CN110489862A - A kind of complex electromechanical systems life-span prediction method - Google Patents
A kind of complex electromechanical systems life-span prediction method Download PDFInfo
- Publication number
- CN110489862A CN110489862A CN201910765988.3A CN201910765988A CN110489862A CN 110489862 A CN110489862 A CN 110489862A CN 201910765988 A CN201910765988 A CN 201910765988A CN 110489862 A CN110489862 A CN 110489862A
- Authority
- CN
- China
- Prior art keywords
- component
- electromechanical systems
- failure rate
- life
- failure
- 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
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention designs complex electromechanical systems life prediction field, and in particular to a kind of complex electromechanical systems life-span prediction method.This method carries out life prediction to complex electromechanical systems based on the method that experience wavelet transformation and Petri network combine.Subcomponent nodes statistical model is constructed to system first, determines the critical component of system, experience wavelet decomposition is carried out to the detection signal of critical component and obtains multiple experience wavelet function components, critical component is carried out to accelerate experiment of degenerating, the variation of record detection signal.The functional image that each frequency band amplitude changes over time in the experience wavelet function component of original signal is obtained with this.The coupled relation between component is obtained according to the physical model of each key point of system, establishes the Petri net model of complex electromechanical systems.Corresponding unit failure rate r is determined according to characteristic signali(t), the Petri network after being changed determines the malfunction of system.In the correlation function R (r of system failure rate and unit failure rate1,r2,...,rs) in corresponding image, system current failure rate is obtained to the time difference Δ t of failure threshold, time difference Δ t is the remaining life of system, to achieve the purpose that complex electromechanical systems life prediction.
Description
Technical field
Disclosure herein refer to complex electromechanical systems life prediction fields, and in particular to a kind of complex electromechanical systems life prediction
Method.
Background technique
Complex electromechanical systems refer generally to that structure is complicated, collect mechanical, electrical, liquid and are controlled in integrated large-sized power equipment.With
The scale of scientific and technological progress, modern production constantly expands, core of the complex electromechanical systems as equipment manufacture, and safety is reliable
Property problem limits industry development.Due to the cost problem of equipment, complete replacement system needs to expend a large amount of manpower and material resources, together
When plant downtime during also along with a large amount of economic loss.Therefore, reasonable prediction lifetime of system can to raising security of system
By property, reduces economic loss and be of great significance.
The method of existing life prediction research can be mainly divided into the method based on mechanism model and data-driven.Based on machine
The method of reason model is to establish the mathematical model of failure mechanism or damage principle to describe the performance degradation of system or component
Process.Method based on data-driven is based on Probability Statistics Theory, by establishing the statistical model based on Heuristics
Or stochastic model, solution obtains the probability distribution of remaining life, to quantify the uncertainty of predicting residual useful life result.And due to
System structure is complicated, and mechanism model is difficult to obtain, therefore the method for data-driven is widely used.
And the object of life prediction at present is mostly the life prediction of triangular web component/part.It is mostly system-level prediction
By component-level to system-level, life prediction first is carried out to various parts, recycles Copula function to obtain system lifetim pre-
Survey model.And how to handle the coupled relation between each component becomes the Major Difficulties for establishing lifetime of system prediction model.
Summary of the invention
In view of this, it is an object of the invention to propose a kind of life-span prediction method of complex electromechanical systems.This method is logical
The experience wavelet transformation of mistake carries out feature extraction to measured signal, then establishes system for the coupled relation between system unit
Petri net model, to achieve the purpose that lifetime of system is predicted.
In order to achieve the above objectives, technical solution of the present invention provides a kind of complex electromechanical systems life-span prediction method, described
Method the following steps are included:
1) subcomponent nodes statistical model is constructed to system, determines the critical component of system, realizes the dimension-reduction treatment of system.It closes
Key member is defined as: the sum of each critical component number of nodes/component total node number >=ε, and critical component is able to achieve system major part
Function.
2) EWT decomposition is carried out by the detection signal of experience wavelet transformation (EWT) to complex electromechanical systems critical component, obtained
To the experience wavelet function component (EWF) of original signal.
3) critical component is carried out accelerating experiment of degenerating, the variation of record detection signal.The experience of original signal is obtained with this
The functional image that each frequency band amplitude changes over time in wavelet function component.Definition component performance degradation curve be frequency band amplitude with
Time attenuation curve.Feature band amplitude attenuation reaches δ than for the first timeiWhen assert component failure.
