CN107273632B - The non-accumulated impact pectus part lifetime of system prediction technique of one kind and device - Google Patents
The non-accumulated impact pectus part lifetime of system prediction technique of one kind and device Download PDFInfo
- Publication number
- CN107273632B CN107273632B CN201710509410.2A CN201710509410A CN107273632B CN 107273632 B CN107273632 B CN 107273632B CN 201710509410 A CN201710509410 A CN 201710509410A CN 107273632 B CN107273632 B CN 107273632B
- Authority
- CN
- China
- Prior art keywords
- predicted
- component
- degradation
- emulation
- impact
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a kind of non-accumulated impact pectus part lifetime of system prediction technique and devices, this method comprises: the degraded data and simulation parameter data of S1, acquisition system to be predicted, the normal deterioration model and impact degradation model of each component in system to be predicted are determined according to degraded data；S2, it treats forecasting system and carries out degeneration emulation, according to normal deterioration model and impact degradation model, obtain the emulation amount of degradation of each component in system to be predicted in current simulation time section；If S3, judging that the emulation amount of degradation of each component is less than the corresponding degradation of each component, total degree of degeneration of the system to be predicted in current simulation time section is obtained；If S4, judging that total degree of degeneration is less than degree of degeneration threshold value, stepping is emulated, returns to S2 until thrashing to be predicted, obtains the bimetry of system to be predicted.The embodiment of the present invention is suitble to the life prediction of high reliability long-life multiple component system, improves the accuracy of multiple component system life prediction.
Description
Technical field
The present embodiments relate to reliability engineering fields, and in particular to a kind of non-accumulated impact pectus part lifetime of system
Prediction technique and device.
Background technique
It is greatly improved with advances in technology with design level, high reliability long life component starts to be applied to each neck
Domain, but for these high reliability long-life components, due to the high reliability and long-life property of component, it is difficult to estimate it
Service life, this problem are brought a lot of trouble to military industry, aerospace industry.More than military enterprise, in many industrial and commercial enterprises
Industry there is also component life it is difficult to predict problem, obtain high reliability component service life all there is huge meaning to many fields
Justice.Therefore, how component life is estimated and predicts to become as an important research contents.
The method of initial component life prediction is typically all life test, obtains certain sample and carries out life test, record
The out-of-service time of each sample is analyzed by out-of-service time of the mathematical method to collection, predicts life of product, but with reliable
Property raising, the out-of-service time of component is difficult to collect, and therefore, the life-span prediction method based on life test is no longer applicable in.Later,
It has been proposed that accelerated life test shortens the component failure time, this method does not have for general component by increasing proof stress
It is problematic, but component very high for some reliabilities still so there are problems that being difficult to collect the out-of-service time.Due to characterization portion
The performance parameter of part performance can occur gradually degenerating with the increase for using the time, the failure of system can regard as performance by
The result gradually degenerated.Therefore, in the prior art, people attempt to go to study this performance degradation process, pre- in research performance degradation
In terms of surveying the service life, the relevant personnel propose the technique study production using nonlinear regression, neural network or least square vector
Moral character energy degenerative process.But in these researchs, need to obtain historical data and real time data simultaneously, required data volume is larger and neglects
Having omited system degradation is a physical process, these methods are all no longer suitable for the life prediction of high reliability long-life component
With influencing the accuracy of multiple component system life prediction.
Therefore, how to propose a kind of scheme, the life prediction of high reliability long-life multiple component system can be suitble to, improve
The accuracy of multiple component system life prediction, becomes urgent problem to be solved.
Summary of the invention
For the defects in the prior art, the embodiment of the invention provides a kind of non-accumulated impact pectus part lifetime of system
Prediction technique and device.
On the one hand, the embodiment of the invention provides a kind of non-accumulated impact pectus part lifetime of system prediction techniques, comprising:
S1, the degraded data and simulation parameter data for obtaining system to be predicted, according to the degraded data determine it is described to
The normal deterioration model of each component and impact degradation model in forecasting system, the simulation parameter data include the system to be predicted
The degree of degeneration threshold value of the degradation of each component and the system to be predicted in system；
S2, degeneration emulation is carried out to the system to be predicted, according to the normal deterioration model and impact degradation model, obtained
Take the emulation amount of degradation of each component in the system to be predicted in current simulation time section；
If S3, judgement know that the emulation amount of degradation of each component is less than the corresponding degradation of each component, obtain
Take total degree of degeneration of the system to be predicted in the current simulation time section；
If S4, judgement know that total degree of degeneration is less than the degree of degeneration threshold value, stepping is emulated, returns to S2, directly
To the thrashing to be predicted, the bimetry of the system to be predicted is obtained.
Further, the degraded data and simulation parameter data for obtaining system to be predicted, according to the degraded data
Determine the normal deterioration model of each component in the system to be predicted, comprising:
The test normal deterioration amount that the system to be predicted is not affected by each component when impact is obtained, according to the test
Normal deterioration amount carries out curve fitting；
According to the curve-fitting results, the normal deterioration function of each component is obtained, is normally moved back according to the test
Change amount obtains the parameter of the normal deterioration function of each component using moments estimation method, obtains the normal deterioration of each component
Model.
Further, the degraded data and simulation parameter data for obtaining system to be predicted, according to the degraded data
Determine the impact degradation model of each component in the system to be predicted, comprising:
The test impact amount of degradation of each component when the system to be predicted is impacted is obtained,
It presets the impact degradation model and meets normal distribution, amount of degradation is impacted according to the test, is estimated using the square
Meter method obtains the mean value and variance of the test impact amount of degradation of the various parts to be predicted, obtains rushing for each component
Hit degradation model.
Further, the method also includes:
If judgement knows that the amount of degradation of at least one each component is greater than the corresponding degradation of each component, really
The fixed thrashing to be predicted, obtains the bimetry of the system to be predicted.
Further, the method also includes:
If judgement knows that total degree of degeneration is greater than the degree of degeneration threshold value, it is determined that the system to be predicted is lost
Effect obtains the bimetry of the system to be predicted.
Further, the simulation parameter data further include the corresponding weight of each component in the system to be predicted, accordingly
Ground, the total degree of degeneration for obtaining the system to be predicted, comprising:
According to the emulation amount of degradation of each component, the degree of degeneration of each component is obtained；
According to the degree of degeneration of each component and the corresponding weight of each component, by the degree of degeneration of each component
Total degree of degeneration as the system to be predicted of the sum of weighting.
It is further, described that degeneration emulation is carried out to the system to be predicted, comprising:
The attack time interval and simulation time step-length of the system to be predicted, Mei Gesuo are determined according to the degraded data
It states attack time interval and emulation impact is carried out to the system to be predicted, determined according to the simulation time step-length described current imitative
The true period；
Wherein, it is described according to the degraded data determine the system to be predicted attack time interval and simulation time walk
Length includes:
The impact sum that the system to be predicted is subject to when the system to be predicted is impacted is obtained, is preset described to pre-
The attack time interval index of coincidence of examining system is distributed；
The system shock time interval index of coincidence to be predicted is obtained using moments estimation method according to the impact sum
The exponential distribution parameter of distribution obtains the attack time gap model of the system to be predicted；
The attack time interval is determined according to the attack time gap model；
The annealing time for reaching the corresponding degradation of each component under each component normal deterioration is obtained, according to institute
Annealing time and the attack time interval are stated, determines the simulation time step-length.
