CN103310051B - Board information terminal Failure Rate Forecasting Method in a kind of life cycle management - Google Patents
Board information terminal Failure Rate Forecasting Method in a kind of life cycle management Download PDFInfo
- Publication number
- CN103310051B CN103310051B CN201310221921.6A CN201310221921A CN103310051B CN 103310051 B CN103310051 B CN 103310051B CN 201310221921 A CN201310221921 A CN 201310221921A CN 103310051 B CN103310051 B CN 103310051B
- Authority
- CN
- China
- Prior art keywords
- information terminal
- board information
- life
- stress
- model
- 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.)
- Expired - Fee Related
Links
Landscapes
- Testing Resistance To Weather, Investigating Materials By Mechanical Methods (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
Board information terminal Failure Rate Forecasting Method in a kind of life cycle management of the present invention, belongs to forecasting technique in life span field, the prediction of the board information terminal failure rate in especially a kind of life cycle management.Comprise 1) determine board information terminal life cycle management; 2) board information terminal failure criterion is determined; 3) synthetic chemistry laboratory speedup factor model is constructed; 4) synthetic chemistry laboratory high-acceleration life test data is gathered; 5) failure rate model is constructed; 6) board information terminal Performance Degradation Model is built in conjunction with failure rate model; 7) structure is based on the life-span dynamic prediction model of gray theory; 8) failure rate and the residual life of the real time node under board information terminal life cycle management is predicted.The present invention by systematically applying the environmental stress that increases gradually and working stress, shorten sense cycle to excite fault, the weak link exposed in design; By analyzing the performance condition of each operation, provide the reliability index of each operation of product.
Description
Technical field
Board information terminal Failure Rate Forecasting Method in a kind of life cycle management of the present invention, belongs to forecasting technique in life span field, the prediction of the board information terminal failure rate in especially a kind of life cycle management.
Background technology
In normal working conditions, the normal various reliability characteristics adopting life test method to remove to estimate product.The product that this method was grown especially concerning the life-span, it not a kind of suitable method.Because the test period that it needs cost very long, even have little time to finish durability test, new product designs again, and old product will have been eliminated.Therefore this method and developing rapidly of product are incompatible.
Accelerated life test is a kind of effective way of rapid evaluation long-life high reliability life of product and reliability index.Accelerated life test data are utilized the key that the life characteristics under product normal stress level is assessed to be to the relation set up between life characteristics and stress level.
At present, in the research of acceleration model, single stress acceleration model is relatively ripe.But the environmental stress that product is subject in actual use is complicated, such as can be subject to the impact of temperature, appropriateness and electric stress equal stress simultaneously.In fact, also the resultant effect of these stress have impact on the life-span of product just.Therefore, in accelerated test, introduce combined stress, not only can shorten test period, improve test efficiency, and actual environment condition can be simulated more accurately, obtain more believable result.
Gray system theory is the new method of a kind of research " small sample ", " poor information " uncertain problem, by the information that the generation to " part " Given information, exploitation, extraction are useful, realizes the correct description to system running state, development law.
Gray prediction realizes by setting up GM model (GreyModel).Wherein most widely used is GM(1 about an equation, single order variable, 1) and model.GM(1,1) model is based on random original time series, the new time series formed after temporally cumulative.The rule presented can be approached by the solution of linear first-order differential equation, and approaches disclosed original time ordered series of numbers exponentially Changing Pattern through the solution of linear first-order differential equation.Be provided with
individual original time ordered series of numbers data are:
Accumulating generation is carried out to it, obtains 1-AGO(AccumulatedGeneratingOperation) sequence is:
Wherein,
Set up albefaction equation:
Whitening non trivial solution carries out regressive reduction, can obtain grey forecasting model:
When being predicted by given historical data (white system), because the partial information of " gray system " is known, the results contrast of short-term forecasting is close to real data; And when carrying out long-term forecasting according to historical data, information due to " darky system " is completely unknown, and the step number of prediction is too many, the inherent development law between historical data and " darky system " becomes Relative Fuzzy, thus it is large with the deviation change of real data curve to cause predicting the outcome.