4) coupled relation between component is obtained according to the physical model of each critical component of system, determines component status change
Condition and stream direction.It is directed to the degeneration of system mode simultaneously, system mode is subjected to discrete division.Defining the system failure can not
Selfreparing, can only transfer from from low fault case to high fault case.Comprehensive library institute, flow relation establish system with conditions such as transition
Petri pessimistic concurrency control.
5) the failure threshold ω for defining system, thrashing is assert when system reaches failure threshold.It is netted using Patri
To the correlation function R (r of system failure rate and unit failure rate1,r2,...,rs)。
6) corresponding unit failure rate r is determined according to characteristic signali(t), the Petri network after being changed, determine be
The malfunction of system.In the correlation function R (r of system failure rate and unit failure rate1,r2,...,rs) in corresponding image, obtain
To system current failure rate to the time difference Δ t of failure threshold, according to the definition in system spare service life, time difference Δ t is to be
The remaining life L of system realizes system lifetim prediction.
The beneficial conditions that the present invention reaches are as follows: the present invention is crucial by carrying out statistical modeling extraction system to system node number
Component carries out feature extraction to component measured signal using experience wavelet transformation, establishes the Petri model of system then to realize
Consider the system-level life prediction of component coupled relation.Experience Wavelet transformation is decomposed single signal by establishing adaptive wavelet
For the component of different amplitude-frequencies, the feature extraction of signal is realized.The advantage of experience wavelet transformation is that its theoretical basis is complete, calculates
Complexity is low, and calculation amount is small, good to high-frequency characteristic component extraction effect, and has obtained practical application in multiple fields.And Perti
Netcom cross transition cause to indicate library flowing can indicate different conditions with transition with the dynamic operation of simulation system
Between change condition solve the problems, such as legacy system life time of the level predict founding mathematical models, it is ensured that system-level life prediction can
By Du Genggao.
Detailed description of the invention
In order to keep the purpose of the present invention, technical scheme and beneficial effects clearer, the present invention provides following attached drawing and carries out
Illustrate:
Fig. 1 is the flow diagram of complex electromechanical systems life-span prediction method of the present invention.
Fig. 2 is the Petri network structural schematic diagram of complex electromechanical systems life-span prediction method of the present invention.
Specific embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is further elaborated.
1) subcomponent nodes statistical model is constructed to system, altogether n component inputoutput section is counted and pressed for department of statistic respectively
It is arranged with according to descendingThen the total nodal point number of component is as shown in Equation 1.
2) critical component for determining system realizes the dimension-reduction treatment of system.Critical component is defined as: each critical component node
The sum of number/component total node number >=ε, and critical component is able to achieve system major part function.To setIn preceding s component section
Points summation obtains Ks, so that Ks/K≥ε.Under the premise of the combination of s component is able to achieve system basic functions, s value is taken most
It is small, obtain the critical component set P of system:
P={ p1,p2,...,ps} (2)
For a certain complex electromechanical systems, it includes: electrical, driving, control, sensing, machinery 5 that we, which set its critical component,
A critical component.
3) Fourier transformation is carried out to the measured signal of critical component and obtains frequency spectrum F (ω), one group of frequency is marked off to frequency spectrum
Compose line of demarcation.It first determines that the boundary ω is [0, π], the local maximum of frequency spectrum is searched in range and is arranged according to descending, normalizing
Changing processing is 0 to 1.When local maximum number m < n, frequency spectrum line of demarcation number is modified to local maximum number, that is, takes n
=m；N-1 maximum before retaining when m >=n.Taking two local maximum intermediate frequencies is frequency spectrum line of demarcation, this n-1 line of demarcation
Divide the spectrum into n part.
4) according to obtained line of demarcation, experience scaling function and experience small echo letter are constructed using Meyer wavelet construction method
Number, obtains orthogonal wavelet filter.Shown in Meyer small echo experience scaling function and experience wavelet function such as formula (3) and formula (4).
Wherein:
β (x)=x4(35-84x+70x2-20x3) (5)
τn=γ ωn (6)
5) detail coefficients and approximation coefficient are determined with the method for inner product, detail coefficients and approximation coefficient respectively such as formula (8) and
Shown in formula (9).
