagnostic method that can adopt here comprises the diagnostic method of various neural networks, based on the method for gray theory, based on method of rough set and support vector machine etc.
Three, judge based on the diagnostic result of D-S evidence theory:
After using distinct methods equipment failure is diagnosed, can adopt following steps that the diagnostic result of each method is carried out comprehensively, and provide result of determination:
Step 1: according to the result of different diagnostic methods, the basic probability assignment function of structure D-S evidence theory.Building method is as shown in Equation (1):
Here,
iExpression the
iPlant diagnostic method,
jThe
jPlant fault type,
C i (j)Expression uses the
iKind of diagnostic method is for the
jPlant the diagnostic result that fault type is made,
R (i)The reliability coefficient that represents i kind diagnostic method,
R (i)Can obtain by calculating independent accuracy when carrying out fault diagnosis with the method.
Step 2: after having obtained the basic probability assignment function of different diagnostic methods for the different faults type, can use the D-S fusion rule diagnostic result that the whole bag of tricks obtains is carried out comprehensively.
Step 3: after calculating the comprehensive diagnos result of various diagnostic methods, the criterion during according to concrete use can obtain the net result of fault diagnosis.
For the equipment characteristic of condenser, in conjunction with expertise and pertinent literature, 14 kinds of fault types have been chosen and 16 kinds of failure symptoms are set up the fault signature collection of this method research.Here, x1-x16 is 16 selected in fault diagnosis of condenser failure symptoms, and what Y1-Y14 represented is 14 fault types of condenser, and particular content is as follows:
X1 one condenser absolute pressure; X2 one circulation-water pump electric machine electric current; X3 one water circulating pump intake pressure; X4 one condenser water resistance .x5 one condensate pump top hole pressure; X6 one condensate pump current of electric; X7 one condensate pump conductivity; X8 one cooling water temperature rise; X9 one condenser terminal difference; XlO one condensate undercooling; Air and the coolant outlet temperature difference that X11 one air ejector is extracted out; Xl2 one condenser bleeding point is to air ejector import pressure reduction; Xl3 one condenser water level; Xl4 one vaccum pump motor electric current; The x15 one low water level that adds; X16 one rotor differential expansion.
Catastrophic failure appears in Y1 one water circulating pump; Y2 one rear shaft seal steam supply interrupts suddenly; Y3 one condenser full water; Y4 one vacuum system pipe breakage; Y5 one condenser outlet hydroecium is free air cooling, and but tube sheet is dirty; Y6 one condensate pump is undesired; Y7 one afterbody hangs down and adds pipeline breaking; Y8 one Cooling Tubes of Condenser breaks; Y9 one cooling tube stops up; Y1O one Cooling Tubes of Condenser is dirty; Y11 one aspiration pump work is undesired; Y12 one vacuum system imprecision; Y13 one quantity of circulating water is not enough; Y14 one water-jet pump catastrophic failure.
In order to test this method, selected the fault of 40 kinds of known fault types as test data here.The data that three groups of data failure symptoms as shown in table 1 are wherein concentrated.
Table 1
| Failure symptom | Test data 1 | Test data 2 | Test data 3 |
| X1 | 0.75 | 1.00 | 0.75 |
| X2 | 0.56 | 0.50 | 0.50 |
| X3 | 0.50 | 0.45 | 0.50 |
| X4 | 0.75 | 0.50 | 0.50 |
| X5 | 0.50 | 0.50 | 0.50 |
| X6 | 0.50 | 0.55 | 0.50 |
| X7 | 0.55 | 0.50 | 0.55 |
| X8 | 0.75 | 0.25 | 0.75 |
| X9 | 0.75 | 0.75 | 0.75 |
| X10 | 0.50 | 0.70 | 0.50 |
| X11 | 0.70 | 0.50 | 0.75 |
| X12 | 0.50 | 0.48 | 0.45 |
| X13 | 0.50 | 0.50 | 0.50 |
| X14 | 0.50 | 0.50 | 0.50 |
| X15 | 0.50 | 0.53 | 0.50 |
| X16 | 0.50 | 0.55 | 0.50 |
| Fault type | Cooling tube stops up | The vacuum system pipe breakage | Quantity of circulating water is not enough |
Adopt respectively the method for BP neural network and RBF neural network to carry out fault diagnosis for the data that provide, the diagnostic result of above three groups of data is shown in table 2-table 4, and the fault diagnosis of condenser that is respectively distinct methods is test data 1,2,3 as a result.