Summary of the invention
The object of the invention is to provide the Failure Rate Forecasting Method of the board information terminal in a kind of life cycle management for above-mentioned weak point, using the rule out-of-service time data that obtain through the pre-service training sample as Performance Degradation Model, the data of 3 moving average methods to grey accumulated generating operation are adopted to process, by the self study of grey forecasting model, automatic Rule Summary from test figure, and predict unknown information by the rule summed up, information after prediction exports can obtain life-span under board information terminal normal stress level and failure rate information after inverse transformation.By the life cycle management of predict electronic product, and the residual life of each operation, set up the mathematical model of life cycle management; By synthetic chemistry laboratory strenuous test, set up the failure rate model in life cycle management, the failure rate of each operation of prediction board information terminal.
Board information terminal Failure Rate Forecasting Method in a kind of life cycle management takes following technical scheme to realize, and the board information terminal Failure Rate Forecasting Method in life cycle management, comprises following step:
1) board information terminal life cycle management is determined
The life cycle management of board information terminal is defined as board information terminal from its service time, or resumes work after overhauling, until the working time before ultimate limit state.
2) board information terminal failure criterion is determined
Described board information terminal fault is divided into hard fault and soft fault, described hard fault represent a board information terminal applying harassing and wrecking during and afterwards, the one or more function that it designs in advance can not be performed, if and do not repair or do not replace components and parts wherein, then can not recover it and normally work.
Described soft fault represents that a board information terminal is during applying harassing and wrecking, can not perform the one or more function that it designs in advance, but can return to normal operating state after stopping applies harassing and wrecking.
3) synthetic chemistry laboratory speedup factor model is constructed
Implement synthetic chemistry laboratory to board information terminal and do Highly Accelerated Life Test, described synthetic chemistry laboratory comprises temperature, humidity and electric stress.
The speedup factor model formula of temperature is as follows:
Wherein,
represent room temperature absolute temperature,
represent the absolute temperature under high temperature,
Ea represents the energy of activation (eV) of reaction of losing efficacy,
K represents Boltzmann constant.
The speedup factor model formula of humidity is as follows:
Wherein,
represent accelerated test relative humidity,
represent the relative humidity that normally works,
N represents the rate of acceleration constant of humidity.
The formula of the speedup factor model of electric stress is as follows:
Wherein,
for high electric stress,
for normal electrical stress, c is constant.
Highly Accelerated Life Test carries out under the synthetic chemistry laboratory of the multiple combinations such as temperature, humidity and electric stress, is obtained the comprehensive speedup factor tested by the environmental stress speedup factor being multiplied applied.
4) synthetic chemistry laboratory high-acceleration life test data is gathered
In Highly Accelerated Life Test process, the real-time communication daily record provided by enterprise operation platform, gathers the fault-time point data of board information terminal in process of the test.
5) failure rate model is constructed
The failure rate model function built is:
If the sample number participating in above testing scheme is M, the speedup factor under each temperature stage is
, i-th sample in the test period of each temperature stage is
, running time is
, under high environmental stress running time equal test period be multiplied by current environment stress under speedup factor, then the running time of i-th sample is
(4)
By the running time of M sample
summation draws total working time T, then the actual net cycle time computing formula of M sample is:
(5)
In test process, after finding that sample data exception or pilot lamp are extremely, first carry out fault diagnosis by the method for remote debugging and restart online, if remote diagnosis is by Resolving probiems, then record trouble 1 time; If have n sample fault, then record trouble n time; In the long-range insurmountable situation of confirmation, the total degree F of record trouble
i, all the other tests proceed, and after this, the test period of this fault sample does not calculate as the working time.After having tested, the number of stoppages of whole sample is:
Thus, the board information terminal failure rate function of structure is:
（6）
6) board information terminal Performance Degradation Model is built in conjunction with failure rate model
Step-Stress Accelerated Life Testing based is adopted to test, described Step-Stress Accelerated Life Testing based is tested from low stress level by sample, stress level and test period constant speed improve paramount one-level stress level, so increase progressively step by step, until terminal to the limit of stress level, then test stopping.The time that sample experiences from on-test to complete failure is tested out by the method, calculate the particular product performance parameters that each stage of stress of stepping is corresponding, utilize this parameter, adopt Matlab Software on Drawing to go out properties of sample evaluation analysis curve, and the performance of board information terminal working condition is analyzed.