Wherein, ψnFor experience wavelet function, upper horizontal line indicates its complex conjugate, and superscript indicates that its corresponding Fourier becomes
It changes.Empirical modal component function may be expressed as:
6) critical component is carried out accelerating experiment of degenerating, in the failure mechanism for not changing component and increases new Failure Factor
Under the premise of, proof stress is improved, such as environmental stress, thermal stress, the failure process of acceleration components promote component in a short time
Failure.The variation of record detection signal, obtains each frequency band amplitude in the experience wavelet function component of original signal with this and becomes at any time
The functional image of change.Definition component performance degradation curve is frequency band amplitude attenuation curve at any time, unit failure rate are as follows:
ri(t)=δ (t) × 100% (12)
Feature band amplitude attenuation ratio δ (t) reaches δ for the first timeiWhen assert component failure.
7) coupled relation between component is obtained according to the physical model of each key point of system, determines component status change
Condition and stream direction.It is directed to the degeneration of system performance simultaneously, system fault condition be divided into intermittent fault, hard fault,
Failure.Define the system failure can not selfreparing, can only be changed from low fault case to high fault case.In the library that Petri network has been determined
We establish out the Petri net model of complex electromechanical systems as shown in Figure 2 after institute, transition and flow relation.Shown in Fig. 2
In Petri network, p1,p2,p3,p4,p5For source library institute, p12For purpose library institute.Original state M0It is represented by as shown in formula (13).
M0=(1,1,1,1,1,0,0,0,0,0,0,0) (13)
M for library Vector Groups, " 1 " indicates to contain Tokken in the institute of the library, empty carried out by " 0 " library representation.Library containing Tokken
It is expressed as current state.The variation with static Petri network dynamic display system state is realized by the transformation of M.
p1,p2,p3,p4,p5Respectively indicate electrical, driving, control, sensing, the normal operating conditions of mechanical each component, when certain
Unit failure rate reaches the transfer that transition trigger condition then causes Tokken, the i.e. variation of state, and corresponding component or component are constituted
Subsystem fault, show as library institute p6,p7,p8,p9Contain Tokken.Change t5Input and output showed by component to system
Conversion, library institute p10,p11,p12Three kinds of malfunctions of expression system: intermittent fault, hard fault, failure.Change t6,t7Corresponding thing
Part is defined as the variation of system failure rate.
8) the failure threshold ω for defining system, thrashing is assert when system failure rate reaches failure threshold.It utilizes
Patri net obtains the correlation function of system failure rate Yu unit failure rate:
R (t)=R (r1(t),r2(t),...,r(t)s) (14)
9) life prediction is carried out to system.By taking a kind of complex electromechanical systems Petri net model shown in Fig. 2 as an example, electrically
The subsystems couple that component and driving part are constituted, corresponding Petri net model are mostly to change t because of a fruit1Shooting condition definition
For { r1(t)≥ρ1|r2(t)≥ρ2, wherein ρ1,ρ2The respectively signal attenuation ratio δ of componentiCorresponding failure threshold.Equally
Ground, for changing t5, event F { r can be defined1,r2,r3,r4,r5, t is changed when event F occurs5Enabled, state occurs to turn
It moves, realizes the conversion by component to system.Corresponding unit failure rate r is determined according to the state of each component current demand signali(t),
Petri network after being changed determines the malfunction of system.Remaining life is defined as:
L=inf { l > 0:R (l+tk)≥ω|R0:k,R(tk) < ω (15)
In correlation function R (t)=R (r of system failure rate and unit failure rate1(t),r2(t),...,r(t)s) corresponding
In image, current failure rate is obtained to the time difference Δ t of failure threshold, according to the definition in system spare service life, the time difference, Δ t was
For the remaining life L of system.
Claims (6)
1. a kind of complex electromechanical systems life-span prediction method, which comprises the steps of:
1) subcomponent nodes statistical model is constructed to system, determines the critical component of system, realizes the dimension-reduction treatment of system.Crucial portion
Part is defined as: the sum of each critical component number of nodes/component total node number >=ε, and critical component is able to achieve system major part function.
2) EWT decomposition is carried out by the detection signal of experience wavelet transformation (EWT) to complex electromechanical systems critical component, obtains original
The experience wavelet function component (EWF) of signal.