In table 2, two kinds of different diagnostic methods have been provided for the diagnostic result of test data 1.Wherein, using the method for BP neural network, be 0.36 for the output valve of Y9 fault, and use the method for RBF neural network, is 0.20 for the output valve of Y9 fault.Although all be the maximal value in each fault type output valve.But gap between the two is larger.For the RBF neural network, more approaching between each output valve especially.Thereby can't be out of order and be definite conclusion of Y9.In table 3, two kinds of different diagnostic methods have been provided for the diagnostic result of test data 2.Two kinds of algorithms all are maximum in each output valve for the output valve of Y4 fault, but maximal value wherein only is 0.25, and not obvious output valve greater than other fault.Although can be out of order like this for the conclusion of Y4 but not convincing.Equally, use respectively the BP neural network and the RBF neural network is diagnosed for test data 3, the result is as shown in table 4.In table 4, two kinds of diverse ways for fault be the output of Y14 all near 0.36, and obviously greater than the output of other each fault, be the conclusion of Y14 so can be out of order.Data 3 are mainly used in test, when additive method can draw correct conclusion, use method of the present invention and whether can draw correct diagnostic result.
Next, application of formula (1) obtains distinct methods for the elementary probability difference function of different faults type.Utilize separately the method for BP neural network and RBF neural network to test for 40 groups of data, wherein the correct diagnosis of BP neural network is 32 groups, 35 groups of the correct diagnosis of RBF neural network.Here with 1 method that represents the BP neural network, with 2 methods that represent the RBF neural network.So,＜i TranNum=" 230 "〉and R (1)=0.800, R (2)=0.875.</i>
Corresponding to table 2-table 4, can obtain table 5 distinct methods for the basic probability assignment function of test data 1 like this, table 6 distinct methods is for the basic probability assignment function of test data 2, and table 7 distinct methods is for the basic probability assignment function of test data 3.
After drawing table 5-table 7, can merge according to the D-S fusion rule respectively, through the diagnostic result fault diagnosis result based on the D-S evidence theory as shown in table 8 after merging.
In test data 1, be 0.4288 for the output valve of fault Y9; In test data 2, be 0.3593 for the output valve of fault Y4; In test data 3, be 0.5487 for the output valve of fault Y13.The output valve that provides when by above result as can be known, the output valve of these several faults is with a kind of method of independent employing all is significantly improved.
At last, use following rule for the diagnostic result that draws, carry out the judgement of fault type:
(1) output valve of this fault is the maximal value in all fault output valves;
(2) output valve of this fault is greater than 0.3;
(3) and the difference of other each fault output valve greater than 0.2.
Thus, can determine to judge that fault type is Y9 in test data 1, in test data 2, fault type is Y4, and in test data 3, fault type is Y13.As seen use the well uncertainty in the abatement apparatus fault diagnosis of method of the present invention, improve the accuracy of diagnostic result.
Claims (2)
1. the equipment fault diagnosis method based on the D-S evidence theory is characterized in that, comprises following concrete steps:
1) foundation of fault signature collection: equipment is carried out fault diagnosis, at first will be by the analysis of fault type and failure symptom, the fault signature collection of apparatus for establishing;
2) fault diagnosis of distinct methods: after having set up the failure symptom collection, adopt different diagnostic methods that equipment is carried out independently fault diagnosis, the diagnostic method that can adopt comprises the diagnostic method of various neural networks, based on the method for gray theory, based on the method for rough set and support vector machine;
3) judge based on the diagnostic result of D-S evidence theory: after using distinct methods equipment failure is diagnosed, the diagnostic result of each method is carried out comprehensively, step is as follows:
A: according to the result of different diagnostic methods, the basic probability assignment function of structure D-S evidence theory, building method as the following formula shown in:
Wherein
iExpression the
iPlant diagnostic method,
jThe
jPlant fault type,
C i (j)Expression uses the
iKind of diagnostic method is for the
jPlant the diagnostic result that fault type is made,
R (i)The reliability coefficient that represents i kind diagnostic method,
R (i)Can obtain by calculating independent accuracy when carrying out fault diagnosis with the method;
B: after having obtained the basic probability assignment function of different diagnostic methods for the different faults type, the diagnostic result that application D-S fusion rule obtains the whole bag of tricks carries out comprehensively;
C: after calculating the comprehensive diagnos result of various diagnostic methods, the criterion during according to concrete use obtains the net result of fault diagnosis.
