CN108985329A - A kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling - Google Patents
A kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling Download PDFInfo
- Publication number
- CN108985329A CN108985329A CN201810594868.7A CN201810594868A CN108985329A CN 108985329 A CN108985329 A CN 108985329A CN 201810594868 A CN201810594868 A CN 201810594868A CN 108985329 A CN108985329 A CN 108985329A
- Authority
- CN
- China
- Prior art keywords
- probability
- prior probability
- axial flow
- operating condition
- flow blower
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
Abstract
The present invention provides a kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling, comprising: clear axial flow blower fault type, clear failure may corresponding operating condition and signs；Bayesian network is constructed according to axial flow blower fault type and its corresponding operating condition and sign；The prior probability of fan trouble is modeled according to actual physical experience and reasonable assumption；Error analysis is carried out to model, guarantees the feasibility of model.The prior probability of failure can be divided into two-part linear superposition, operating condition is directed toward and historical information, and establishes the direct function relationship between prior probability and work information on the basis of the prior probability to axial flow blower failure carries out Rational Model by the present invention.Under different operating conditions, prior probability is different, so that network topology probability is more objective, has study property, while each diagnostic message is added to historical data base, updates network topology probability, make system intelligent.
Description
Technical field
The present invention relates to fault diagnosis field more particularly to a kind of axial flow blower failure pattra leaves based on prior probability modeling
This probability analysis method.
Background technique
Process fluid machinery is the machine that energy conversion, processing are carried out using the mixture of fluid or fluid and solid as object
Tool is the important component of Process Equipment.Process fluid machinery takes part in most of links in plant produced, is a work
Heart, power and the key equipment of factory.Once failure occur in fluid machines, huge wealth certainly will be caused to entire production
Loss is produced, serious person may cause safety accident.According to the strategic plan of made in China 2025, high-end equipment intelligence manufacture pair
The higher more urgent requirement of the propositions such as performance, safety and the efficiency of fluid machinery.For this purpose, how to prevent and diagnose process fluid machine
The failure that tool occurs seems particularly necessary.
In recent years, the fluid machinery method for diagnosing faults based on Bayesian Network achieves important breakthrough, but this
The network topology probability of a little methods be rule of thumb it is assumed that and immobilize so that whole network is excessively subjective.This method is adopted
With the mode modeled to prior probability, in conjunction with influencing in terms of historical information and operating condition reason two, building state value function and general
Rate function keeps prior probability more objective, and has certain study property.
Detailed description of the invention
Fig. 1 is axial flow blower parameter list；
Fig. 2 is the directed acyclic graph of the failure Bayesian network of axial flow blower；
Fig. 3 is that extent of corrosion changes over time curve；
Fig. 4 is that fastening force decline degree changes over time curve.
Summary of the invention
The present invention provides a kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling, by failure
Prior probability be divided into two-part linear superposition, operating condition is directed toward and historical information, and establish prior probability and work information it
Between direct function relationship.Under different operating conditions, prior probability is different, so that network topology probability is more objective, and will be each
Diagnostic message is added to historical data base, updates network topology probability, makes system intelligent.
A kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling, comprising:
Step 1: clear axial flow blower fault type, clear failure may corresponding operating condition and signs；
Step 2: Bayesian network is constructed according to axial flow blower fault type and its corresponding operating condition and sign；
Step 3: the prior probability of fan trouble is modeled according to actual physical experience and reasonable assumption；
Step 4: error analysis is carried out to model, guarantees the feasibility of model.
Preferably, in step 1 two quasi-representative of high spot reviews axial flow blower failure: rotor unbalance and foundation bolt
It loosens.
Preferably, the Bayesian network in step 2 is directed acyclic graph, it is divided into three node layers, respectively operating condition layer section
Point, failure node layer, sign node layer.