3) critical component is carried out accelerating experiment of degenerating, the variation of record detection signal.The experience small echo of original signal is obtained with this
The functional image that each frequency band amplitude changes over time in function component.Definition component performance degradation curve be frequency band amplitude at any time
Attenuation curve.Feature band amplitude attenuation reaches δ than for the first timeiWhen assert component failure.
4) coupled relation between component is obtained according to the physical model of each critical component of system, determines the item of component status change
Part and stream direction.It is directed to the degeneration of system mode simultaneously, system mode is subjected to discrete division.Defining the system failure can not review one's lessons by oneself
It is multiple, it can only transfer from from low fault case to high fault case.Comprehensive library institute, flow relation establish the Petri of system with conditions such as transition
Pessimistic concurrency control.
5) the failure threshold ω for defining system, thrashing is assert when system reaches failure threshold.Utilize Patri net system
Correlation function R (the r for the failure rate and unit failure rate of uniting1,r2,...,rs)。
6) corresponding unit failure rate r is determined according to characteristic signali(t), the Petri network after being changed determines the event of system
Barrier state.In the correlation function R (r of system failure rate and unit failure rate1,r2,...,rs) in corresponding image, obtain system
Current failure rate is to the time difference Δ t of failure threshold, and according to the definition in system spare service life, time difference Δ t is the surplus of system
Remaining service life L realizes system lifetim prediction.
2. a kind of according to claim 1, complex electromechanical systems life-span prediction method, which is characterized in that the step 1)
System core component is determined by constructing subcomponent nodes statistical model to system: department of statistic's altogether n component inputoutput section
It counts and is arranged with according to descending{k1,k2,...,kn, and obtain component total node number K.To setIn preceding s component
Number of nodes sums to obtain Ks, so that Ks/K≥ε.Under the premise of the combination of s component is able to achieve system basic functions, s value is taken most
It is small.Method according to this obtains the critical component set P of system.
3. a kind of according to claim 1, complex electromechanical systems life-span prediction method, which is characterized in that the step 2)
EWT decomposition is carried out by measured signal of the experience wavelet transformation to complex electromechanical systems key point.First determine the boundary ω be [0,
π], the local maximum of frequency spectrum is searched in range and is arranged according to descending, and normalized is 0 to 1.Local maximum number
When m < n, frequency spectrum line of demarcation number is modified to local maximum number, that is, takes n=m；N-1 maximum before retaining when m >=n.
Taking two local maximum intermediate frequencies is frequency spectrum line of demarcation, this n-1 line of demarcation divides the spectrum into n part.
4. a kind of according to claim 1, complex electromechanical systems life-span prediction method, which is characterized in that the step 3)
The method for obtaining component capabilities degenerated curve are as follows: in the premise of the new Failure Factor of the failure mechanism and increase for not changing component
Under, proof stress is improved, such as environmental stress, thermal stress, the failure process of acceleration components promote component to fail in a short time.
The variation of record detection signal, obtains the letter that each frequency band amplitude changes over time in the experience wavelet function component of original signal with this
Number image.Definition component performance degradation curve is frequency band amplitude attenuation curve at any time.
5. a kind of according to claim 1, complex electromechanical systems life-span prediction method, which is characterized in that the step 4)
Petri net model establish.The coupled relation between component is obtained according to the physical model of each key point of system, determines component shape
The condition of state transition and stream direction.It is directed to the degeneration of system performance simultaneously, system fault condition is carried out to be divided into intermittent fault,
Hard fault, failure.Define the system failure can not selfreparing, can only be changed from low fault case to high fault case.Comprehensive library institute, stream close
System establishes the Petri net model of system with conditions such as transition.Library containing Tokken is expressed as current state.Pass through petri
Net transformation uses the variation of static Petri network dynamic display system state to realize.