Adopt Step-Stress Accelerated Life Testing based to carry out in test process, sample needs all energisings, and ensures that omnidistance sample is connected normally with enterprise platform center; Once after finding that individual samples data exception or pilot lamp are extremely, first carry out fault diagnosis by the method for remote debugging and restart online, if remote diagnosis is by Resolving probiems, then record trouble 1 time; If have n sample fault, then record trouble n time; In the long-range insurmountable situation of confirmation, the total degree of record trouble, all the other tests proceed, and after this, the test period of this fault sample does not calculate as the working time.
The performance evaluation of described board information terminal working condition, calculates the performance number of board information terminal real time node according to formula (7).
（7）
Wherein,
for the performance number of each stress node, computing formula is as follows:
（8）
represent the
itest period under individual stage of stress;
represent the
isynthetic chemistry laboratory speedup factor under individual stage of stress;
7) structure is based on the life-span dynamic prediction model of gray theory
The performance degradation result obtained by step 6), adopts the life-span dynamic prediction model based on gray theory to carry out further forecast analysis.
The specific implementation process of the described life-span dynamic prediction model based on gray theory is as follows:
7.1) the 1st step prediction, T-N+1 ~ T-1 moment N-1 point actual value before the known T moment
, by this N-1 point actual value input grey forecasting model, obtain the predicted value in the T moment
;
7.2) the 2nd step prediction, by the T moment performance number of prediction
the input end returning model, as the last point of N-1 point time series moving window, removes T-N+1 moment performance number, is reassembled into N-1 point time series moving window
input grey forecasting model, obtains in T+1 moment predicted value
;
7.3) repeat prediction until carry out the n-th step prediction, combine new N-1 point time series moving window
input grey forecasting model, obtains in T+1 moment predicted value
.
Adopt and predict based on the life-span dynamic prediction model of gray theory, the forecasting sequence comprising overall trend can be obtained, obtain the residual life of each operation of board information terminal simultaneously.
When constructing life-span dynamic prediction model based on gray theory, pre-service need be carried out to raw data, to adjust original changing trend of data, and the intrinsic propesties of restoring system to greatest extent, thus weaken the fluctuation change of data.
Described pre-service can adopt 3 moving average method process data, and newly-generated data are:
（9）
Wherein, be provided with
nindividual given data
（10）
In the development of any one gray system, As time goes on, some random perturbations or driving factors constantly will be had to enter system, the development of system is affected by it.More toward future development, away from timeorigin, the prediction significance of GM (1,1) model is more weak.Therefore, in actual applications, constantly consideration must As time goes on enter the data of system, at any time new data be inserted in model, set up new forecast model.
8) failure rate and the residual life of the real time node under board information terminal life cycle management is predicted
By the performance evaluation value of the synthetic chemistry laboratory speedup factor and each timing node of board information terminal that calculate each temperature stage
adopt the residual life doped based on the life-span dynamic prediction model of gray theory under each working condition of board information terminal, calculate the failure rate under corresponding working condition with it by formula (6), both merging result just can obtain residual life under the real-time working situation in board information terminal life cycle management and failure rate.
The present invention by systematically applying the environmental stress that increases gradually and working stress, shorten sense cycle to excite fault, the weak link exposed in design; By analyzing the performance condition of each operation, providing the reliability index of each operation of product, possessing following advantage:
1) design Highly Accelerated Life Test scheme, significantly shorten the time of electronic product accelerated life test.
2) adopt gray prediction theory to set up life-span dynamic prediction model, grasp the residual life of each operation of product.
3) in conjunction with the performance condition of each operation of board information terminal, failure rate model is set up, the board information terminal failure rate under prediction life cycle management.