Preferably, the specific steps of step 3 are as follows:
The prior probability of step a. clear failure node layer is influenced to constitute with historical information two parts by operating condition factor, if
It is prior probability modeling for the first time, then historical information is zero；
Step b. rationally establishes influence mould of the operating condition node layer state to failure prior probability according to knowledge and experience
Type；
Step c. usage history information is modified prior probability, it is ensured that the reasonability of probability.
As further preferred, in step b, by operating condition node layer serialization, node state is described with " degree ", degree
The case where value range is [0,1], degree bigger expression operating condition factor is more serious.
As further preferred, in step b, influence of the operating condition layer to malfunctioning node prior probability is determined by degree, probability
Value and degree meet certain functional relation.
As further preferred, in step c, the failure posterior probability that the last time solves is believed as this history solved
Breath.
Preferably, prior probability is made of the linear combination that operating condition is influenced with historical information in step 4, pass through weight
Coefficient adjustment reduces error.
The present invention provides a kind of axial flow blower failure Bayesian probability analysis methods based on prior probability modeling, compare
The advantages of conventional method and current other methods is that prior probability of the invention is ART network, is run from Fan Equipment
Operating condition and historical experience information are set out, so that the equipment fault prior probability modeled by prior probability is with more objective
Property, and there is study property.In addition to this, portability of the invention is good, naive Bayesian network prior probability model more
Intelligence can be generally applicable to the failure prior probability modeling of various axial flow blowers.
Specific embodiment
In order to more specifically describe the present invention, with reference to the accompanying drawing and specific example application mode is to technology of the invention
Scheme is described in detail.
Example: it is a kind of based on prior probability modeling axial flow blower failure Bayesian probability analysis method in laboratory room small-sized
The application of axial flow blower failure prior probability modeling.
S01, as shown in Figure 1, being six rachis streaming fan parameter tables of the invention.
The fault type that may occur to it and the possible corresponding operating condition of specific failure are analyzed with sign, here emphasis
It investigates rotor unbalance and foundation bolt loosens these two types of typical axial flow blower failures, corresponding operating condition layer is because being known as product
Dirt, rotor machining or installation error, Unit Commitment be improper, the decline of dielectric corrosion, external load impacting, fastening force, related sign
Layer show as high frequency multiplication amplitude is big, horizontal vertical phase difference is 90 degree, fuselage shaking, power frequency ingredient obviously, amplitude becomes with revolving speed
Change occur half frequency ingredient when fundamental frequency advantage increases, loosens strong when sensitive, frequency increases and level, vertical phase difference be 0 degree or
180 degree；
S02, as shown in Fig. 2, the Bayesian network of building axial flow blower failure, the first-level nodes are operating condition layer, the second layer
Node is failure layer, and third node layer is sign layer.State model is established to operating condition layer, it is notable that in order to closer to
It is practical, it introduces " degree " this concept here to describe the node state of operating condition layer, for six nodes of operating condition layer, removes rotor
Except processing or installation error (b), Unit Commitment improper (c), it is accordingly to be regarded as continuous nodes.Defining " degree " is becoming certainly for node
Amount, the measurement as a kind of operating condition.The value range of degree is [0,1], and taking 0 expression degree is nothing, can be considered that event does not occur,
Such as a brand-new blower, without incrustation, then the degree a=0 of incrustation.Take 1 expression degree for maximum, such as one rotten
Deteriorate bad blower, and corrosion has reached enough seriously, at this time degree d=1, and degree is greater than 1 and is accordingly to be regarded as 1, because continuing to increase
Greatly without practical significance.The degree direct influence of node is to the prior probability of failure, and degree is bigger, and the probability that failure occurs is got over
Greatly.How degree solves, can contact in it continuous latent variable, establish functional relation；
S03 considers this node of incrustation: rule of thumb it is found that incrustation and time t (day) are in be incremented by relationship, and incrustation journey
Degree is bigger, and incrustation speed accordingly slows down, and describes this relationship using power function.Again according to the use environment of blower, apparent wind machine exists
In the case where any processing, (for the sake of convenient, take 720 days) degree of fouling maximum when validity period was up to 2 years, i.e. a=1, so its
The expression formula of degree at any time are as follows:
S04, consider rotor machining or installation error this node: the node is related with blower factory state, when blower is given
Periodically, error is i.e. given, and blower factory state used in this experiment is good, error 0, i.e. degree b=0.