6. a kind of according to claim 1, complex electromechanical systems life-span prediction method, which is characterized in that the step 6)
Predicting residual useful life.Corresponding unit failure rate r is determined according to the state of each component current demand signali(t), after being changed
Petri network determines the malfunction of system.Remaining life is defined as:
L=inf { l > 0:R (l+tk)≥ω|R0:k,R(tk) < ω (1)
In correlation function R (t)=R (r of system failure rate and unit failure rate1(t),r2(t),...,r(t)s) corresponding image
In, current failure rate is obtained to the time difference Δ t of failure threshold, and according to the definition in system spare service life, time difference Δ t is to be
The remaining life L of system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910765988.3A CN110489862A (en) | 2019-08-19 | 2019-08-19 | A kind of complex electromechanical systems life-span prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910765988.3A CN110489862A (en) | 2019-08-19 | 2019-08-19 | A kind of complex electromechanical systems life-span prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110489862A true CN110489862A (en) | 2019-11-22 |
Family
ID=68552041
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910765988.3A Pending CN110489862A (en) | 2019-08-19 | 2019-08-19 | A kind of complex electromechanical systems life-span prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110489862A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112001545A (en) * | 2020-08-24 | 2020-11-27 | 中国石油大学(华东) | Digital twin-driven marine oil underwater production system fault prediction method and system |
| CN112329339A (en) * | 2020-10-27 | 2021-02-05 | 河北工业大学 | Short-term wind speed prediction method for wind power plant |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110202227A1 (en) * | 2010-02-17 | 2011-08-18 | Gm Global Technology Operations, Inc. | Health prognosis for complex system using fault modeling |
| CN102393922A (en) * | 2011-06-23 | 2012-03-28 | 山西省电力公司晋中供电分公司 | Fuzzy Petri inference method of intelligent alarm expert system of transformer substation |
| CN102447735A (en) * | 2011-11-09 | 2012-05-09 | 重庆大学 | Reliability analyzing method for DAML-S (Darpa Agent Markup Language for Services) composite services |
| US20140208287A1 (en) * | 2013-01-18 | 2014-07-24 | Harbin Institute Of Technology | Energy Consumption Simulation and Evaluation System for Embedded Device |
| WO2014124049A2 (en) * | 2013-02-06 | 2014-08-14 | The Board Of Trustees Of The University Of Illinois | Stretchable electronic systems with containment chambers |
| CN104184632A (en) * | 2014-09-03 | 2014-12-03 | 重庆大学 | Method for analyzing performance of communication protocol router |
| AU2014202322A1 (en) * | 2014-04-29 | 2015-11-12 | Canon Kabushiki Kaisha | Wavelet denoising of fringe image |
| CN105488537A (en) * | 2015-12-11 | 2016-04-13 | 中国航空工业集团公司西安飞机设计研究所 | Method for representing component fault attributes based on Petri net |
| CN105694079A (en) * | 2016-01-26 | 2016-06-22 | 天津大学 | Method for stopping wrinkles from being formed on surface of azobenzene thin film by light illumination |
| CN106599352A (en) * | 2016-11-07 | 2017-04-26 | 西北工业大学 | Reliability analysis method for aircraft fly-by-wire control system |
-
2019
- 2019-08-19 CN CN201910765988.3A patent/CN110489862A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110202227A1 (en) * | 2010-02-17 | 2011-08-18 | Gm Global Technology Operations, Inc. | Health prognosis for complex system using fault modeling |
| CN102393922A (en) * | 2011-06-23 | 2012-03-28 | 山西省电力公司晋中供电分公司 | Fuzzy Petri inference method of intelligent alarm expert system of transformer substation |
| CN102447735A (en) * | 2011-11-09 | 2012-05-09 | 重庆大学 | Reliability analyzing method for DAML-S (Darpa Agent Markup Language for Services) composite services |
| US20140208287A1 (en) * | 2013-01-18 | 2014-07-24 | Harbin Institute Of Technology | Energy Consumption Simulation and Evaluation System for Embedded Device |
| WO2014124049A2 (en) * | 2013-02-06 | 2014-08-14 | The Board Of Trustees Of The University Of Illinois | Stretchable electronic systems with containment chambers |
| AU2014202322A1 (en) * | 2014-04-29 | 2015-11-12 | Canon Kabushiki Kaisha | Wavelet denoising of fringe image |
| CN104184632A (en) * | 2014-09-03 | 2014-12-03 | 重庆大学 | Method for analyzing performance of communication protocol router |