4) the present invention propose a kind of life cycle management in board information terminal Failure Rate Forecasting Method, be applicable to small sample test figure, be convenient to practical engineering application.
5) the present invention propose a kind of life cycle management in board information terminal Failure Rate Forecasting Method, have stronger engineering adaptability and versatility to different board information terminals or stress kind.
Embodiment
Board information terminal Failure Rate Forecasting Method in life cycle management, comprises following step:
1) board information terminal life cycle management is determined
The life cycle management of board information terminal is defined as board information terminal from its service time, or resumes work after overhauling, until the working time before ultimate limit state.
2) board information terminal failure criterion is determined
Described board information terminal fault is divided into hard fault and soft fault, described hard fault represent a board information terminal applying harassing and wrecking during and afterwards, the one or more function that it designs in advance can not be performed, if and do not repair or do not replace components and parts wherein, then can not recover it and normally work.
Described soft fault represents that a board information terminal is during applying harassing and wrecking, can not perform the one or more function that it designs in advance, but can return to normal operating state after stopping applies harassing and wrecking.
3) synthetic chemistry laboratory speedup factor model is constructed
Implement synthetic chemistry laboratory to board information terminal and do Highly Accelerated Life Test, described synthetic chemistry laboratory comprises temperature, humidity and electric stress.
The speedup factor model formula of temperature is as follows:
Wherein,
represent room temperature absolute temperature,
represent the absolute temperature under high temperature,
Ea represents the energy of activation (eV) of reaction of losing efficacy,
K represents Boltzmann constant.
The speedup factor model formula of humidity is as follows:
Wherein,
represent accelerated test relative humidity,
represent the relative humidity that normally works,
N represents the rate of acceleration constant of humidity.
The formula of the speedup factor model of electric stress is as follows:
Wherein,
for high electric stress,
for normal electrical stress, c is constant.
Highly Accelerated Life Test carries out under the synthetic chemistry laboratory of the multiple combinations such as temperature, humidity and electric stress, is obtained the comprehensive speedup factor tested by the environmental stress speedup factor being multiplied applied.
4) synthetic chemistry laboratory high-acceleration life test data is gathered
In Highly Accelerated Life Test process, the real-time communication daily record provided by enterprise operation platform, gathers the fault-time point data of board information terminal in process of the test.
5) failure rate model is constructed
The failure rate model function built is:
If the sample number participating in above testing scheme is M, the speedup factor under each temperature stage is
, i-th sample in the test period of each temperature stage is
, running time is
, under high environmental stress running time equal test period be multiplied by current environment stress under speedup factor, then the running time of i-th sample is
(4)
By the running time of M sample
summation draws total working time T, then the actual net cycle time computing formula of M sample is:
(5)
In test process, after finding that sample data exception or pilot lamp are extremely, first carry out fault diagnosis by the method for remote debugging and restart online, if remote diagnosis is by Resolving probiems, then record trouble 1 time; If have n sample fault, then record trouble n time; In the long-range insurmountable situation of confirmation, the total degree F of record trouble
i, all the other tests proceed, and after this, the test period of this fault sample does not calculate as the working time.After having tested, the number of stoppages of whole sample is:
Thus, the board information terminal failure rate function of structure is:
（6）
6) board information terminal Performance Degradation Model is built in conjunction with failure rate model
Step-Stress Accelerated Life Testing based is adopted to test, described Step-Stress Accelerated Life Testing based is tested from low stress level by sample, stress level and test period constant speed improve paramount one-level stress level, so increase progressively step by step, until terminal to the limit of stress level, then test stopping.The time that sample experiences from on-test to complete failure is tested out by the method, calculate the particular product performance parameters that each stage of stress of stepping is corresponding, utilize this parameter, adopt Matlab Software on Drawing to go out properties of sample evaluation analysis curve, and the performance of board information terminal working condition is analyzed.
According to the testing program parameter of design, concrete test parameters is arranged in table 1.
