CN108256700A - A kind of maintenance of equipment method for predicting residual useful life and system - Google Patents
A kind of maintenance of equipment method for predicting residual useful life and system Download PDFInfo
- Publication number
- CN108256700A CN108256700A CN201810332285.7A CN201810332285A CN108256700A CN 108256700 A CN108256700 A CN 108256700A CN 201810332285 A CN201810332285 A CN 201810332285A CN 108256700 A CN108256700 A CN 108256700A
- Authority
- CN
- China
- Prior art keywords
- maintenance
- equipment
- stage
- degradation model
- degraded data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of maintenance of equipment method for predicting residual useful life and systems.This method includes：Establish the initial degradation model in each stage in maintenance of equipment degenerative process；Obtain the degraded data of maintenance of equipment in each stage；The parameter of the initial degradation model is determined according to the degraded data of the maintenance of equipment in each stage, obtains degradation model；According to the degradation model predictive maintenance equipment remaining life.Combined influence of the maintenance to degradation ratio and amount of degradation has been fully considered by this method and this system, has further enriched the life prediction theory based on modeling of degenerating, the precision of life prediction can be effectively improved.
Description
Technical field
The present invention relates to reliability engineering technique field, more particularly to a kind of maintenance of equipment method for predicting residual useful life and
System.
Background technology
With the continuous promotion of scientific and technological level, engineering key equipment is just towards the side of non-linear, complication, scale
To development.This kind equipment is divided the work according to task, and there is othernesses for difference, composition, load, environment etc., identical,
The synergy of internal factor and external environment will be undergone in operational process, and then by the gradual performance degradation of occurrence of equipment.When
When performance degradation accumulation is to certain amount grade, degeneration will develop into failure, and equipment performance can not meet task needs at this time.If
Engineers and technicians do not have found such failure in time, and the mankind will face the consequence that environmental disruption, property loss etc. can not be retrieved.
To avoid the generation of above-mentioned consequence, there is an urgent need to improve the safety and reliability of equipment, thus, in the past few decades, equipment
Life prediction receive paying close attention to for domestic and international researcher and engineers and technicians.
The service life that the method for current age prediction can be divided into life-span prediction method and data-driven based on failure mechanism is pre-
Two major class of survey method.Due to being difficult to obtain the potential failure mechanism of equipment in real process, thus, compared to based on failure machine
The life-span prediction method of reason, the life-span prediction method of data-driven are more exposed to the favor of scholars.The service life of data-driven is pre-
Survey method can be divided into traditional life-span prediction method, the life-span prediction method based on modeling of degenerating and the longevity based on data fusion
Order Forecasting Methodology.Traditional life-span prediction method is mainly fitted service life distribution using fail data, obtains corresponding
Service life distributed constant, service life distribution mainly have Weibull distributions, exponential distribution, log series model etc..However, with production skill
The continuous improvement of art and design level, the acquisition of the lifetime data of most of equipment become further difficult.Thus, traditional service life
Forecasting Methodology can face the defects of fail data is insufficient, it is difficult to ensure the result of life prediction.Service life based on modeling of degenerating is pre-
Survey method only needs Condition Monitoring Data (degraded data), and the cost that degraded data needs are obtained in engineering is lost much smaller than acquisition
The cost that data need is imitated, therefore, the life-span prediction method based on modeling of degenerating becomes the mainstream side in current age prediction field
Method.When carrying out life prediction, the degradation model of generally use mainly has：Gamma processes, Wiener processes, inverse Gaussian process, horse
Er Kefu processes etc..It is not difficult to find out in above-mentioned degradation model by analysis and has ignored influence of the maintenance to degenerative process.
Invention content
In view of the problems of the existing technology, the present invention provides a kind of maintenance of equipment method for predicting residual useful life and it is
System, to improve the accuracy to maintenance of equipment predicting residual useful life.
To achieve the above object, the present invention provides following schemes：
A kind of maintenance of equipment method for predicting residual useful life, including：
Establish the initial degradation model in each stage in maintenance of equipment degenerative process；
Obtain the degraded data of maintenance of equipment in each stage；
The parameter of the initial degradation model is determined according to the degraded data of the maintenance of equipment in each stage, is degenerated
Model；
According to the degradation model predictive maintenance equipment remaining life.
Optionally, it is further included before the initial degradation model in each stage in described establish in maintenance of equipment degenerative process：
Obtain the degraded data of maintenance of equipment；
Judge whether the degraded data is less than failure threshold；
If not, then it represents that maintenance of equipment fails；
If so, judging whether degraded data is less than repair threshold value；
If so, perform the degraded data for obtaining maintenance of equipment.
If it is not, then build the initial degradation model in maintenance of equipment degenerative process each stage.