Further, described according to the normal deterioration model and impact degradation model, it obtains in current simulation time section
The emulation amount of degradation of each component in the system to be predicted, comprising:
According to the attack time interval and the simulation time step-length, judge whether the emulation impact moment falls in described work as
In preceding simulation time section；
If the impact moment falls in the current simulation time section, according to normal deterioration model acquisition
The emulation normal deterioration amount of each component impacts amount of degradation according to the emulation that the impact degradation model obtains each component, will
The emulation amount of degradation of the sum of the emulation normal deterioration amount and the emulation impact amount of degradation as each component；
If the impact moment does not fall in the current simulation time section, institute is obtained according to the normal deterioration model
The emulation normal deterioration amount of each component is stated, and using the emulation normal deterioration amount as the emulation amount of degradation of each component.
On the other hand, the embodiment of the present invention provides a kind of non-accumulated impact pectus part lifetime of system prediction meanss, including
At least one processor；And
At least one processor being connect with the processor communication, in which:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to refer to
Order is able to carry out above-mentioned non-accumulated impact pectus part lifetime of system prediction technique.
On the one hand, the embodiment of the present invention provides a kind of non-transient computer readable storage medium, the non-transient calculating
Machine readable storage medium storing program for executing stores computer instruction, and it is more under above-mentioned non-accumulated impact that the computer instruction executes the computer
Component system life-span prediction method.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention and device, according to be predicted
The degraded data of system, establish system to be predicted normal deterioration model and impact degradation model, then to the system to be predicted into
Row degradation emulation, predicts the bimetry of the system to be predicted, improves the accuracy of multiple component system life prediction.In addition,
The embodiment of the present invention only needs the historical data or test data according to system to be predicted, does not need to carry out life test, can
It is suitble to the life prediction of high reliability long-life multiple component system.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without creative efforts, can be with root
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the flow diagram of non-accumulated impact pectus part lifetime of system prediction technique in the embodiment of the present invention；
Fig. 2 is the process signal of another non-accumulated impact pectus part lifetime of system prediction technique in the embodiment of the present invention
Figure；
Fig. 3 is that the Degradation path of component 1 in the embodiment of the present invention adds up schematic diagram；
Fig. 4 is that the Degradation path of component 2 in the embodiment of the present invention adds up schematic diagram；
Fig. 5 is that the Degradation path of component 3 in the embodiment of the present invention adds up schematic diagram；
Fig. 6 is that the Degradation path of mechanical system in the embodiment of the present invention adds up schematic diagram；
Fig. 7 is the structural schematic diagram of non-accumulated impact pectus part lifetime of system prediction meanss in the embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Fig. 1 is the flow diagram of non-accumulated impact pectus part lifetime of system prediction technique in the embodiment of the present invention, such as
Shown in Fig. 1, non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention includes:
S1, the degraded data and simulation parameter data for obtaining system to be predicted, according to the degraded data and the emulation
Supplemental characteristic determines the normal deterioration model of each component and impact degradation model, the simulation parameter number in the system to be predicted
According to including the degradation of each component and the degree of degeneration threshold value of the system to be predicted in the system to be predicted；
Specifically, the embodiment of the present invention is directed to multipart system, obtains the degraded data of system to be predicted and imitates
True supplemental characteristic, simulation parameter data may include in system to be predicted the degradation of each component and system to be predicted move back
Change degree threshold value, certainly, degraded data as needed can also include the type of other data system for example to be predicted, component
Quantity, intensity etc., the embodiment of the present invention is not especially limited.Degraded data may include in system to be predicted all parts
The amount of degradation of different time, can also include other data certainly, and the embodiment of the present invention is not specifically limited.Wherein degraded data
Acquisition with simulation parameter data can be obtained according to the historical data of the system to be predicted, or to the system to be predicted into
Row degradation test obtains degraded data, and the specific acquisition methods embodiment of the present invention is not especially limited.Get system to be predicted
Degraded data and simulation parameter data after, determined in system to be predicted according to the degraded data and simulation parameter data that get
The normal deterioration model and impact degradation model of all parts.
S2, degeneration emulation is carried out to the system to be predicted, according to the normal deterioration model and impact degradation model, obtained
Take the emulation amount of degradation of each component in the system to be predicted in current simulation time section；
Specifically, after the normal deterioration model and emulation degradation model that establish system all parts to be predicted, this is waited for
Forecasting system carries out degeneration emulation, obtains current simulation time according to the normal deterioration model of all parts and emulation degradation model
The emulation amount of degradation of each component in the interior system to be predicted of section.
If S3, judgement know that the emulation amount of degradation of each component is less than the corresponding degradation of each component, obtain
Take total degree of degeneration of the system to be predicted in the current simulation time section；
Specifically, judge whether the emulation amount of degradation of each component got is both less than the corresponding degradation of each component,
If being judged as YES, total degree of degeneration of the system to be predicted in current simulation time section is obtained, which always moves back
Change degree can indicate the amount of degradation of system to be predicted in current simulation time section.
If S4, judgement know that total degree of degeneration is less than the degree of degeneration threshold value, stepping is emulated, returns to S2, directly
To the thrashing to be predicted, the bimetry of the system to be predicted is obtained.
Specifically, after getting total degree of degeneration of system to be predicted, judge the forecasting system in current simulation time section
Interior total degree of degeneration is less than the degree of degeneration threshold value of the system to be predicted, then emulates stepping, return to above-mentioned S2, until to be predicted
Thrashing obtains the bimetry of system to be predicted.Wherein, emulation stepping refers to that simulation time increases, and returns to above-mentioned S2 and obtains
The emulation amount of degradation of each component in the system to be predicted in current simulation time section after taking simulation time increase, and continue to execute
S3-S4.It is corresponding that thrashing to be predicted refers to that the emulation amount of degradation of at least one component in system to be predicted is greater than the component
Degradation or total degree of degeneration of the system to be predicted are greater than degree of degeneration threshold value, then it represents that the thrashing to be predicted.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention further include: if judgement is known
Total degree of degeneration is greater than the degree of degeneration threshold value, it is determined that the thrashing to be predicted obtains the system to be predicted
The bimetry of system.If judgement knows that the amount of degradation of at least one each component is greater than the corresponding degeneration threshold of each component
Value, it is determined that the thrashing to be predicted obtains the bimetry of the system to be predicted.Forecasting system is treated to be moved back
Change emulation until total degree of degeneration of the system to be predicted in current simulation time section is greater than degree of degeneration threshold value, or should be to
Amount of degradation of at least one component in current simulation time section in forecasting system is greater than its corresponding degradation, it is determined that
The thrashing to be predicted, obtains the bimetry of the system to be predicted, which is pre- to this since emulation of degenerating
Examining system fails the corresponding emulation moment.