2. described equipment fault diagnosis method based on the D-S evidence theory according to claim 1 is characterized in that, the criterion that in the described step 3) diagnostic result of each method is carried out among the comprehensive step C can be:
The output valve of fault is the maximal value in all fault output valves; The output valve of fault is greater than the threshold value that presets; The difference range of regulation and other each fault output valve.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012104511399A CN102928231A (en) | 2012-11-13 | 2012-11-13 | Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012104511399A CN102928231A (en) | 2012-11-13 | 2012-11-13 | Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102928231A true CN102928231A (en) | 2013-02-13 |
Family
ID=47643084
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012104511399A Pending CN102928231A (en) | 2012-11-13 | 2012-11-13 | Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102928231A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104457903A (en) * | 2014-12-31 | 2015-03-25 | 上海电力学院 | Method for protecting water level of boiler vapor drum |
| CN105758450A (en) * | 2015-12-23 | 2016-07-13 | 西安石油大学 | Fire protection pre-warning sensing system building method based on multiple sensor emergency robots |
| CN105899929A (en) * | 2013-11-05 | 2016-08-24 | 斯内克马公司 | Technical testing method |
| CN106154182A (en) * | 2016-08-26 | 2016-11-23 | 上海电力学院 | A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory |
| CN106447040A (en) * | 2016-09-30 | 2017-02-22 | 湖南科技大学 | Method for evaluating the health state of mechanical equipment based on heterogeneous multi-sensor data fusion |
| CN106503643A (en) * | 2016-10-18 | 2017-03-15 | 上海电力学院 | Tumble detection method for human body |
| CN107274112A (en) * | 2017-07-07 | 2017-10-20 | 许寅卿 | Improve the diagnosis algorithm model of oil dissolved gas |
| CN107545339A (en) * | 2016-06-27 | 2018-01-05 | 南京理工大学 | The wind power generating set on-line fault diagnosis method of DS evidence theories based on SCADA alarm signals |
| CN107729920A (en) * | 2017-09-18 | 2018-02-23 | 江苏海事职业技术学院 | A kind of method for estimating state combined based on BP neural network with D S evidence theories |
| CN108761263A (en) * | 2018-05-24 | 2018-11-06 | 深圳大图科创技术开发有限公司 | A kind of fault diagnosis system based on evidence theory |
| CN109060398A (en) * | 2018-09-11 | 2018-12-21 | 上海电力学院 | A kind of multi-source information equipment fault diagnosis method |
| WO2019184557A1 (en) * | 2018-03-29 | 2019-10-03 | 华为技术有限公司 | Method and device for locating root cause alarm, and computer-readable storage medium |
| CN111723341A (en) * | 2020-06-15 | 2020-09-29 | 中国船舶重工集团公司第七0三研究所 | Multi-subset probability processing method for boiler fault diagnosis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034389A (en) * | 2007-03-19 | 2007-09-12 | 江西省电力科学研究院 | Electrical energy power quality disturbance automatic identification method and system based on information fusion |
| CN102589890A (en) * | 2012-03-01 | 2012-07-18 | 上海电力学院 | Integrated fault diagnostic method of steam turbine based on CPN (counter-propagation network) and D-S (dempster-shafer) evidences |
-
2012
- 2012-11-13 CN CN2012104511399A patent/CN102928231A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034389A (en) * | 2007-03-19 | 2007-09-12 | 江西省电力科学研究院 | Electrical energy power quality disturbance automatic identification method and system based on information fusion |
| CN102589890A (en) * | 2012-03-01 | 2012-07-18 | 上海电力学院 | Integrated fault diagnostic method of steam turbine based on CPN (counter-propagation network) and D-S (dempster-shafer) evidences |
Non-Patent Citations (4)
| Title |
|---|
| 张国军等: "《基于D-S理论的综合故障诊断决策方法》", 《后勤工程学院学报》 * |
| 杨广等: "《神经网络与D-S证据理论分层融合的柴油机综合故障诊断方法研究》", 《武汉理工大学学报（交通科学与工程版）》 * |
| 王娜等: "《基于神经网络与D-S证据理论的故障诊断》", 《仪器仪表学报》 * |
| 董辛等: "《基于矢谱和D-S理论的旋转机械故障诊断方法研究》", 《煤矿机械》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105899929A (en) * | 2013-11-05 | 2016-08-24 | 斯内克马公司 | Technical testing method |
| CN104457903A (en) * | 2014-12-31 | 2015-03-25 | 上海电力学院 | Method for protecting water level of boiler vapor drum |
| CN105758450B (en) * | 2015-12-23 | 2017-11-24 | 西安石油大学 | Met an urgent need based on multisensor the fire-fighting early warning sensory perceptual system construction method of robot |
| CN105758450A (en) * | 2015-12-23 | 2016-07-13 | 西安石油大学 | Fire protection pre-warning sensing system building method based on multiple sensor emergency robots |
| CN107545339A (en) * | 2016-06-27 | 2018-01-05 | 南京理工大学 | The wind power generating set on-line fault diagnosis method of DS evidence theories based on SCADA alarm signals |