Optionally, the ginseng that the initial degradation model is determined according to the degraded data of the maintenance of equipment in each stage
Number, obtains degradation model, specifically includes：
The degeneration mould in maintenance of equipment degenerative process each stage is calculated according to the degraded data of maintenance of equipment in each stage
The remaining degeneration factor of type；
Likelihood function is determined according to the degraded data of maintenance of equipment in each stage；
It maximizes to the likelihood function, calculates the drift of the degradation model in maintenance of equipment degenerative process each stage
Move coefficient and diffusion coefficient；
The degradation model is determined according to the remaining degeneration factor, the coefficient of deviation and the diffusion coefficient.
Optionally, it is described according to the degradation model predictive maintenance equipment remaining life, it specifically includes：
The amount of degradation head of maintenance of equipment is calculated up to the time of preventative maintenance threshold value according to the degradation model, obtains first
Time；
The amount of degradation head of maintenance of equipment is calculated up to the time of failure threshold according to the degradation model, obtained for the second time；
According to the first time and second time, the maintenance of equipment remaining life is predicted.
A kind of maintenance of equipment predicting residual useful life system, including：
Modeling module, for establishing the initial degradation model in each stage in maintenance of equipment degenerative process；
First acquisition module, for obtaining the degraded data of maintenance of equipment in each stage；
Parameter determination module, for determining the initial degeneration mould according to the degraded data of the maintenance of equipment in each stage
The parameter of type, obtains degradation model；
Prediction module, for according to the degradation model predictive maintenance equipment remaining life.
Optionally, it further includes：
Second acquisition module, for obtaining the degraded data of maintenance of equipment；
First judgment module, for judging whether the degraded data is less than failure threshold；
First result determining module, for when the degraded data is more than or equal to failure threshold, representing that maintenance of equipment loses
Effect；
Second judgment module is connect with first judgment module, for when the degraded data be less than failure threshold when,
Judge whether degraded data is less than repair threshold value；
Second result determining module, for when degraded data is less than repair threshold value, obtaining the degraded data of maintenance of equipment；
For when degraded data is more than or equal to repair threshold value, building the initial degradation model in maintenance of equipment degenerative process each stage.
Optionally, the parameter determination module includes：
First computing unit was degenerated for calculating the maintenance of equipment according to the degraded data of maintenance of equipment in each stage
The remaining degeneration factor of the degradation model in journey each stage；
Determination unit, for determining likelihood function according to the degraded data of maintenance of equipment in each stage；
For maximizing to the likelihood function, it is each to calculate the maintenance of equipment degenerative process for second computing unit
The coefficient of deviation and diffusion coefficient of the degradation model in stage；
Degradation model determination unit, for according to the remaining degeneration factor, the coefficient of deviation and the diffusion coefficient
Determine the degradation model.
Optionally, the prediction module includes：
First time computing unit reaches preventative dimension for calculating the amount of degradation head of maintenance of equipment according to the degradation model
The time of threshold value is repaiied, is obtained at the first time；
Second time calculating unit reaches failure threshold for calculating the amount of degradation head of maintenance of equipment according to the degradation model
Time, obtained for the second time；
Predicting unit, for according to the first time and second time, predicting the maintenance of equipment remaining life.
Compared with prior art, the present invention has following technique effect：The present invention establishes each stage in equipment degenerative process
Initial degradation model；The parameter of the initial degradation model is determined according to the degraded data of the equipment in each stage, is obtained
Degradation model；According to the pre- measurement equipment remaining life of the degradation model.The present invention fully considered maintenance to degradation ratio and
It is theoretical further to enrich the life prediction based on modeling of degenerating for the combined influence of amount of degradation so that degradation model more close to
Practical Project effectively increases the precision of life prediction, while has established solid foundation for health control activity, has wide
Wealthy application space.
Description of the drawings
It in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the present invention
Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow chart of maintenance of equipment life-span prediction method of the embodiment of the present invention；
Fig. 2 is the structure diagram of maintenance of equipment life prediction system of the embodiment of the present invention；
Fig. 3 is to consider the emulation Degradation path figure that maintenance influences；
Fig. 4 is the histogram frequency distribution diagram in service life；
Fig. 5 is the Cumulative Distribution Function figure in service life.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without making creative work
Embodiment shall fall within the protection scope of the present invention.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, it is below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
Maintenance of equipment of the present invention is the random degeneration equipment obtained after imperfect repair, and equipment performance passes through
Imperfect repair can restore to the state between " repairing as new " and " repairing as usual ", such as add lubricant, to steel mill for bearing
Wind turbine carries out dynamic balancing adjustment, and correction is adjusted to gyroscopic drift coefficient.Thus, imperfect repair with more general and
Universality.