Table 1 test parameters
| Test period | Temperature/humidity |
| 24 hours | 25℃+60% |
| 24 hours | 35℃+65% |
| 24 hours | 45℃+70% |
| 24 hours | 55℃+75% |
| 24 hours | 65℃+80% |
| 24 hours | 75℃+85% |
| 24 hours | 85℃+90% |
| 24 hours | 95℃+95% |
Adopt Step-Stress Accelerated Life Testing based to carry out in test process, sample needs all energisings, and ensures that omnidistance sample is connected normally with enterprise platform center; Once after finding that individual samples data exception or pilot lamp are extremely, first carry out fault diagnosis by the method for remote debugging and restart online, if remote diagnosis is by Resolving probiems, then record trouble 1 time; If have n sample fault, then record trouble n time; In the long-range insurmountable situation of confirmation, the total degree of record trouble, all the other tests proceed, and after this, the test period of this fault sample does not calculate as the working time.
The performance evaluation of described board information terminal working condition, calculates the performance number of board information terminal real time node according to formula (7).
（7）
Wherein,
for the performance number of each stress node, computing formula is as follows:
（8）
represent the
itest period under individual stage of stress;
represent the
isynthetic chemistry laboratory speedup factor under individual stage of stress;
7) structure is based on the life-span dynamic prediction model of gray theory
The performance degradation result obtained by step 6), adopts the life-span dynamic prediction model based on gray theory to carry out further forecast analysis.
From pre-measuring angle, dynamic prediction model is optimal model.Along with the development of system, the informative of historical data will progressively reduce, and while constantly supplementing fresh information, remove old data timely, and forecast model also just more can reflect the feature that system is current.In addition, remove old data timely, can also calculated amount be reduced, be conducive to actual computing.In addition, often predict a step, grey parameter does once to be revised, and forecast model is upgraded.Along with the continuous correction of grey parameter, model also progressively perfect, thus improve the precision of prediction.
The specific implementation process of the described life-span dynamic prediction model based on gray theory is as follows:
7.1) the 1st step prediction, T-N+1 ~ T-1 moment N-1 point actual value before the known T moment
, by this N-1 point actual value input grey forecasting model, obtain the predicted value in the T moment
;
7.2) the 2nd step prediction, by the T moment performance number of prediction
the input end returning model, as the last point of N-1 point time series moving window, removes T-N+1 moment performance number, is reassembled into N-1 point time series moving window
input grey forecasting model, obtains in T+1 moment predicted value
;
7.3) repeat prediction until carry out the n-th step prediction, combine new N-1 point time series moving window
input grey forecasting model, obtains in T+1 moment predicted value
.
Adopt and predict based on the life-span dynamic prediction model of gray theory, the forecasting sequence comprising overall trend can be obtained, obtain the residual life of each operation of board information terminal simultaneously.
Can supplement by this model in time and utilize new information, improve the Whitened degree of gray zone.And often predict that a step gray model parameter is done and once revise, model is improved, and the grey forecasting model of structure further increases the accuracy of prediction.
When constructing life-span dynamic prediction model based on gray theory, pre-service need be carried out to raw data, to adjust original changing trend of data, and the intrinsic propesties of restoring system to greatest extent, thus weaken the fluctuation change of data.
Described pre-service can adopt 3 moving average method process data, and newly-generated data are:
（9）
Wherein, be provided with
nindividual given data
（10）
In the development of any one gray system, As time goes on, some random perturbations or driving factors constantly will be had to enter system, the development of system is affected by it.More toward future development, away from timeorigin, the prediction significance of GM (1,1) model is more weak.Therefore, in actual applications, constantly consideration must As time goes on enter the data of system, at any time new data be inserted in model, set up new forecast model.
8) failure rate and the residual life of the real time node under board information terminal life cycle management is predicted
By the performance evaluation value of the synthetic chemistry laboratory speedup factor and each timing node of board information terminal that calculate each temperature stage
adopt the residual life doped based on the life-span dynamic prediction model of gray theory under each working condition of board information terminal, calculate the failure rate under corresponding working condition with it by formula (6), both merging result just can obtain residual life under the real-time working situation in board information terminal life cycle management and failure rate.