It should be noted that treating forecasting system in the embodiment of the present invention carries out degeneration emulation, judge in emulation to be predicted
System does not fail, when carrying out emulation stepping, the amount of degradation of next each component of simulation time section and previous each portion of simulation time section
The amount of degradation of part is cumulative without impacting, the embodiment of the present invention consider impact injury that system to be predicted is subject to only with currently by
The impact arrived is related, this case unrelated with the impact being subject to before, so that non-accumulated impact pectus part system lifetim is pre-
It is more accurate to survey.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention, according to system to be predicted
Degraded data, establishes the normal deterioration model and impact degradation model of system to be predicted, then degenerates to the system to be predicted
Emulation, predicts the bimetry of the system to be predicted, improves the accuracy of multiple component system life prediction.In addition, of the invention
Embodiment only needs the historical data or test data according to system to be predicted, does not need to carry out life test, can be suitble to height
The life prediction of reliability long-life multiple component system.
On the basis of the above embodiments, the degraded data and simulation parameter data for obtaining system to be predicted, according to
The degraded data determines the normal deterioration model of each component in the system to be predicted, comprising:
The test normal deterioration amount that the system to be predicted is not affected by each component when impact is obtained, according to the test
Normal deterioration amount carries out curve fitting；
According to the curve-fitting results, the normal deterioration function of each component is obtained, is normally moved back according to the test
Change amount obtains the parameter of the normal deterioration function of each component using moments estimation method, obtains the normal deterioration of each component
Model.
Specifically, degradation experiment can be carried out by treating forecasting system, or according to the historical data of the system to be predicted,
Obtain the test normal deterioration amount that system to be predicted is not affected by each component when impact.It is normally moved back according to the test of each component of acquisition
Change amount carries out curve fitting, and the normal deterioration function of all parts is obtained according to the result of curve matching.According to the test of acquisition
Normal deterioration amount.Using moments estimation method, the parameter of the normal deterioration function of each component in system to be predicted is obtained, each component is obtained
Normal deterioration model.
The specific acquisition methods of normal deterioration model are as follows:
Degradation experiment, t at the time of not impacting are carried out to the instance element that a collection of quantity is q1,t2,...,
ta,...,trThe test amount of degradation of each component is collected, t is recordedaThe test amount of degradation data at moment are as follows:
ya1,ya2...yac...yaq
A=1,2 ..., r
C=1,2 ... q
Instance element is y in the amount of degradation data for not occurring to impact the momenta1,ya2...yac...,yaq, utilize least square
Method respectively to q instance element different moments test amount of degradation data ya1,ya2...yac...,yaqIt carries out curve fitting, root
According to curve-fitting results, the formula for obtaining all parts normal deterioration function is expressed as follows:
yi=fi(t), i=1,2 ..., q
In formula, yi--- component i is not affected by test amount of degradation when impact；
fi(t) --- the expression formula of normal deterioration function when component i is not affected by impact；
T --- the time variable during each component degradation；
The quantity of q --- instance element.
Assume that the normal deterioration function of each component in system to be predicted meets same deterioration law in the embodiment of the present invention,
The normal deterioration function of i.e. each component is identical, but the parameter of the normal deterioration function of each component may be different, it is therefore desirable to root
The parameter of the normal deterioration function of each component is obtained according to the test normal deterioration amount of all parts.
According to the test amount of degradation that the degraded data of each component in the examining system to be checked of acquisition is when being not affected by impact, utilize
Moment estimation method obtains the parameters of normal deterioration function.Slope is obeyed as the linear regression of β with normal deterioration function below
For yi=fi(t)=β t, wherein β~(uβ,σβ),
Wherein,
In formula, μβ--- the mean coefficient of component linear regression；
σβ--- the coefficient of variation of component linear regression；
R --- it is divided into r moment test, carries out data collection；
ta--- the time at a-th of moment；
yac--- in taThe test amount of degradation of moment c component；
--- in taThe average test amount of degradation of all components in moment system to be predicted；
da 2--- in taThe test amount of degradation variance of all components in moment system to be predicted.
After the parameter for obtaining each component normal deterioration function, it can obtain the normal deterioration model of each component.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention, not according to system to be predicted
By the test normal deterioration amount of impact moment each component, the normal deterioration model of each component is obtained, system to be predicted is combined
Practical degraded data, improve the accuracy of normal deterioration model foundation, further improve multiple component system life prediction
Accuracy.In addition, the embodiment of the present invention only needs the historical data or test data according to system to be predicted, do not need to carry out
Life test can be suitble to the life prediction of high reliability long-life multiple component system.
On the basis of the above embodiments, the degraded data and simulation parameter data for obtaining system to be predicted, according to
The degraded data determines the impact degradation model of each component in the system to be predicted, comprising:
The test impact amount of degradation of each component when the system to be predicted is impacted is obtained,
It presets the impact degradation model and meets normal distribution, amount of degradation is impacted according to the test, is estimated using the square
Meter method obtains the mean value and variance of the test impact amount of degradation of the various parts to be predicted, obtains rushing for each component
Hit degradation model.
Specifically, degradation experiment can be carried out by treating forecasting system, or according to the historical data of the system to be predicted,
Obtain the test impact amount of degradation of each component when system to be predicted is impacted.The embodiment of the present invention assumes rushing for system to be predicted
It hits amount of degradation and meets normal distribution, the impact amount of degradation for setting system to be predicted as needed certainly meets other points
Cloth such as exponential distribution, Poisson distribution etc., the embodiment of the present invention is not specifically limited.It is moved back according to the test of each component of acquisition impact
Change amount obtains the mean value and variance of the test impact amount of degradation of various parts to be predicted using moments estimation method, thus can be with
Obtain the impact degradation model of the various parts to be predicted.
The embodiment of the present invention impacts degradation model foundation, and the specific method is as follows:
Degradation experiment is carried out to the instance element that a collection of quantity is q, [0, tr] period by impact at the time of t1,
t2,…,tb,…,tl(tlIt is [0, tr] last time impact in section time) test of collecting each component impacts amount of degradation.
Measure tbMoment, amount of degradation data were impacted in each component test are as follows:
zb1,zb2...zbc...zbq
B=1,2 ... l
C=1,2 ... q
According to impact moment t1,t2,…,tb,…,tlAmount of degradation data z is impacted in the test of each componentb1,zb2,...,zbq,b
=1,2 ... l obtains the mean value and variance of the test impact amount of degradation of various parts to be predicted using moments estimation method:
Wherein,
In formula, tb--- b-th of moment；
Q --- q component；
The parameter of exponential distribution is obeyed (when the embodiment of the present invention pre-supposes that the impact of emulation in λ --- attack time interval
Between interval index of coincidence distribution)；
uβ--- component linear regression coefficient；
L --- number of shocks；
--- in tbThe average test amount of degradation of all components in moment system to be predicted；
zbc--- tbMoment, the test degeneration total amount of c-th of component；
sb 2--- in tbThe variance of the degeneration total amount of all components in moment system to be predicted.