| CN105488537A (en) * | 2015-12-11 | 2016-04-13 | 中国航空工业集团公司西安飞机设计研究所 | Method for representing component fault attributes based on Petri net |
| CN105694079A (en) * | 2016-01-26 | 2016-06-22 | 天津大学 | Method for stopping wrinkles from being formed on surface of azobenzene thin film by light illumination |
| CN106599352A (en) * | 2016-11-07 | 2017-04-26 | 西北工业大学 | Reliability analysis method for aircraft fly-by-wire control system |
Non-Patent Citations (8)
| Title |
|---|
| JÉRÔME GILLES: "Empirical Wavelet Transform", 《IEEE TRANSACTIONS ON SIGNAL PROCESSING 》 * |
| NOUREDINE OUELAA等: "Application of the Empirical Mode Decomposition method for the prediction of the tool wear in turning operation", 《MECHANICAL TECHNOLOGIES》 * |
| TRAN VAN TUNG等: "Machine Fault Diagnosis and Prognosis: The State of The Art", 《INTERNATIONAL JOURNAL OF FLUID MACHINERY AND SYSTEMS》 * |
| 刘义艳等: "基于 EWT 和 FESN 的结构健康状态趋势预测", 《应用力学学报》 * |
| 刘建华: "故障Petri网在机械系统剩余寿命预测中的应用研究", 《石家庄铁道学院学报》 * |
| 化建宁: "故障Petri网及其在机械设备剩余寿命预测中的应用", 《中国知网硕士电子期刊》 * |
| 李传涛: "基于状态预测的设备管理功能模块设计与实现", 《中国知网硕士电子期刊》 * |
| 王 宁等: "基于改进 EWT的模拟电路故障诊断研究", 《计算机技术与发展》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112001545A (en) * | 2020-08-24 | 2020-11-27 | 中国石油大学(华东) | Digital twin-driven marine oil underwater production system fault prediction method and system |
| CN112001545B (en) * | 2020-08-24 | 2022-03-15 | 中国石油大学(华东) | Digital twin-driven marine oil underwater production system fault prediction method and system |
| CN112329339A (en) * | 2020-10-27 | 2021-02-05 | 河北工业大学 | Short-term wind speed prediction method for wind power plant |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Lithium-ion batteries remaining useful life prediction based on BLS-RVM | |
| Zhao et al. | Remaining useful life prediction of aircraft engine based on degradation pattern learning | |
| Li et al. | An adaptive data fusion strategy for fault diagnosis based on the convolutional neural network | |
| Guo et al. | A recurrent neural network based health indicator for remaining useful life prediction of bearings | |
| CN105701337B (en) | Fatigue life prediction method and device for wind turbine generator | |
| Han et al. | Intelligent fault diagnosis method for rotating machinery via dictionary learning and sparse representation-based classification | |
| CN110162857A (en) | A kind of flexible measurement method for surveying parameter towards complex industrial process difficulty | |
| CN105096053A (en) | Health management decision-making method suitable for complex process system | |
| CN110489862A (en) | A kind of complex electromechanical systems life-span prediction method | |
| Jaramillo et al. | A Bayesian approach for fatigue damage diagnosis and prognosis of wind turbine blades | |
| Liao et al. | Forecasting dissolved gases content in power transformer oil based on weakening buffer operator and least square support vector machine–Markov | |
| CN103439594B (en) | SF6Fault diagnosis System and method for | |
| CN109189834A (en) | Elevator Reliability Prediction Method based on unbiased grey fuzzy Markov chain model | |
| CN104298843A (en) | Complex mechanism dynamic cascading reliability modeling method based on colored stochastic Petri network | |
| Bi et al. | Bifurcation mechanism of bursting oscillations in parametrically excited dynamical system | |
| Zhao et al. | Research on digital twin driven rolling bearing model-data fusion life prediction method | |
| Shi et al. | Optimization and decomposition methods in network traffic prediction model: A review and discussion | |
| Storlie et al. | Modeling and predicting power consumption of high performance computing jobs | |
| Feizollahi et al. | Robust quadratic assignment problem with budgeted uncertain flows | |
| Tirovolas et al. | Introducing fuzzy cognitive map for predicting engine’s health status | |
| CN116050190B (en) | Product performance and degradation state analysis method based on digital twinning | |
| Limnios et al. | Ch. 14. an introduction to semi-markov processes with application to reliability | |
| Zhao et al. | Integrated equipment health prognosis considering crack initiation time uncertainty and random shock | |
| CN104753718A (en) | Deficiency service throughput rate complementing method and device based on non-negative polymerization | |
| Liu et al. | Uncertainty processing in prognostics and health management: An overview |
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 |