Claims (4)
1. the board information terminal Failure Rate Forecasting Method in life cycle management, is characterized in that, comprises following step:
1) board information terminal life cycle management is determined
The life cycle management of board information terminal is defined as board information terminal from its service time, or resumes work after overhauling, until the working time before ultimate limit state;
2) board information terminal failure criterion is determined
Described board information terminal fault is divided into hard fault and soft fault, described hard fault represent a board information terminal applying harassing and wrecking during and afterwards, the one or more function that it designs in advance can not be performed, if and do not repair or do not replace components and parts wherein, then can not recover it and normally work;
Described soft fault represents that a board information terminal is during applying harassing and wrecking, can not perform the one or more function that it designs in advance, but can return to normal operating state after stopping applies harassing and wrecking;
3) synthetic chemistry laboratory speedup factor model is constructed
Implement synthetic chemistry laboratory to board information terminal and do Highly Accelerated Life Test, described synthetic chemistry laboratory comprises temperature, humidity and electric stress;
Carry out under the synthetic chemistry laboratory that Highly Accelerated Life Test combines at temperature, humidity and electric stress, obtained the comprehensive speedup factor tested by the environmental stress speedup factor being multiplied applied;
The speedup factor model formula of described temperature is as follows:
Wherein,
represent room temperature absolute temperature,
represent the absolute temperature under high temperature,
Ea represents the energy of activation (eV) of reaction of losing efficacy,
K represents Boltzmann constant;
The speedup factor model formula of humidity is as follows:
Wherein,
represent accelerated test relative humidity,
represent the relative humidity that normally works,
N represents the rate of acceleration constant of humidity;
The formula of the speedup factor model of electric stress is as follows:
Wherein,
for high electric stress,
for normal electrical stress, c is constant;
4) synthetic chemistry laboratory high-acceleration life test data is gathered
In Highly Accelerated Life Test process, the real-time communication daily record provided by enterprise operation platform, gathers the fault-time point data of board information terminal in process of the test;
5) failure rate model is constructed
The failure rate model function built is
If the sample number participating in above testing scheme is M, the speedup factor under each temperature stage is
, i-th sample in the test period of each temperature stage is
, running time is
, under high environmental stress running time equal test period be multiplied by current environment stress under speedup factor, then the running time of i-th sample is
(4)
By the running time of M sample
summation draws total working time T, then the actual net cycle time computing formula of M sample is:
(5)
In test process, after finding that sample data exception or pilot lamp are extremely, first carry out fault diagnosis by the method for remote debugging and restart online, if remote diagnosis is by Resolving probiems, then record trouble 1 time; If have n sample fault, then record trouble n time; In the long-range insurmountable situation of confirmation, the total degree F of record trouble
i, all the other tests proceed, and after this, the test period of this fault sample does not calculate as the working time; After having tested, the number of stoppages of whole sample is:
Thus, the board information terminal failure rate function of structure is
（6）
6) board information terminal Performance Degradation Model is built in conjunction with failure rate model
Step-Stress Accelerated Life Testing based is adopted to test, the time that sample experiences from on-test to complete failure is tested out by the method, calculate the particular product performance parameters that each stage of stress of stepping is corresponding, utilize this parameter, adopt Matlab Software on Drawing to go out properties of sample evaluation analysis curve, and the performance of board information terminal working condition is analyzed;
Adopt Step-Stress Accelerated Life Testing based to carry out in test process, sample needs all energisings, and ensures that omnidistance sample is connected normally with enterprise platform center; Once after finding that individual samples data exception or pilot lamp are extremely, first carry out fault diagnosis by the method for remote debugging and restart online, if remote diagnosis is by Resolving probiems, then record trouble 1 time; If have n sample fault, then record trouble n time; In the long-range insurmountable situation of confirmation, the total degree of record trouble, all the other tests proceed, and after this, the test period of this fault sample does not calculate as the working time;
The performance evaluation of described board information terminal working condition, calculates the performance number of board information terminal real time node according to formula (7);
（7）
Wherein,
for the performance number of each stress node, computing formula is as follows:
（8）
represent the
itest period under individual stage of stress;
represent the
isynthetic chemistry laboratory speedup factor under individual stage of stress;
7) structure is based on the life-span dynamic prediction model of gray theory
The performance degradation result obtained by step 6), adopts the life-span dynamic prediction model based on gray theory to carry out further forecast analysis;
Adopt and predict based on the life-span dynamic prediction model of gray theory, the forecasting sequence comprising overall trend can be obtained, obtain the residual life of each operation of board information terminal simultaneously;
8) failure rate and the residual life of the real time node under board information terminal life cycle management is predicted
By the performance evaluation value of the synthetic chemistry laboratory speedup factor and each timing node of board information terminal that calculate each temperature stage
adopt the residual life doped based on the life-span dynamic prediction model of gray theory under each working condition of board information terminal, calculate the failure rate under corresponding working condition with it by formula (6), namely both merging result obtains residual life under the real-time working situation in board information terminal life cycle management and failure rate.