The impact degradation model of each component can be obtained.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention, according to system to be predicted by
Amount of degradation is impacted in test to impact moment each component, obtains the impact degradation model of each component, combines system to be predicted
Practical degraded data improves the accuracy of normal deterioration model foundation, further improves multiple component system life prediction
Accuracy.In addition, the embodiment of the present invention only needs the historical data or test data according to system to be predicted, do not need to carry out the longevity
Life test, can be suitble to the life prediction of high reliability long-life multiple component system.
On the basis of the above embodiments, the simulation parameter data further include each component correspondence in the system to be predicted
Weight, correspondingly, the total degree of degeneration for obtaining the system to be predicted, comprising:
According to the emulation amount of degradation of each component, the degree of degeneration of each component is obtained；
According to the degree of degeneration of each component and the corresponding weight of each component, by the degree of degeneration of each component
Total degree of degeneration as the system to be predicted of the sum of weighting.
Specifically, when the emulation amount of degradation for judging various parts to be predicted is both less than its corresponding degradation, then
The total degree of degeneration for obtaining system to be predicted, further judges whether the system to be predicted fails.Wherein, system to be predicted is total
The acquisition methods of degree of degeneration are: according to the emulation amount of degradation of component each in system to be predicted, obtaining each portion in system to be predicted
The degree of degeneration of part, specifically can be using the ratio of the emulation amount of degradation degradation corresponding with each component of each component as each portion
The degree of degeneration of part, then the weight of the degree of degeneration of each component and all parts in system to be predicted is weighted summation,
Obtain total degree of degeneration of system to be predicted.The specific method is as follows:
Dimensionless processing is done to the emulation amount of degradation of each component, obtains the degree of degeneration of all parts, i.e., each component
Degree of degeneration indicates are as follows:
System to be predicted is equal to the degree of degeneration weighted sum of all parts in total degree of degeneration of moment t+ Δ t, it may be assumed that
In formula, Zi(emulation amount of degradation of t+ Δ t) --- the component i in moment t+ Δ t；
K1,K2,…Km--- the degradation of each component；
F(Z1(t+Δt),Z2(t+Δt),…ZmTotal degeneration journey of (t+ Δ t)) --- the system to be predicted in moment t+ Δ t
Degree；
w1,w2,…wm--- weight of the emulation amount of degradation of each component in system to be predicted, experience gained.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention, according in system to be predicted
The emulation amount of degradation of each component obtains total degree of degeneration of system to be predicted using weighted sum method, it is contemplated that different components
Degree of degeneration total degree of degeneration for treating forecasting system influence it is different, the total degree of degeneration for improving system to be predicted obtains
Accuracy, further improve the accuracy of multiple component system life prediction.In addition, the embodiment of the present invention only need according to
The historical data or test data of forecasting system do not need to carry out life test, can be suitble to high reliability long-life multi-part
System lifetim prediction.
It is on the basis of the above embodiments, described that degeneration emulation is carried out to the system to be predicted, comprising:
The attack time interval and simulation time step-length of the system to be predicted, Mei Gesuo are determined according to the degraded data
It states attack time interval and emulation impact is carried out to the system to be predicted, determined according to the simulation time step-length described current imitative
The true period；
Wherein, it is described according to the degraded data determine the system to be predicted attack time interval and simulation time walk
Length includes:
The impact sum that the system to be predicted is subject to when the system to be predicted is impacted is obtained, is preset described to pre-
The attack time interval index of coincidence of examining system is distributed；
The system shock time interval index of coincidence to be predicted is obtained using moments estimation method according to the impact sum
The exponential distribution parameter of distribution obtains the attack time gap model of the system to be predicted；
According to attack time interval described in the attack time gap model；
The annealing time for reaching the corresponding degradation of each component under each component normal deterioration is obtained, according to institute
Annealing time and the attack time interval are stated, determines the simulation time step-length.
Specifically, the embodiment of the present invention treat forecasting system carry out degenerate emulation when, according to the system to be predicted got
Degraded data determine attack time interval and simulation time step-length, attack time interval refer to carry out every time Impact Simulation when
Between time interval between the time of last Impact Simulation, simulation time step-length refers to that emulation stepping is increased imitative every time
The true time.One-shot emulation is carried out every the attack time interval according to determining attack time interval, and when according to emulation
Between step-length determine current simulation time section, current simulation time section be current emulation moment to the current emulation moment plus emulating when
Between the step-length corresponding period, such as current emulation moment is t, and simulation time step-length is Δ t, then current simulation time section is [t, t
+Δt]。
Wherein, according to degraded data determine system to be predicted attack time interval and simulation time step-length can be using such as
Lower method:
The embodiment of the present invention presets attack time interval index of coincidence and is distributed Xj~exp (λ) presets emulation number of shocks symbol
It closes Poisson distribution N (t)~p (λ t).The impact sum for obtaining system to be predicted system to be predicted when receiving impact, specifically may be used
To be obtained by test.Moments estimation method is utilized according to the impact sum of acquisition, between the attack time that can determine system to be predicted
Every meeting the corresponding exponential distribution parameter lambda of exponential distribution, that is, obtain the attack time gap model of system to be predicted.Certain basis
It needs to preset the attack time interval and meets other distribution such as Poisson distributions, rushed further according to the system to be predicted of acquisition
Time interval model is hit, the corresponding attack time interval of system to be predicted can be obtained.The wherein tool of attack time gap model
Body expression way is as follows:
Xj=exp rnd (1/ λ), j=1,2 ...
In formula, Xj--- the interval time of j-th of Impact Simulation, i.e. j-th Impact Simulation and -1 Impact Simulation of jth it
Between time interval；
λ --- the exponential distribution parameter of attack time interval obedience exponential distribution.
Behind the attack time interval for getting emulation of degenerating, obtains in system to be predicted and reach this under each component normal deterioration
The annealing time of component degradation threshold value determines the emulation for emulation of degenerating according to the annealing time at attack time interval and each component
Time step determines that the specific method is as follows for simulation time step-length:
Obtain the annealing time for reaching the component degradation threshold value in system to be predicted under each component normal deterioration, specific component
Reach the annealing time of degradation under i normal deterioration, expression way is as follows:
Hi=f-1(Ki)
In formula: Hi--- (i.e. normal deterioration) reaches its corresponding degradation K in the case that component i is not affected by impacti's
Annealing time；
Ki--- the degradation of component i.