S05, consider this improper node of Unit Commitment: it is mainly the improper number of start and stop that the operating condition, which influences rotor unbalance,
When number is more, the probability that rotor unbalance occurs is bigger, when might as well set the improper number of start and stop and reaching 10 times, the event degree
It is 1, then degree and the improper number of start and stop are linear:
When the improper number of start and stop equals or exceeds 10 times, influence of the operating condition to rotor unbalance reaches maximum.
S06 considers this node of dielectric corrosion: since fan operation process might have oil droplet or corrosive gas seeps
Enter, therefore need to consider the influence of dielectric corrosion.Extent of corrosion d is related with corrosion rate and time, and use environment is different, corrosion rate
Moment variation, for the sake of convenient, it is believed that constant rate remains average corrosion rate, then the monodrome letter that degree d portrays as time t
Number.Rule of thumb, if blower is under without any disposition, extent of corrosion maximum is reached when using 4 years, and before corrosion process
Face is slow, intermediate fast, slow below, can be described with half period sine curve, such as Fig. 3 is apparent from according to trigonometric function knowledge:
From the figure 3, it may be seen that blower when using starting and at 4 years or so, corrode it is slower, it is fast in intermediate stage corrosion.
S07, consider this node of external load impacting: the degree of load impacting is improper similar with start and stop, depends on impact
Number, when might as well think blower by impact 20000 times, impact cumulative effect is maximum, i.e. impact degree e=1.Assuming that external load
Lotus is impacted in the form of recurrent pulse, and the period is one hour, then impacts 24 times within one day, be easy to be calculated, when impact 834 days
When, degree is maximum, the expression formula of degree are as follows:
S08 considers that fastening force declines this node: based on practical experience, with the increase of use time, bolted
Fastening force can be gradually reduced, and when bolt is more and more loose, decline increasingly severe, therefore the degree f of the operating condition is increase with time
And increase, increased rate constantly increases, and is described herein with exponential function.In the case where thinking no any processing, two are used
Bolt looseness degree reaches maximum f=1 when year:
As shown in figure 4, being degree curve of the invention.
S09 carries out function modelling to the prior probability of failure node layer.For rotor unbalance, there are two types of states, occur
(IB) and do not occurFoundation bolt is loosened, there are three types of states, strongly (LS), generallyDo not occur
Rotor unbalance occur prior probability will be made of two parts, one is according to the degree of each node of operating condition layer into
The assessment of row apriority, the second is provided by historical information, and often to account for weight bigger by the latter.Such as the operating condition of blower characterizes and occurs
Unbalanced probability is small, and in last time aerator supervision, the uneven probability obtained is larger, then this time unbalanced prior probability
It should be bigger than normal.Prior probability expression formula is as follows:
P (IB)=λ P (IB)Operating condition+(1-λ)P(IB)History
Fault diagnosis is carried out with Bayesian network, the above formula left side is the prior probability of IB.The right is made of two parts, right
Side formula first is that the probability determined by operating condition, formula second is that the probability determined by historical information (i.e. last time Bayesian inference is calculated
The probability that IB occurs, if last detection has found imbalance fault and repairs, then this historical probabilities are 0, first
It tests probability only to be determined by operating condition).
The probability determined by operating condition can be determined that different models has different take by functional relation f (a, b, c, d)
Method, this patent takes exponential relationship, and four operating condition degree weights are different, considers the influence to rotor unbalance, rotor machining (peace
Dress) error > dielectric corrosion > incrustation > start and stop are improper, and weighting value ratio is 4:3:2:1, and expression formula is as follows:
P(IB)Operating condition=f (a, b, c, d)=1-e-(a+2b+0.5c+1.5d)
Work as a=b=c=d=1, i.e., when all operating condition degree are all the most serious, the uneven probability of happening that operating condition is directed toward is big
About 99.3%.Equally, when all operating condition degree are 0, imbalance fault does not occur for probability 0, i.e. operating condition layer direction, meets
Logic.