Fig. 1 is the flow chart of maintenance of equipment life-span prediction method of the embodiment of the present invention.As shown in Figure 1, a kind of maintenance of equipment
Life-span prediction method includes the following steps：
Step 101：Establish the initial degradation model in each stage in maintenance of equipment degenerative process.
Wiener (Wiener) process has certain advantage in description degenerative process, can portray actually moving back for equipment well
Change path.The equipment that consideration maintenance influences is modeled into row degradation using Wiener processes by the present invention and durability analysis, i.e.,
Degenerative process is modeled stage by stage based on Wiener processes, the i+1 rank of life cycle is entered after i maintenance
Section, deterioration level of the equipment in the i+1 stage are represented by：
Xi(t)=ηiwp+Λ(λ,i)t+σBB(t) 0≤t≤Ti+1-Ti (1)
Wherein, Xi(t) amount of degradations of t durations is run again after undergoing ith maintenance for equipment, and i (0≤i≤N) is
The maintenance frequency that current device has been undergone, sums of the N for preventative maintenance, ηiRemaining degeneration factor after being repaired for ith is carved
Influence of the maintenance to amount of degradation is drawn, due to ηiValue in the range of (0,1), thus naturally consider ηiObey Beta points
Cloth, corresponding probability density function are represented by：
A, b (a ＞ 1, b ＞ 1) are remaining degeneration factor parameter, special, η0=0.wpPreventative maintenance threshold for equipment
Value, can generally be determined by industrial standard and expertise.Λ (λ, i) represents the coefficient of deviation after ith maintenance, for retouching
Influence of the maintenance to degradation ratio is stated, wherein, λ is the parameter of coefficient of deviation.TiFor the time of ith maintenance, Ti+1For
The time of i+1 time maintenance, σBFor diffusion coefficient, B (t) is standard Brownian movement.
According to model defined in formula (1) can be seen that remaining degeneration factor and coefficient of deviation with maintenance frequency phase
It closes, with the increase of maintenance frequency, the desired value of remaining degeneration factor gradually increases, i.e., maintenance effect is gradually deteriorated, drift system
Number can be specifically chosen corresponding form according to actual application background.
Step 102：Obtain the degraded data of maintenance of equipment in each stage.
Step 103：The parameter of the initial degradation model is determined according to the degraded data of the maintenance of equipment in each stage,
Obtain degradation model.
It specifically includes：
The degeneration mould in maintenance of equipment degenerative process each stage is calculated according to the degraded data of maintenance of equipment in each stage
The remaining degeneration factor of type；
Likelihood function is determined according to the degraded data of maintenance of equipment in each stage；
It maximizes to the likelihood function, calculates the drift of the degradation model in maintenance of equipment degenerative process each stage
Move coefficient and diffusion coefficient；
The degradation model is determined according to the remaining degeneration factor, the coefficient of deviation and the diffusion coefficient.
According to model defined in formula (1) it is found that model parameter is broadly divided into remaining degeneration factor and degradation parameter two
Major class is specifically described the process of this two classes parameter Estimation below.
It enablesRepresent s (1≤s≤M, s ∈+) platform historical Device every time repair after remaining degraded data, wherein,S represents the number of historical Device, and M is the sum of historical Device.Meanwhile remaining degeneration factor
Relationship between remaining degraded data is represented by：
Wherein,The remaining degeneration factor after i repair is undergone for s platforms equipment,I dimension is undergone for s platforms equipment
Remaining degraded data after repairing, wpFor preventative maintenance threshold value.According to formula (3), s platform historical Devices can be obtained and repaired every time
Remaining degeneration factor afterwardsWherein,From the foregoing, it will be observed that ηiObey the Beta with parameter (a, b)
Distribution selects Maximum-likelihood estimation to estimate parameter (a, b) here.Corresponding likelihood function L (a, b) is represented by：
Wherein,ForProbability density function, M be historical Device sum, N be preventative maintenance total degree, Γ
() represents Gamma functions,The remaining degeneration factor after i repair is undergone for s platforms equipment.It takes the logarithm, can obtain to formula (4)：
Wherein,The remaining degeneration factor after i repair is undergone for s platforms equipment, M is the sum of historical Device, and N is pre-
Anti- property repairs total degree, and Γ () represents Gamma functions, and (a, b) is ηiThe parameter of distribution.Pass through the log-likelihood function that maximizes
LnL (a, b) can obtain the maximum likelihood estimation of parameter (a, b), and since log-likelihood function is complex, intelligence can be used
Algorithm or multidimensional searching method carry out optimizing.