The calculation method of simulation time step-length is as follows:
N is positive integer, adjusts the size of n, so that:
Obtain Δ t ':
Simulation step length Δ t are as follows:
Δ t=Δ t '-Δ t ' mod50
In formula, n --- segmentation HiInterval number, the size of n is adjusted, to find out Δ t；
Xj--- the attack time interval of jth Secondary Shocks.
--- it is rounded symbol downwards；
Δ t ' mod50 --- Δ t ' divided by 50 remainder；
Δ t --- simulation step length.
Non-accumulated impact pectus part lifetime of system prediction technique provided in an embodiment of the present invention, according to system to be predicted
Degraded data determines the attack time interval and simulation time step-length for emulation of degenerating, and further treats forecasting system and degenerate
Emulation, obtains the bimetry of system to be predicted, has fully considered the practical data degenerated of system to be predicted, improved to be predicted
The accuracy of the degeneration emulation of system, further improves the accuracy of multiple component system life prediction.In addition, the present invention is implemented
Example only needs the historical data or test data according to system to be predicted, does not need to carry out life test, can be suitble to highly reliable
The life prediction of property long-life multiple component system.
On the basis of the above embodiments, described according to the normal deterioration model and impact degradation model, it obtains current
In simulation time section in the system to be predicted each component emulation amount of degradation, comprising:
According to the attack time interval and the simulation time step-length, judge whether the emulation impact moment falls in described work as
In preceding simulation time section；
If the impact moment falls in the current simulation time section, according to normal deterioration model acquisition
The emulation normal deterioration amount of each component impacts amount of degradation according to the emulation that the impact degradation model obtains each component, will
The emulation amount of degradation of the sum of the emulation normal deterioration amount and the emulation impact amount of degradation as each component；
If the impact moment does not fall in the current simulation time section, institute is obtained according to the normal deterioration model
The emulation normal deterioration amount of each component is stated, and using the emulation normal deterioration amount as the emulation amount of degradation of each component.
Specifically, treat forecasting system carry out degenerate emulation when, judge emulation impact the moment whether fall in current emulation
In period, i.e., according to attack time interval and simulation time step-length in above-described embodiment, judge in current simulation time section
Whether [t, t+ Δ t] has emulation to impact.If judging that emulation impact has occurred in current simulation time section, basis is normally moved back
Change the emulation normal deterioration amount that model obtains each component in system to be predicted, is obtained in system to be predicted according to impact degradation model
Amount of degradation is impacted in the emulation of each component, by the sum of the emulation normal deterioration amount of each component and emulation impact amount of degradation as to be predicted
The emulation amount of degradation of each component in system.If judging, there is no emulation to impact in current simulation time section, and basis is normally moved back
Change the emulation normal deterioration amount that model obtains each component in system to be predicted, using the emulation normal deterioration amount as system to be predicted
In each component emulation amount of degradation.Whether the emulation amount of degradation of each component is greater than each component in judging the system to be predicted obtained
Corresponding degradation, then carry out corresponding movement in next step.
The specific acquisition methods of the emulation amount of degradation of system to be predicted are as follows:
If judgement impact is fallen in [t, t+ Δ t] this current simulation time section, then emulation of the component i in moment t+ Δ t
Amount of degradation expression formula is as follows:
Zi(t+ Δ t)=f (t+ Δ t)+Wij
In formula: Zi(emulation amount of degradation of t+ Δ t) --- the component i in moment t+ Δ t；
F (t+ Δ t) --- component i moment t+ Δ t normal deterioration amount, according to normal deterioration model in above-described embodiment
It obtains；
Wij--- component i is impacted amount of degradation by caused by jth Secondary Shocks, according to impacting degradation model in above-described embodiment
It obtains；
Δ t --- simulation time step-length.
If judgement impact is not fallen in [t, t+ Δ t] this current simulation time section, then component i is in the imitative of moment t+ Δ t
True amount of degradation expression formula is as follows:
Zi(t+ Δ t)=f (t+ Δ t)
In formula: in formula: Zi(emulation amount of degradation of t+ Δ t) --- the component i in moment t+ Δ t；
F (t+ Δ t) --- component i moment t+ Δ t normal deterioration amount, according to normal deterioration model in above-described embodiment
It obtains；
Δ t --- simulation time step-length.
After getting the emulation amount of degradation of each component in system to be predicted, judge in the current simulation time section obtained to pre-
The emulation amount of degradation of each component degradation whether corresponding all less than each component in examining system.If both less than, according to upper
The method for stating embodiment offer, obtains total degree of degeneration of system to be predicted in current simulation time section, then judges total degeneration
Whether degree is less than degree of degeneration threshold value.If being less than, emulation stepping is carried out, i.e., simulation time is progressive, and determines next imitative
The true impact moment, and amount of degradation is emulated using each component of t moment emulation normal deterioration amount as each component, concrete operations are as follows:
The moment is impacted in next emulation:
X=X+Xj=X+exp rnd (1/ λ)
Time is progressive:
T=t+ Δ t
Emulate normal deterioration amount:
Zi(t)=f (t)
In formula, X --- the moment is impacted in next emulation at j-th of emulation impact moment；
Xj--- the attack time interval at j-th of emulation impact moment and next emulation impact moment；
λ --- the parameter of attack time interval obedience exponential distribution；
Zi(t) --- emulation amount of degradation of the component i in moment t；
F (t) --- component i is in the function expression of moment t emulation normal deterioration amount, i.e. normal deterioration model.
As can be seen that, if system to be predicted does not fail, being carried out in the embodiment of the present invention after step-by-step simulation
When emulating stepping, the emulation amount of degradation of each component is all emulation normal deterioration amount in the system to be predicted, and degeneration next time is imitative
Whether true only basis includes that emulation is impacted when secondary emulation of degenerating, if not including, emulating amount of degradation is to emulate normal deterioration amount.
I.e. without the accumulative of emulation impact amount of degradation, it is believed that one-shot is degenerated caused by the system to be predicted, as long as not causing
The failure of the system does not impact the emulation degeneration in the system later period to be predicted then.
Fig. 2 is the process signal of another non-accumulated impact pectus part lifetime of system prediction technique in the embodiment of the present invention
Figure specifically introduces the realization of the non-accumulated impact pectus part lifetime of system prediction technique of the embodiment of the present invention below with reference to Fig. 2
Journey:
Simulation time t is set to 0, number of shocks j and is set to 1 by simulation initialisation, emulation impact moment X is set to 0, and
The emulation amount of degradation Z of i-th of component in initial time system to be predictedi(0) it is set to 0.According to test data, normal deterioration is determined
Model and impact degradation model, then treat forecasting system and carry out degeneration emulation, determine attack time interval.When obtaining current emulation
Between in section each component emulation normal deterioration amount, judge whether the emulation impact moment falls in current simulation time section [t, t+ Δ t]
Interior, i.e. whether t < X < t+ Δ t is true, if so:
Amount of degradation is impacted in the emulation for successively obtaining each component in current simulation time section, by the emulation normal deterioration of each component
The emulation amount of degradation of the sum of amount and emulation impact amount of degradation as each component.Judge whether the emulation amount of degradation of each component is less than respectively
The corresponding degradation of component, if there is the emulation amount of degradation of at least one component to be greater than or equal to the corresponding degeneration threshold of the component
Value, it is determined that thrashing to be predicted, the bimetry for exporting system to be predicted is T=t+ Δ t.If the emulation of each component is degenerated
Amount is respectively less than the corresponding degradation of the component, then obtains system to be predicted according to the emulation amount of degradation of each component and emulate currently
Total degree of degeneration in period, judges whether total degree of degeneration is less than degree of degeneration threshold value, if being not less than, emulates step
Into: determine next emulation impact moment X=X+Xj, number of shocks j=j+1, simulation time t=t+ Δ t are emulated, is obtained next imitative
The emulation amount of degradation etc. of various parts to be predicted of true time.If total degeneration journey of the system to be predicted in current simulation time section
Degree is greater than degree of degeneration threshold value, it is determined that thrashing to be predicted, the bimetry for exporting system to be predicted is T=t+ Δ t.