Foundation bolt loosens probability and is codetermined by operating condition layer and historical information, and there are three types of value states for the node, strongly,
Generally and do not occur, probability is expressed as follows:
P (LS)=λ P (LS)Operating condition+(1-λ)P(LS)History
Historical information still take it is last diagnose as a result, i.e. foundation bolt loosen it is strong, general, do not occur it is corresponding general
Rate, if being repaired to equipment, strong and general historical information probability takes 0.Now establish operating condition next -event estimator and operating condition
Functional relation between degree should meet a certain specific operation m (d, e, f):
Probability strong, general and not occurring is described with trigonometric function, it is contemplated that the probability not occurred generally is larger, therefore:
P(LS)Operating condition=sin2(2πx)sin2(2πy)|
Wherein x, y are the amounts determined by operating condition degree d, e, f, and value range is [0,1].X's in this patent, y follows the example of
Reason is to think that fastening force decline is identical on foundation bolt loosening influence with dielectric corrosion, but compared to external load impacting, influence
It is smaller, therefore e should be leading factor.For above formula due to the sinusoidal monotonicity with cosine, y is the variable of main derived function value size, therefore
And set y=e.
S10 carries out error analysis to model, guarantees the feasibility of model.By taking incrustation as an example, if in actual life, incrustation
Time when being 1000 days degree just reach 1, then 720 days of this model foundation and really there is error, now solve this
Influence of the error to prior probability:
B=c=d=0 might as well be set, blower uses 2 years, according to the degree of fouling a=1 that step S03 is calculated, really
Degree of fouling a*=0.849 takes operating condition and historical information weight coefficient=0.3:
P(IB)Operating condition=1-e-0.9=0.593
P*(IB)Operating condition=1-e-0.9*0.849=0.534
P (IB)=0.3P (IB)Operating condition+0.7P(IB)History=0.1779
P*(IB)=0.3P*(IB)Operating condition+0.7P*(IB)History=0.1602
Δ=P (IB)-P*(IB)=0.0177 ≈ 1.8%
By above-mentioned calculating process, after the weighting between operating condition restricts and the weighting of history and operating condition restricts, error quilt
The very little of scaling.Time phase difference 280 days in example, last probable error but only have 1.8%, illustrate that this weighting function solves
Prior probability model is with certain rational.
Technical solution of the present invention and beneficial effect is described in detail in above-described specific embodiment, Ying Li
Solution is not intended to restrict the invention the foregoing is merely presently most preferred embodiment of the invention, all in principle model of the invention
Interior done any modification, supplementary, and equivalent replacement etc. are enclosed, should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling, comprising the following steps:
Step 1: clear axial flow blower fault type, clear failure may corresponding operating condition and signs；
Step 2: Bayesian network is constructed according to axial flow blower fault type and its corresponding operating condition and sign；
Step 3: the prior probability of fan trouble is modeled according to actual physical experience and reasonable assumption；
Step 4: error analysis is carried out to model, guarantees the feasibility of model.
2. the axial flow blower failure Bayesian probability analysis method according to claim 1 based on prior probability modeling,
It is characterized in that, high spot reviews rotor unbalance and foundation bolt loosen this two classes axial flow blower failure in the step 1.
3. the axial flow blower failure Bayesian probability analysis method according to claim 1 based on prior probability modeling,
It is characterized in that, the Bayesian network in step 2 is directed acyclic graph, is divided into three node layers, respectively operating condition node layer, failure
Node layer, sign node layer.