Here mainly degradation parameter is estimated using M group history degraded datas.It enablesThe observation data of s platform historical Devices are represented, due to every in life cycle
A stage be all it is independent, can sublevel piecewise analysis degenerative process.According to Wiener process fundamental propertys, in ti,j-1The degeneration at moment
It is horizontalUnder conditions of given, ti,jThe deterioration level at momentNormal Distribution, i.e.,：
Wherein,It is s platform equipment in ti,j-1The deterioration level at moment, after Λ (λ, i) represents ith maintenance
Coefficient of deviation, σBFor diffusion coefficient.Therefore,Conditional probability density be represented by：
Wherein,It is s platform equipment in ti,jThe deterioration level at moment,It is s platform equipment in ti,j-1The degeneration at moment
Level, Λ (λ, i) represent the coefficient of deviation after ith maintenance, σBFor diffusion coefficient.Further, log-likelihood function can
It is expressed as：
Wherein, M be historical Device sum, N be preventative maintenance total degree, ri sIt is s platforms equipment total in the i+1 stages
Monitor number,It is s platform equipment in ti,jThe deterioration level at moment,It is s platform equipment in ti,j-1The degeneration water at moment
Flat, Λ (λ, i) represents the coefficient of deviation after ith maintenance, σBFor diffusion coefficient.Similarly, the logarithm in maximization formula (8)
Likelihood function can obtain degradation parameter (λ, σB) maximum likelihood estimation.
Step 104：According to the degradation model predictive maintenance equipment remaining life.
It specifically includes：
The amount of degradation head of maintenance of equipment is calculated up to the time of preventative maintenance threshold value according to the degradation model, obtains first
Time；
The amount of degradation head of maintenance of equipment is calculated up to the time of failure threshold according to the degradation model, obtained for the second time；
According to the first time and second time, the maintenance of equipment remaining life is predicted.
Concept based on first-hitting time, the remaining degeneration factor η after i-th (0≤i ＜ N) secondary preventative maintenanceiGiven feelings
Under condition, the amount of degradation head of equipment reaches preventative maintenance threshold value wpTime Ri|ηiIt may be defined as：
Ri|ηi=inf (ri|Xi(ri)≥wp|ηi,ri＞ 0) (9)
Wherein, inf represents the infimum factor, Xi(ri) it is that equipment undergoes operation r againiThe amount of degradation of duration, wpFor prevention
Property repair threshold value, ηiFor degeneration factor remaining after ith maintenance.Remaining amount of degradation η after n-th preventative maintenanceNGiven
In the case of, the amount of degradation head of equipment reaches the time R of failure threshold wN|zNIt may be defined as：
RN|ηN=inf (rN|XN(rN)≥w|ηN, rN＞ 0) (10)
Wherein, inf represents the infimum factor, XN(rN) to run r again after equipment experience n-th maintenanceNDuration
Amount of degradation, w are failure threshold, ηNFor degeneration factor remaining after n-th maintenance.
According to Wiener process fundamental characteristics, two class first-hitting time R can be obtainedi|ηi(0≤i ＜ N) and RN|ηNCondition
Probability density function (ProbabilityDensityFunction, PDF).
Wherein, wpFor preventative maintenance threshold value, w is failure threshold, ηiFor degeneration factor remaining after ith maintenance, ηN
For degeneration factor remaining after n-th maintenance, Λ (λ, i) represents the coefficient of deviation after ith maintenance, Λ (λ, N) table
Show the coefficient of deviation after n-th maintenance, σBFor diffusion coefficient.According to total probability formula, R can be obtainedi(0≤i ＜ N) and RN's
Probability density function.
Wherein, fi(ri|ηi) for after ith maintenance, the amount of degradation head of equipment reaches preventative maintenance threshold value wpCondition
Time Ri|ηiProbability density function, h (ηi) probability density function for degeneration factor remaining after ith maintenance, fN(rN|
ηN) for after n-th maintenance, the amount of degradation head of equipment reaches the condition time R of failure threshold wN|ηNProbability density function, h
(ηN) probability density function for degeneration factor remaining after n-th maintenance.Based on variable defined above, can be set
Standby service life T is：
Wherein, RiAfter ith maintenance, the amount of degradation head of equipment reaches preventative maintenance threshold value wpTime, N is pre-
Anti- property repairs total degree.In order to derive the probability density function of service life T, lemma is provided here as theoretical foundation.
Lemma 1：If stochastic variable X1,X2,…,XnProbability density function be respectivelyThen stochastic variable X=X1+X2+…+XnProbability density function be：
Wherein,Respectively stochastic variable X1,X2,…,XnProbability density function, f
(x1,x2,…,xn) it is stochastic variable X1,X2,…,XnJoint probability density function.Work as X1,X2,…,XnIt, can when mutual indepedent
Further it is written as：
Wherein,Respectively stochastic variable X1,X2,…,XnProbability density function.Draw
Reason 1 can derive that specific proof procedure is repeated no more according to probability theory knowledge.