If the judgement emulation impact moment does not fall in current simulation time section [t, t+ Δ t], then:
Successively using the emulation normal deterioration amount of each component as the emulation amount of degradation of each component, and judge the emulation of each component
Whether amount of degradation is less than the corresponding degradation of each component, if there is the emulation amount of degradation of at least one component to be greater than or equal to the portion
The corresponding degradation of part, it is determined that thrashing to be predicted, the bimetry for exporting system to be predicted is T=t+ Δ t.If each
The emulation amount of degradation of component is respectively less than the corresponding degradation of the component, then is obtained according to the emulation amount of degradation of each component to be predicted
Total degree of degeneration of the system in current simulation time section, judges whether total degree of degeneration is less than degree of degeneration threshold value, if not
It is less than, then emulates stepping: determining simulation time t=t+ Δ t, the emulation for obtaining next simulation time various parts to be predicted is moved back
Change amount etc..If total degree of degeneration of the system to be predicted in current simulation time section is greater than degree of degeneration threshold value, it is determined that pre-
Examining system failure, the bimetry for exporting system to be predicted is T=t+ Δ t.
Detailed process is not as shown in Fig. 2, the embodiment of the present invention is further elaborated.
Below with reference to specific case, the concrete scheme of the embodiment of the present invention is introduced:
The mechanical system of certain model is made of 3 components, and 1,2 amount of degradation of component shows as abrasion loss, wherein component 1
Performance degradation shows as that abrasion loss is bigger, and braking ability is poorer；It is bigger that 2 performance degradation of component shows as abrasion loss, sealing performance
Also poorer.The amount of degradation of component 3 shows as crack length, and crack length has an impact to the intensity of mechanical part, shows as splitting
Line is longer, and the intensity of component is lower.Wherein, m=3, according to the degradation of component each known to the degraded data of the mechanical system
It is respectively such as the following table 1 with weight:
The degradation and weight detail list of each component of table 1
K in table 1iIndicate the degradation of each component, KNIndicate the degree of degeneration threshold value of mechanical system, ωiIndicate each component
Weight.The step of utilizing simulation algorithm bimetry to above-mentioned mechanical system is as follows:
Step 1: the related data in degeneration simulation process is collected, comprising: collect in certain period of time by degradation experiment
Amount of degradation is impacted in the test normal deterioration amount of each component and test, the degradation K of all parts in multiple component systemi, i=1,
2,3 ... and system entirety degradation KN.To the test normal deterioration amount curve matching of collection, and obtain all parts
Degradation model:
(1) the multiple component system normal deterioration process, which follows linear regression process, to be obtained to test normal deterioration amount fitting,
Therefore, the normal deterioration function of each component is obtained are as follows:
F (t)=βit
The embodiment of the present invention assumes that the impact degeneration of each component meets normal distribution.
(2) parameter Estimation
According to experimental data, the parameter of normal deterioration model is obtained using moments estimation method, impacts the parameter of degradation model such as
The following table 2:
The parameter detail list of 2 normal deterioration model of table and impact degradation model
Step 2: simulation initialisation processing determines emulation impact moment and simulation time step delta t；
In the embodiment of the present invention, simulation time step-length is calculated equal to Δ t=1250h
Step 3: judge whether the emulation impact moment falls into [t, t+ Δ t] this time interval；
Step 4: all parts are calculated in moment t+ Δ t] degeneration total amount, if having impact in [t, t+ Δ t], respectively
The emulation amount of degradation of component is equal to the sum of emulation impact amount of degradation caused by the emulation normal deterioration amount and impact of each component, if
Without impact in [t, t+ Δ t], the emulation amount of degradation of each component is equal to the emulation normal deterioration amount of each component.And judge all components
Emulation amount of degradation whether be both less than corresponding degradation, if it exists either component emulation amount of degradation be more than degradation, then
Thrashing, output system service life T=t+ Δ t；If all component emulation amount of degradations are respectively less than degradation, step is considered
Five；
(1) judgement impact falls in [t, t+ Δ t] this time interval
Emulation amount of degradation expression formula of the component i in time t+ Δ t is as follows:
Z1(t+ Δ t)=β1(t+Δt)+W1j=8.4823 × 10-8(t+Δt)+1×10-3
Z2(t+ Δ t)=β2(t+Δt)+W2j=8.4936 × 10-9(t+Δt)+0.9×10-4
Z3(t+ Δ t)=β3(t+Δt)+W3j=8.6 × 10-7(t+Δt)+0.011
(2) at [t, t+ Δ t], this time interval does not impact
Then component i is as follows in the emulation amount of degradation expression formula of time t+ Δ t:
Z1(t+ Δ t)=β1(t+ Δ t)=8.4823 × 10-8(t+Δt)
Z2(t+ Δ t)=β2(t+ Δ t)=8.4936 × 10-9(t+Δt)
Z3(t+ Δ t)=β3(t+ Δ t)=8.6 × 10-7(t+Δt)
Step 5: total degree of degeneration of system is calculated, and judges whether total degree of degeneration is less than the degree of degeneration of system
Threshold value, if total degree of degeneration of system is more than the degree of degeneration threshold value of system, thrashing, output system service life T=t+ Δ t.
If total degree of degeneration is less than degree of degeneration threshold value, step 6 is considered；
System is as follows in the degree of degeneration in time t+ Δ t:
Step 6: emulation stepping calculates the emulation impact moment of next non-accumulated impact, and the emulation of each component is degenerated
Amount is reduced to emulate normal deterioration amount caused by the normal deterioration of each component, and the time is progressive, repetition step 3 to step 6, until
Thrashing, output system service life T=t+ Δ t, emulation terminate.