4. the axial flow blower failure Bayesian probability analysis method according to claim 1 based on prior probability modeling,
It is characterized in that, the specific steps of step 3 are as follows:
The prior probability of step a. clear failure node layer is influenced to constitute with historical information two parts by operating condition factor, if it is head
Secondary prior probability modeling, then historical information is zero；
Step b. rationally establishes influence model of the operating condition node layer state to failure prior probability according to knowledge and experience；
Step c. usage history information is modified prior probability, it is ensured that the reasonability of probability.
5. the axial flow blower failure Bayesian probability analysis method according to claim 4 based on prior probability modeling,
Be characterized in that, in step b, by operating condition node layer serialization, node state is described with " degree ", degree value range be [0,
1], the case where degree bigger expression operating condition factor, is more serious.
6. the axial flow blower failure Bayesian probability analysis method according to claim 4 based on prior probability modeling,
It is characterized in that, in step b, influence of the operating condition layer to malfunctioning node prior probability is determined by degree, and probability value and degree meet one
Fixed functional relation.
7. the axial flow blower failure Bayesian probability analysis method according to claim 4 based on prior probability modeling,
It is characterized in that, in step c, the historical information that is solved as this of failure posterior probability that the last time solves.
8. the axial flow blower failure Bayesian probability analysis method according to claim 1 based on prior probability modeling,
It is characterized in that, in step 4, prior probability is made of the linear combination that operating condition is influenced with historical information, is adjusted by weight coefficient
Reduce error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810594868.7A CN108985329B (en) | 2018-06-11 | 2018-06-11 | Bayesian probability analysis method for fan fault based on prior probability modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810594868.7A CN108985329B (en) | 2018-06-11 | 2018-06-11 | Bayesian probability analysis method for fan fault based on prior probability modeling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108985329A true CN108985329A (en) | 2018-12-11 |
| CN108985329B CN108985329B (en) | 2020-10-30 |
Family
ID=64540164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810594868.7A Active CN108985329B (en) | 2018-06-11 | 2018-06-11 | Bayesian probability analysis method for fan fault based on prior probability modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108985329B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111080039A (en) * | 2020-03-17 | 2020-04-28 | 浙江上风高科专风实业有限公司 | Fan cluster fault prediction method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915234A (en) * | 2010-07-16 | 2010-12-15 | 西安交通大学 | Method for diagnosing compressor-associated failure based on Bayesian network |
| CN104005917A (en) * | 2014-04-30 | 2014-08-27 | 叶翔 | Method and system for predicting wind machine state based on Bayesian reasoning mode |
| CN106875053A (en) * | 2017-02-15 | 2017-06-20 | 济中节能技术（苏州）有限公司 | Fault Diagnosis of Fan method and diagnostic system based on bayes algorithms |
| CN107045584A (en) * | 2017-05-11 | 2017-08-15 | 河海大学 | A kind of power frequency vibration abnormal failure diagnostic method suitable for water pump rotor system |
| CN107796609A (en) * | 2017-10-17 | 2018-03-13 | 西安建筑科技大学 | A kind of handpiece Water Chilling Units method for diagnosing faults based on DBN model |
-
2018
- 2018-06-11 CN CN201810594868.7A patent/CN108985329B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101915234A (en) * | 2010-07-16 | 2010-12-15 | 西安交通大学 | Method for diagnosing compressor-associated failure based on Bayesian network |
| CN104005917A (en) * | 2014-04-30 | 2014-08-27 | 叶翔 | Method and system for predicting wind machine state based on Bayesian reasoning mode |
| CN106875053A (en) * | 2017-02-15 | 2017-06-20 | 济中节能技术（苏州）有限公司 | Fault Diagnosis of Fan method and diagnostic system based on bayes algorithms |