Based on lemma 1, then the service life probability density of equipment can be further represented as：
fT(t)=∫ ∫ ... ∫ f0(r0)f1(r1)…fN-1(rN-1)fN(t-r0-…-rN-1)dr0…drN-1 (18)
Wherein, f0(r0),f1(r1),…,fN(rN) it is respectively R0,R1,…,RNProbability density function.Due in formula (18)
Integral operation it is complex, the present invention is emulated using Monte-Carlo methods, is selected a fully big positive number M, is divided
Not from f0(r0)、f1(r1)…fN-1(rN-1) in sample, pth time sampled result is Wherein 0≤p≤M.This
When equipment service life probability density function can be approximately：
Wherein, M is abundant big positive number, fN(rN) it is RNProbability density function,For from fi(ri) in sample i-th
Secondary sampled result.
It is further included before the initial degradation model in each stage in described establish in maintenance of equipment degenerative process：
Obtain the degraded data of maintenance of equipment；
Judge whether the degraded data is less than failure threshold；
If not, then it represents that maintenance of equipment fails；
If so, judging whether degraded data is less than repair threshold value；
If so, perform the degraded data for obtaining maintenance of equipment.
If it is not, then build the initial degradation model in maintenance of equipment degenerative process each stage.
According to specific embodiment provided by the invention, the invention discloses following technique effects：The present invention establishes equipment and moves back
The initial degradation model in each stage during change；The initial degeneration mould is determined according to the degraded data of the equipment in each stage
The parameter of type, obtains degradation model；According to the pre- measurement equipment remaining life of the degradation model.The present invention has fully considered that repair is lived
The dynamic combined influence to degradation ratio and amount of degradation it is theoretical further to enrich the life prediction based on modeling of degenerating so that degenerate
Model effectively increases the precision of life prediction more close to Practical Project, while has established for health control activity solid
Basis, have wide application space.
Fig. 2 is the structure diagram of maintenance of equipment life prediction system of the embodiment of the present invention.As shown in Fig. 2, a kind of repair is set
Standby predicting residual useful life system, including：
Modeling module 201, for establishing the initial degradation model in each stage in maintenance of equipment degenerative process.
First acquisition module 202, for obtaining the degraded data of maintenance of equipment in each stage.
Parameter determination module 203, for determining described initially to move back according to the degraded data of the maintenance of equipment in each stage
Change the parameter of model, obtain degradation model.
The parameter determination module 203 specifically includes：
First computing unit was degenerated for calculating the maintenance of equipment according to the degraded data of maintenance of equipment in each stage
The remaining degeneration factor of the degradation model in journey each stage；
Determination unit, for determining likelihood function according to the degraded data of maintenance of equipment in each stage；
For maximizing to the likelihood function, it is each to calculate the maintenance of equipment degenerative process for second computing unit
The coefficient of deviation and diffusion coefficient of the degradation model in stage；
Degradation model determination unit, for according to the remaining degeneration factor, the coefficient of deviation and the diffusion coefficient
Determine the degradation model.
Prediction module 204, for according to the degradation model predictive maintenance equipment remaining life.
The prediction module 204 specifically includes：
First time computing unit reaches preventative dimension for calculating the amount of degradation head of maintenance of equipment according to the degradation model
The time of threshold value is repaiied, is obtained at the first time；
Second time calculating unit reaches failure threshold for calculating the amount of degradation head of maintenance of equipment according to the degradation model
Time, obtained for the second time；
Predicting unit, for according to the first time and second time, predicting the maintenance of equipment remaining life.
The system also includes：
Second acquisition module, for obtaining the degraded data of maintenance of equipment；
First judgment module, for judging whether the degraded data is less than failure threshold；
First result determining module, for when the degraded data is more than or equal to failure threshold, representing that maintenance of equipment loses
Effect；
Second judgment module is connect with first judgment module, for when the degraded data be less than failure threshold when,
Judge whether degraded data is less than repair threshold value；
Second result determining module, for when degraded data is less than repair threshold value, obtaining the degraded data of maintenance of equipment；
For when degraded data is more than or equal to repair threshold value, building the initial degradation model in maintenance of equipment degenerative process each stage.
Fig. 3 is to consider the emulation Degradation path figure that maintenance influences；Fig. 4 is the histogram frequency distribution diagram in service life；Fig. 5 is
The Cumulative Distribution Function figure in service life.
The present invention verifies the validity and reasonability of institute's extracting method using emulation experiment.To obtain the degeneration of equipment
Track, it is necessary to the suitable form of Λ (λ, i) is selected according to actual application background.Therefore, may be selected Λ (λ, i)=(i+1) λ into
Row illustrates, and without loss of generality, other forms also may be selected.Table 1 gives the setting value of model parameter, based on table 1, you can
Corresponding simulation track is obtained, Fig. 1 gives one and considers the emulation Degradation path that maintenance influences.