At the time of next impact:
X=X+Xj=X+exp rnd (1/ λ)
Time is progressive:
T=t+ Δ t
The emulation normal deterioration amount of time t moment:
Z1(t)=β1T=8.4823 × 10-8t
Z2(t)=β2T=8.4936 × 10-9t
Z3(t)=β3T=8.6 × 10-7t
Fig. 3 is that the Degradation path of component 1 in the embodiment of the present invention adds up schematic diagram, and Fig. 4 is component 2 in the embodiment of the present invention
Degradation path add up schematic diagram, Fig. 5 be the embodiment of the present invention in component 3 Degradation path add up schematic diagram, Fig. 6 be the present invention
The Degradation path of mechanical system adds up schematic diagram in embodiment, as shown in figures 3 to 6, under the impact of non-accumulated Poisson, works as system
It works to 1.3255 × 105The h moment receives impact, damages after causing the amount of degradation of component 1 and component 2 to reach degradation
It is bad.For component 3, before impacting, system is due to linear degenerative process, 1.245 × 105The h moment
Degenerate to degradation.The degree of degeneration of whole system is 1.279 × 105The h moment degenerates to degree of degeneration threshold value, and system will be by
Reach degradation in the amount of degradation of component 3 and damage, then the failure start time of system is 1.245 × 105The h moment.So
For not repairable system, it is assumed that the bimetry of system is 1.245 × 10 under the operating condition of parameter5h。
Fig. 7 is the structural schematic diagram of non-accumulated impact pectus part lifetime of system prediction meanss in the embodiment of the present invention, such as
Shown in Fig. 7, the apparatus may include: processor (processor) 71, memory (memory) 72 and communication bus 73,
In, processor 71, memory 72 completes mutual communication by communication bus 73.Processor 71 can call in memory 72
Logical order, to execute following method: S1, the degraded data for obtaining system to be predicted determine institute according to the degraded data
The normal deterioration model and impact degradation model of each component in system to be predicted are stated, the degraded data includes the system to be predicted
The degree of degeneration threshold value of the degradation of each component and the system to be predicted in system；S2, the system to be predicted is moved back
Change emulation, according to the normal deterioration model and impact degradation model, obtains the system to be predicted in current simulation time section
In each component emulation amount of degradation；If it is corresponding that S3, judgement know that the emulation amount of degradation of each component is less than each component
Degradation then obtains total degree of degeneration of the system to be predicted in the current simulation time section；If S4, judgement are known
Total degree of degeneration is less than the degree of degeneration threshold value, then emulates stepping, returns to S2, until the thrashing to be predicted,
Obtain the bimetry of the system to be predicted.
In addition, the logical order in above-mentioned memory 72 can be realized and as only by way of SFU software functional unit
Vertical product when selling or using, can store in a computer readable storage medium.Based on this understanding, this hair
Substantially the part of the part that contributes to existing technology or the technical solution can be with soft in other words for bright technical solution
The form of part product embodies, which is stored in a storage medium, including some instructions are to make
It obtains a computer equipment (can be personal computer, server or the network equipment etc.) and executes each embodiment of the present invention
The all or part of the steps of the method.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is various
It can store the medium of program code.
The embodiment of the present invention provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage
Medium storing computer instruction, the computer instruction make the computer execute side provided by above-mentioned each method embodiment
Method, for example, S1, the degraded data for obtaining system to be predicted are determined in the system to be predicted according to the degraded data
The normal deterioration model of each component and impact degradation model, the degraded data include in the system to be predicted each component move back
Change the degree of degeneration threshold value of threshold value and the system to be predicted；S2, degeneration emulation is carried out to the system to be predicted, according to described
Normal deterioration model and impact degradation model, the emulation for obtaining each component in the system to be predicted in current simulation time section are moved back
Change amount；If S3, judgement know that the emulation amount of degradation of each component is less than the corresponding degradation of each component, institute is obtained
State total degree of degeneration of the system to be predicted in the current simulation time section；If S4, judgement know that total degree of degeneration is small
In the degree of degeneration threshold value, then stepping is emulated, return to S2, until the thrashing to be predicted, obtains the system to be predicted
The bimetry of system.
The above examples are only used to illustrate the technical scheme of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these are modified or replace
It changes, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (7)
1. a kind of non-accumulated impact pectus part lifetime of system prediction technique characterized by comprising
S1, the degraded data and simulation parameter data for obtaining system to be predicted, determine described to be predicted according to the degraded data
The normal deterioration model of each component and impact degradation model in system, the simulation parameter data include in the system to be predicted
The degree of degeneration threshold value of the degradation of each component and the system to be predicted；
S2, degeneration emulation is carried out to the system to be predicted, according to the normal deterioration model and impact degradation model, acquisition is worked as
In preceding simulation time section in the system to be predicted each component emulation amount of degradation；
If S3, judgement know that the emulation amount of degradation of each component is less than the corresponding degradation of each component, institute is obtained
State total degree of degeneration of the system to be predicted in the current simulation time section；
If S4, judgement know that total degree of degeneration is less than the degree of degeneration threshold value, stepping is emulated, returns to S2, until institute
Thrashing to be predicted is stated, the bimetry of the system to be predicted is obtained；
Wherein, the degraded data and simulation parameter data for obtaining system to be predicted, according to degraded data determination
The normal deterioration model of each component in system to be predicted, comprising:
The test normal deterioration amount that the system to be predicted is not affected by each component when impact is obtained, it is normal according to the test
Amount of degradation carries out curve fitting；
According to the curve-fitting results, the normal deterioration function of each component is obtained, according to the test normal deterioration amount,
Using moments estimation method, the parameter of the normal deterioration function of each component is obtained, obtains the normal deterioration model of each component；
Wherein, the degraded data and simulation parameter data for obtaining system to be predicted, according to degraded data determination
The impact degradation model of each component in system to be predicted, comprising:
Obtain the test impact amount of degradation of each component when the system to be predicted is impacted；
It presets the impact degradation model and meets normal distribution, amount of degradation is impacted according to the test, using the moments estimation method,
The mean value and variance of the test impact amount of degradation of the various parts to be predicted are obtained, the impact for obtaining each component is degenerated
Model；
Wherein, described according to the normal deterioration model and impact degradation model, it obtains described to pre- in current simulation time section
The emulation amount of degradation of each component in examining system, comprising:
According to the attack time interval and the simulation time step-length, it is described current imitative to judge whether the emulation impact moment falls in
In the true period；
If the impact moment falls in the current simulation time section, each portion is obtained according to the normal deterioration model
The emulation normal deterioration amount of part impacts amount of degradation according to the emulation that the impact degradation model obtains each component, will be described
Emulate the emulation amount of degradation of the sum of normal deterioration amount and the emulation impact amount of degradation as each component；
If the impact moment does not fall in the current simulation time section, obtained according to the normal deterioration model described each
The emulation normal deterioration amount of component, and using the emulation normal deterioration amount as the emulation amount of degradation of each component.
2. the method according to claim 1, wherein the method also includes:
If judgement knows that the amount of degradation of at least one each component is greater than the corresponding degradation of each component, it is determined that institute
Thrashing to be predicted is stated, the bimetry of the system to be predicted is obtained.