| CN106154182A (en) * | 2016-08-26 | 2016-11-23 | 上海电力学院 | A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory |
| CN106154182B (en) * | 2016-08-26 | 2019-04-09 | 上海电力学院 | A kind of lithium battery method for diagnosing faults based on improvement D-S evidence theory |
| CN106447040A (en) * | 2016-09-30 | 2017-02-22 | 湖南科技大学 | Method for evaluating the health state of mechanical equipment based on heterogeneous multi-sensor data fusion |
| CN106447040B (en) * | 2016-09-30 | 2018-11-23 | 湖南科技大学 | Mechanical equipment health state evaluation method based on Heterogeneous Multi-Sensor Data fusion |
| CN106503643A (en) * | 2016-10-18 | 2017-03-15 | 上海电力学院 | Tumble detection method for human body |
| CN106503643B (en) * | 2016-10-18 | 2019-06-28 | 上海电力学院 | Tumble detection method for human body |
| CN107274112A (en) * | 2017-07-07 | 2017-10-20 | 许寅卿 | Improve the diagnosis algorithm model of oil dissolved gas |
| CN107274112B (en) * | 2017-07-07 | 2021-11-26 | 国网上海市电力公司 | Diagnostic algorithm model for improving dissolved gas in oil |
| CN107729920A (en) * | 2017-09-18 | 2018-02-23 | 江苏海事职业技术学院 | A kind of method for estimating state combined based on BP neural network with D S evidence theories |
| WO2019184557A1 (en) * | 2018-03-29 | 2019-10-03 | 华为技术有限公司 | Method and device for locating root cause alarm, and computer-readable storage medium |
| CN108761263A (en) * | 2018-05-24 | 2018-11-06 | 深圳大图科创技术开发有限公司 | A kind of fault diagnosis system based on evidence theory |
| CN108761263B (en) * | 2018-05-24 | 2021-03-12 | 中电华创(苏州)电力技术研究有限公司 | Fault diagnosis system based on evidence theory |
| CN109060398A (en) * | 2018-09-11 | 2018-12-21 | 上海电力学院 | A kind of multi-source information equipment fault diagnosis method |
| CN111723341A (en) * | 2020-06-15 | 2020-09-29 | 中国船舶重工集团公司第七0三研究所 | Multi-subset probability processing method for boiler fault diagnosis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102928231A (en) | Equipment fault diagnosis method based on D-S (Dempster-Shafer) evidence theory | |
| JP2019040431A (en) | Abnormality determination system, abnormality determination method, program, learnt model, and learnt model production method | |
| CN105158627A (en) | Open-circuit fault diagnosis method for drive system of double-winding fault-tolerant permanent-magnet motor | |
| CN106936627B (en) | Thermal power equipment performance monitoring method based on big data analysis and mining | |
| CN110714869B (en) | Method and device for detecting central offset of rotor, storage medium and equipment | |
| CN103176128A (en) | Method and system for forcasting state of wind generating set and diagnosing intelligent fault | |
| CN105114348A (en) | Device and method for induced draft fan fault early warning based on power station operation data | |
| CN104976139A (en) | Mechanical equipment state diagnosis method based on Gaussian model | |
| CN105182159A (en) | Fault diagnosis method of electric drive system | |
| CN103699514A (en) | Method for detecting steady state and judging operation condition of water treatment process in thermal power plant | |
| CN103925155A (en) | Self-adaptive detection method for abnormal wind turbine output power | |
| CN102418661A (en) | Fault diagnosis method for yaw system for wind driven generator | |
| CN111523260B (en) | Steam turbine rotor digital twin body construction method and monitoring system | |
| CN102706387A (en) | Dirty factor detection device for condenser | |
| CN107433940A (en) | For manipulating the method, apparatus and system of electrovacuum pump | |
| CN116641855B (en) | Wind generating set operation monitoring method, system, equipment and storage medium | |
| CN108490350B (en) | Fault diagnosis system and method for high-power permanent magnet motor | |
| CN117094164A (en) | Sensor layout optimization method based on mixed information entropy | |
| CN102288911A (en) | System and method for monitoring and diagnosing thermal generator set | |
| CN104457903A (en) | Method for protecting water level of boiler vapor drum | |
| CN103713236B (en) | A kind of powerline ice-covering fault automatic judging method | |
| CN111637369B (en) | Water supply pipe network anti-seepage maintenance method and maintenance system | |
| CN113670536B (en) | Power plant electricity water monitoring and informationized management method | |
| CN113298133B (en) | Method for diagnosing explosion tube fault of supercritical unit boiler | |
| CN106815641A (en) | A kind of Gas path fault diagnosis method based on model and Fuzzy Pattern Recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C05 | Deemed withdrawal (patent law before 1993) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130213 |