1 pre-set parameter of table
| a | b | λ | σB | N | wp | w | Δt |
| 2 | 5 | 2 | 0.3 | 2 | 3 | 3.5 | 0.01 |
According to fig. 3 it can be seen that maintenance has a certain impact to degradation ratio and amount of degradation, while this influence meeting
It changes with the increase of maintenance frequency.Further, under the concept of first-hitting time, the longevity of equipment can accordingly be obtained
Life.It repeats to generate 2000 simulation tracks, the histogram frequency distribution diagram in service life can be obtained, as shown in Figure 4.Obtain the frequency in service life
After rate distribution histogram, and then it can determine empirical cumulative distribution function (the Cumulative Distribution in service life
Function,CDF).Meanwhile the probability density function in service life and cumulative distribution letter can be obtained using institute's extracting method of the present invention
Number.The cumulative distribution function that Fig. 5 obtains empirical cumulative distribution function with context of methods is compared.
As shown in figure 5, differ smaller between the cumulative distribution function that empirical cumulative distribution function is obtained with context of methods.Cause
And the service life of equipment can be predicted with institute's extracting method of the present invention.For quantitative comparison simulation result and institute's extracting method of the present invention
As a result, the median in service life is selected to be compared with mean value as two Comparative indices, comparison result is as shown in table 2.
The results contrast of 2 simulation result of table and institute's extracting method
According to table 2 it is found that service life mean value and the relative error of median are respectively less than 2%, thus, this hair is further demonstrated
The validity of bright institute's extracting method.
Therefore, life-span prediction method proposed by the present invention efficiently solves the equipment life prediction influenced there are maintenance
Problem it is theoretical further to enrich the life prediction based on modeling of degenerating so that degradation model more close to Practical Project, has
Effect improves the precision of life prediction, while has established solid foundation for health control activity, has wide application space.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.For system disclosed in embodiment
For, since it is corresponded to the methods disclosed in the examples, so description is fairly simple, related part is said referring to method part
It is bright.
Specific case used herein is expounded the principle of the present invention and embodiment, and above example is said
The bright method and its core concept for being merely used to help understand the present invention；Meanwhile for those of ordinary skill in the art, foundation
The thought of the present invention, in specific embodiments and applications there will be changes.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (8)
1. a kind of maintenance of equipment method for predicting residual useful life, which is characterized in that including：
Establish the initial degradation model in each stage in maintenance of equipment degenerative process；
Obtain the degraded data of maintenance of equipment in each stage；
The parameter of the initial degradation model is determined according to the degraded data of the maintenance of equipment in each stage, obtains degeneration mould
Type；
According to the degradation model predictive maintenance equipment remaining life.
2. maintenance of equipment life-span prediction method according to claim 1, which is characterized in that moved back in the maintenance of equipment of establishing
It is further included before the initial degradation model in each stage during change：
Obtain the degraded data of maintenance of equipment；
Judge whether the degraded data is less than failure threshold；
If not, then it represents that maintenance of equipment fails；
If so, judging whether degraded data is less than repair threshold value；
If so, perform the degraded data for obtaining maintenance of equipment.
If it is not, then build the initial degradation model in maintenance of equipment degenerative process each stage.
3. life-span prediction method according to claim 1, which is characterized in that it is described according to the maintenance of equipment in each stage
Degraded data determine the parameter of the initial degradation model, obtain degradation model, specifically include：
The degradation model in each stage in the maintenance of equipment degenerative process is calculated according to the degraded data of maintenance of equipment in each stage
Remaining degeneration factor；
Likelihood function is determined according to the degraded data of maintenance of equipment in each stage；
It maximizes to the likelihood function, calculates the drift system of the degradation model in maintenance of equipment degenerative process each stage
Number and diffusion coefficient；
The degradation model is determined according to the remaining degeneration factor, the coefficient of deviation and the diffusion coefficient.
4. life-span prediction method according to claim 1, which is characterized in that described according to the degradation model predictive maintenance
Equipment remaining life, specifically includes：
The time of preventative maintenance threshold value is reached according to the amount of degradation head of degradation model calculating maintenance of equipment, when obtaining first
Between；
The amount of degradation head of maintenance of equipment is calculated up to the time of failure threshold according to the degradation model, obtained for the second time；
According to the first time and second time, the maintenance of equipment remaining life is predicted.
5. a kind of maintenance of equipment predicting residual useful life system, which is characterized in that including：
Modeling module, for establishing the initial degradation model in each stage in maintenance of equipment degenerative process；
First acquisition module, for obtaining the degraded data of maintenance of equipment in each stage；
Parameter determination module, for determining the initial degradation model according to the degraded data of the maintenance of equipment in each stage
Parameter obtains degradation model；
Prediction module, for according to the degradation model predictive maintenance equipment remaining life.