| CN107045584A (en) * | 2017-05-11 | 2017-08-15 | 河海大学 | A kind of power frequency vibration abnormal failure diagnostic method suitable for water pump rotor system |
| CN107796609A (en) * | 2017-10-17 | 2018-03-13 | 西安建筑科技大学 | A kind of handpiece Water Chilling Units method for diagnosing faults based on DBN model |
Non-Patent Citations (3)
| Title |
|---|
| DA-ZHUAN WU;XIN-KUO JIANG;NING CHU;ETC.: "Numerical simulation on rotordynamic characteristics of annular seal under uniform and non-uniform flows", 《JOURNAL OF CENTRAL SOUTH UNIVERSITY》 * |
| 刘晓洁，吴家鑫: "贝叶斯网络在故障诊断中的应用", 《北京联合大学学报》 * |
| 胡兆勇, 屈梁生: "一种贝叶斯诊断网络的拓扑结构", 《西安交通大学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111080039A (en) * | 2020-03-17 | 2020-04-28 | 浙江上风高科专风实业有限公司 | Fan cluster fault prediction method and system |
| CN111080039B (en) * | 2020-03-17 | 2023-10-20 | 浙江上风高科专风实业有限公司 | Fan cluster fault prediction method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108985329B (en) | 2020-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103925155B (en) | The self-adapting detecting method that a kind of Wind turbines output is abnormal | |
| CN107609574A (en) | Wind turbines fault early warning method based on data mining | |
| US6587737B2 (en) | Method for the monitoring of a plant | |
| CN107045584B (en) | Power frequency vibration abnormity fault diagnosis method suitable for water pump rotor system | |
| CN102252843B (en) | Assessment method for rolling bearing performance variation | |
| CN103512651B (en) | The Wind turbines state monitoring method reported to the police based on vibration mutation and device | |
| CN103234753A (en) | Gaussian process modeling based wind turbine shafting state monitoring method | |
| CN105913124B (en) | System health status prediction technique based on Bayesian network and base layer data | |
| CN102938024B (en) | A kind of Wind turbines condition monitoring system performance estimating method | |
| CN107315396B (en) | A kind of state monitor maintenance and predictive maintenance Combined maintenance method and system for planning | |
| Djeddi et al. | Operational reliability analysis applied to a gas turbine based on three parameter Weibull distribution | |
| CN108985329A (en) | A kind of axial flow blower failure Bayesian probability analysis method based on prior probability modeling | |
| Helsen et al. | Wind farm operation and maintenance optimization using big data | |
| Manju et al. | A data mining study for condition monitoring on wind turbine blades using hoeffding tree algorithm through statistical and histogram features | |
| CN103808405A (en) | Automatic diagnostic method of gas power generator group vibration fault of device thereof | |
| Das et al. | Performance monitoring and failure prediction of industrial equipments using artificial intelligence and machine learning methods: A survey | |
| CN106126949A (en) | A kind of steam turbine generator running status appraisal procedure | |
| May et al. | Condition based maintenance for offshore wind turbines: the effects of false alarms from condition monitoring systems | |
| Cai et al. | Bayesian networks in fault diagnosis: Practice and application | |
| CN112597607A (en) | Predictive maintenance method and system for high-speed bearing of wind driven generator | |
| Bi et al. | Research on fault diagnosis for pumping station based on TS fuzzy fault tree and Bayesian network | |
| May et al. | Integrating structural health and condition monitoring: a cost benefit analysis for offshore wind energy | |
| CN113446157A (en) | Fault early warning method and equipment for offshore wind generating set | |
| Bezzaoucha et al. | Failure causes based wind turbine components classification and failure propagation: For proactive maintenance implementatio | |
| CN109840563A (en) | A kind of Fault Diagnosis of Fan method based on three-dimensional membership function fuzzy logic system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 1818 Renmin West Road, Dongguan Street, Shangyu District, Shaoxing City, Zhejiang Province Patentee after: Zhejiang Shangfeng high tech special wind industry Co.,Ltd. Address before: 312300 No. 1818 West Renmin Road, Shangyu District, Shaoxing, Zhejiang Patentee before: ZHEJIANG SHANGFENG HI-TECH SPECIALIZED WIND INDUSTRIAL Co.,Ltd. |