3. the method according to claim 1, wherein the method also includes:
If judgement knows that total degree of degeneration is greater than the degree of degeneration threshold value, it is determined that the thrashing to be predicted obtains
Take the bimetry of the system to be predicted.
4. the method according to claim 1, wherein the simulation parameter data further include the system to be predicted
In the corresponding weight of each component, correspondingly, the total degree of degeneration for obtaining the system to be predicted, comprising:
According to the emulation amount of degradation of each component, the degree of degeneration of each component is obtained；
According to the degree of degeneration of each component and the corresponding weight of each component, by adding for the degree of degeneration of each component
Total degree of degeneration of the sum of the power as the system to be predicted.
5. method according to claim 1-4, which is characterized in that described to degenerate to the system to be predicted
Emulation, comprising:
The attack time interval and simulation time step-length that the system to be predicted is determined according to the degraded data, every the punching
It hits time interval and emulation impact is carried out to the system to be predicted, when determining the current emulation according to the simulation time step-length
Between section；
Wherein, the attack time interval and simulation time step-length packet that the system to be predicted is determined according to the degraded data
It includes:
The impact sum that the system to be predicted is subject to when the system to be predicted is impacted is obtained, the system to be predicted is preset
The attack time interval index of coincidence of system is distributed；
The system shock time interval index of coincidence distribution to be predicted is obtained using moments estimation method according to the impact sum
Exponential distribution parameter, obtain the attack time gap model of the system to be predicted；
The attack time interval is determined according to the attack time gap model；
The annealing time for reaching the corresponding degradation of each component under each component normal deterioration is obtained, is moved back according to described
Change time and the attack time interval, determines the simulation time step-length.
6. a kind of non-accumulated impact pectus part lifetime of system prediction meanss characterized by comprising
At least one processor；And
At least one processor being connect with the processor communication, in which:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to instruct energy
It is enough to execute such as method described in any one of claim 1 to 5.
7. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited
Computer instruction is stored up, the computer instruction makes the computer execute such as method described in any one of claim 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710509410.2A CN107273632B (en) | 2017-06-28 | 2017-06-28 | The non-accumulated impact pectus part lifetime of system prediction technique of one kind and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710509410.2A CN107273632B (en) | 2017-06-28 | 2017-06-28 | The non-accumulated impact pectus part lifetime of system prediction technique of one kind and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107273632A CN107273632A (en) | 2017-10-20 |
| CN107273632B true CN107273632B (en) | 2019-09-27 |
Family
ID=60070334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710509410.2A Active CN107273632B (en) | 2017-06-28 | 2017-06-28 | The non-accumulated impact pectus part lifetime of system prediction technique of one kind and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107273632B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104820781A (en) * | 2015-05-06 | 2015-08-05 | 北京航空航天大学 | Prediction method of thermal fatigue life of BGA (Ball Grid Array) welding spot considering influence of sequential temperature cycling load loading |
| CN106372272A (en) * | 2016-08-01 | 2017-02-01 | 北京航空航天大学 | Lithium battery capacity and service life prediction method based on generalized degradation model and multiscale analysis |
| CN106528951A (en) * | 2016-10-18 | 2017-03-22 | 张家港莫特普数据科技有限公司 | Life prediction and safety warning methods and apparatuses for power battery |
-
2017
- 2017-06-28 CN CN201710509410.2A patent/CN107273632B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104820781A (en) * | 2015-05-06 | 2015-08-05 | 北京航空航天大学 | Prediction method of thermal fatigue life of BGA (Ball Grid Array) welding spot considering influence of sequential temperature cycling load loading |
| CN106372272A (en) * | 2016-08-01 | 2017-02-01 | 北京航空航天大学 | Lithium battery capacity and service life prediction method based on generalized degradation model and multiscale analysis |
| CN106528951A (en) * | 2016-10-18 | 2017-03-22 | 张家港莫特普数据科技有限公司 | Life prediction and safety warning methods and apparatuses for power battery |
Non-Patent Citations (2)
| Title |
|---|
| A novel correlative model of failure mechanisms for evaluating MEMS devices reliability;Yaqiu Li 等;《Microelectronics Reliability》;20161231;第669-675页 * |
| Reliability Analysis of Phased-Mission Complex Systems for Warship;Chen Zhiwei 等;《2016 Prognostics and System Health Management Conference》;20161231;第1-6页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107273632A (en) | 2017-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109061504A (en) | Same type difference lithium ion battery remaining life prediction technique and system | |
| CN110348615B (en) | Cable line fault probability prediction method based on ant colony optimization support vector machine | |
| CN111523778A (en) | Power grid operation safety assessment method based on particle swarm algorithm and gradient lifting tree | |
| WO2018112818A1 (en) | Rapid prediction method for cycle life of battery and rapid prediction device therefor | |
| Doagou‐Rad et al. | Leading edge erosion of wind turbine blades: Multiaxial critical plane fatigue model of coating degradation under random liquid impacts | |
| CN111275252A (en) | Blasting peak value speed prediction method, device and medium based on RGA-BPNNG | |
| CN112288147B (en) | Method for predicting insulation state of generator stator by BP-Adaboost strong predictor | |
| CN106501465A (en) | A kind of detection method for detecting Detection of Bolt Bonding Integrity | |
| CN108490806A (en) | Based on the system resilience Simulation Evaluation method for improving fault modes and effect analysis | |
| CN110889155A (en) | Steel bridge deck corrosion prediction model and construction method | |
| CN110399590A (en) | A kind of A.C. contactor state evaluating method based on variable weight coefficient | |
| Gay et al. | Novel resilience assessment methodology for water distribution systems | |
| CN114611372A (en) | Industrial equipment health prediction method based on Internet of things edge calculation | |
| Ding et al. | Structural damage detection based on residual force vector and imperialist competitive algorithm | |
| CN107273632B (en) | The non-accumulated impact pectus part lifetime of system prediction technique of one kind and device | |
| CN115684936A (en) | Power battery health failure probability calculation and health failure map establishment method based on machine learning | |
| CN115545514A (en) | Health degree evaluation-based differentiated operation and maintenance method and device for power distribution fusion equipment | |
| CN113486291B (en) | Petroleum drilling machine micro-grid fault prediction method based on deep learning | |
| CN108959721B (en) | Method for calculating reliability of long-span bridge member under action of windmill load | |
| Kia et al. | Closed-form fragility analysis of the steel moment resisting frames | |
| CN110135085B (en) | Fatigue crack evaluation method based on adaptive kernel density estimation auxiliary particle filter | |
| CN103164627B (en) | Based on the random structure damnification recognition method of genetic algorithm and static force measurement data | |
| CN106407734B (en) | A kind of bridge technology state deterioration appraisal procedure introducing parameter lambda | |
| CN111935188A (en) | KS detection-based method and device for maximizing attack flow in shooting range environment | |
| CN111044808A (en) | Power utilization information acquisition system operation and maintenance quality reliability assessment system and method |
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 |