6. system according to claim 5, which is characterized in that further include：
Second acquisition module, for obtaining the degraded data of maintenance of equipment；
First judgment module, for judging whether the degraded data is less than failure threshold；
First result determining module, for when the degraded data is more than or equal to failure threshold, representing maintenance of equipment failure；
Second judgment module is connect with first judgment module, for when the degraded data is less than failure threshold, judging
Whether degraded data is less than repair threshold value；
Second result determining module, for when degraded data is less than repair threshold value, obtaining the degraded data of maintenance of equipment；For
When degraded data is more than or equal to repair threshold value, the initial degradation model in maintenance of equipment degenerative process each stage is built.
7. system according to claim 5, which is characterized in that the parameter determination module includes：
First computing unit, it is each for calculating the maintenance of equipment degenerative process according to the degraded data of maintenance of equipment in each stage
The remaining degeneration factor of the degradation model in stage；
Determination unit, for determining likelihood function according to the degraded data of maintenance of equipment in each stage；
Second computing unit for maximizing to the likelihood function, calculates maintenance of equipment degenerative process each stage
Degradation model coefficient of deviation and diffusion coefficient；
Degradation model determination unit, for being determined according to the remaining degeneration factor, the coefficient of deviation and the diffusion coefficient
The degradation model.
8. system according to claim 5, which is characterized in that the prediction module includes：
First time computing unit reaches preventative maintenance threshold for calculating the amount of degradation head of maintenance of equipment according to the degradation model
The time of value obtains at the first time；
Second time calculating unit, for according to the degradation model calculate maintenance of equipment amount of degradation head up to failure threshold when
Between, obtained for the second time；
Predicting unit, for according to the first time and second time, predicting the maintenance of equipment remaining life.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810332285.7A CN108256700A (en) | 2018-04-13 | 2018-04-13 | A kind of maintenance of equipment method for predicting residual useful life and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810332285.7A CN108256700A (en) | 2018-04-13 | 2018-04-13 | A kind of maintenance of equipment method for predicting residual useful life and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108256700A true CN108256700A (en) | 2018-07-06 |
Family
ID=62748317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810332285.7A Pending CN108256700A (en) | 2018-04-13 | 2018-04-13 | A kind of maintenance of equipment method for predicting residual useful life and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108256700A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109033569A (en) * | 2018-07-09 | 2018-12-18 | 哈尔滨理工大学 | A method of optimize for carrier-borne aircraft sensing system prophylactic repair threshold intensity and number |
| CN109059989A (en) * | 2018-07-09 | 2018-12-21 | 中国兵器工业第五九研究所 | A kind of method, system and the equipment of instrument residual life evaluation |
| CN110197288A (en) * | 2019-05-30 | 2019-09-03 | 重庆大学 | The remaining life prediction technique of equipment under the influence of failure |
| CN111667078A (en) * | 2020-06-04 | 2020-09-15 | 三一重机有限公司 | Maintenance strategy determination method and device for excavator |
| CN112541597A (en) * | 2019-09-04 | 2021-03-23 | 上海杰之能软件科技有限公司 | Multi-equipment maintenance method and device, storage medium and terminal |
| CN112883549A (en) * | 2021-01-19 | 2021-06-01 | 中国人民解放军火箭军工程大学 | Method for establishing visual condition maintenance model considering random impact influence |
| CN113033882A (en) * | 2021-03-09 | 2021-06-25 | 浙江零跑科技有限公司 | Method for predicting residual life of vehicle-mounted electronic system |
| CN114398775A (en) * | 2022-01-04 | 2022-04-26 | 交通运输部公路科学研究所 | Concrete beam service life prediction method considering cracking damage step stiffness |
-
2018
- 2018-04-13 CN CN201810332285.7A patent/CN108256700A/en active Pending
Non-Patent Citations (4)
| Title |
|---|
| YAGUO LEI 等: "Machine remaining useful life prediction using a Monte Carlo simulation method", 《COMADEM 2016》 * |
| 司小胜 等: "带测量误差的非线性退化过程建模与剩余寿命估计", 《自动化学报》 * |
| 张正新 等: "双时间尺度下的设备随机退化建模与剩余寿命预测方法", 《自动化学报》 * |
| 胡昌华 等: "不完美维护活动干预下的设备剩余寿命估计", 《中国惯性技术学报》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109059989A (en) * | 2018-07-09 | 2018-12-21 | 中国兵器工业第五九研究所 | A kind of method, system and the equipment of instrument residual life evaluation |
| CN109059989B (en) * | 2018-07-09 | 2020-11-06 | 中国兵器工业第五九研究所 | Method, system and equipment for evaluating remaining life of instrument |
| CN109033569A (en) * | 2018-07-09 | 2018-12-18 | 哈尔滨理工大学 | A method of optimize for carrier-borne aircraft sensing system prophylactic repair threshold intensity and number |
| CN109033569B (en) * | 2018-07-09 | 2021-09-17 | 哈尔滨理工大学 | Method for optimizing strength and times of preventive maintenance threshold of shipboard aircraft sensor system |
| CN110197288B (en) * | 2019-05-30 | 2023-06-02 | 重庆大学 | Method for predicting residual service life of equipment under influence of faults |
| CN110197288A (en) * | 2019-05-30 | 2019-09-03 | 重庆大学 | The remaining life prediction technique of equipment under the influence of failure |
| CN112541597A (en) * | 2019-09-04 | 2021-03-23 | 上海杰之能软件科技有限公司 | Multi-equipment maintenance method and device, storage medium and terminal |
| CN111667078A (en) * | 2020-06-04 | 2020-09-15 | 三一重机有限公司 | Maintenance strategy determination method and device for excavator |
| CN111667078B (en) * | 2020-06-04 | 2024-02-09 | 三一重机有限公司 | Maintenance strategy determination method and device for excavator |
| CN112883549A (en) * | 2021-01-19 | 2021-06-01 | 中国人民解放军火箭军工程大学 | Method for establishing visual condition maintenance model considering random impact influence |
| CN112883549B (en) * | 2021-01-19 | 2023-12-19 | 中国人民解放军火箭军工程大学 | Establishment method of condition-dependent maintenance model considering random impact influence |
| CN113033882B (en) * | 2021-03-09 | 2022-05-17 | 浙江零跑科技股份有限公司 | Method for predicting residual life of vehicle-mounted electronic system |
| CN113033882A (en) * | 2021-03-09 | 2021-06-25 | 浙江零跑科技有限公司 | Method for predicting residual life of vehicle-mounted electronic system |
| CN114398775B (en) * | 2022-01-04 | 2022-08-16 | 交通运输部公路科学研究所 | Concrete beam service life prediction method considering cracking damage step stiffness |
| CN114398775A (en) * | 2022-01-04 | 2022-04-26 | 交通运输部公路科学研究所 | Concrete beam service life prediction method considering cracking damage step stiffness |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108256700A (en) | A kind of maintenance of equipment method for predicting residual useful life and system | |
| CN103793752B (en) | A kind of equipment failure number Forecasting Methodology based on modeling of degenerating | |
| CN108520152B (en) | A kind of the service life distribution determination method and system of engineering equipment | |
| Mkandawire et al. | Transformer risk modelling by stochastic augmentation of reliability-centred maintenance | |
| Haddad et al. | Using maintenance options to maximize the benefits of prognostics for wind farms | |
| CN108549769A (en) | A kind of maintenance of equipment method for predicting residual useful life and system | |
| CN108038349A (en) | A kind of repair determining method of aircraft system health status | |
| CN106372299B (en) | A kind of pipeline method for predicting residual useful life based on inverse Gaussian process and EM-PF | |
| CN111582542B (en) | Power load prediction method and system based on anomaly repair | |
| CN107145720A (en) | It is continuous to degenerate and the unknown equipment method for predicting residual useful life impacted under collective effect | |
| Chen et al. | Imperfect maintenance policy considering positive and negative effects for deteriorating systems with variation of operating conditions | |
| Duan et al. | Optimal failure mode-based preventive maintenance scheduling for a complex mechanical device | |
| Kahrobaee et al. | A hybrid analytical-simulation approach for maintenance optimization of deteriorating equipment: Case study of wind turbines | |
| CN112785081A (en) | Oil temperature prediction method and device, electronic equipment and storage medium | |
| Khoury et al. | On the Use of Time‐Limited Information for Maintenance Decision Support: A Predictive Approach under Maintenance Constraints | |
| CN110909442B (en) | Maintenance effect modeling method considering maintenance negative effect | |
| CN113300373A (en) | Stability margin value prediction method and device based on PRMSE evaluation index | |
| CN110826881B (en) | Spacecraft on-orbit health state assessment method and system considering uncertain interference | |
| CN105426665B (en) | Method is determined based on the DYNAMIC RELIABILITY of status monitoring | |
| Aizpurua et al. | Probabilistic power transformer condition monitoring in smart grids | |
| Qian | Control-limit policy of condition-based maintenance optimization for multi-component system by means of Monte Carlo simulation | |
| CN113191506B (en) | Aperiodic condition-based maintenance method considering equipment detection uncertainty | |
| Wang et al. | Maintenance decision based on data fusion of aero engines | |
| Zhang et al. | Model and Data Driven Machine Learning Approach for Analyzing the Vulnerability to Cascading Outages With Random Initial States in Power Systems | |
| CN114330775A (en) | Monte Carlo simulation-based visual maintenance strategy optimization method for accelerated degradation equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180706 |