2. the board information terminal Failure Rate Forecasting Method in a kind of life cycle management according to claim 1, it is characterized in that, in described step 6), Step-Stress Accelerated Life Testing based is tested from low stress level by sample, stress level and test period constant speed improve paramount one-level stress level, so increase progressively step by step, until terminal has arrived the limit of stress level, then test stopping.
3. the board information terminal Failure Rate Forecasting Method in a kind of life cycle management according to claim 1, is characterized in that, the specific implementation process of the life-span dynamic prediction model based on gray theory described in described step 7) is as follows:
7.1) the 1st step prediction, T-N+1 ~ T-1 moment N-1 point actual value before the known T moment
, by this N-1 point actual value input grey forecasting model, obtain the predicted value in the T moment
;
7.2) the 2nd step prediction, by the T moment performance number of prediction
the input end returning model, as the last point of N-1 point time series moving window, removes T-N+1 moment performance number, is reassembled into N-1 point time series moving window
input grey forecasting model, obtains in T+1 moment predicted value
;
7.3) repeat prediction until carry out the n-th step prediction, combine new N-1 point time series moving window
input grey forecasting model, obtains in T+1 moment predicted value
.
4. the board information terminal Failure Rate Forecasting Method in a kind of life cycle management according to claim 1, it is characterized in that, in step 7) when constructing the life-span dynamic prediction model based on gray theory, pre-service need be carried out to raw data, to adjust original changing trend of data, and the intrinsic propesties of restoring system to greatest extent, thus weaken the fluctuation change of data;
Described pre-service adopts 3 moving average method process data, and newly-generated data are:
（9）
Wherein, be provided with
nindividual given data
（10）
。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310221921.6A CN103310051B (en) | 2013-06-05 | 2013-06-05 | Board information terminal Failure Rate Forecasting Method in a kind of life cycle management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310221921.6A CN103310051B (en) | 2013-06-05 | 2013-06-05 | Board information terminal Failure Rate Forecasting Method in a kind of life cycle management |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103310051A CN103310051A (en) | 2013-09-18 |
| CN103310051B true CN103310051B (en) | 2016-03-02 |
Family
ID=49135263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310221921.6A Expired - Fee Related CN103310051B (en) | 2013-06-05 | 2013-06-05 | Board information terminal Failure Rate Forecasting Method in a kind of life cycle management |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103310051B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598029B (en) * | 2016-12-21 | 2019-02-26 | 北京交通大学 | Train control on board equipment Reliability Prediction Method based on performance degradation |
| CN107292415B (en) * | 2017-05-02 | 2021-07-30 | 国网浙江省电力有限公司 | Prediction method for rotation time of intelligent meter |
| CN107333374A (en) * | 2017-06-19 | 2017-11-07 | 江苏昂内斯电力科技股份有限公司 | Indoor street lamp fault alarm method |
| CN108446523B (en) * | 2018-05-11 | 2022-04-08 | 北京航天自动控制研究所 | Method for evaluating and predicting storage life of electronic complete machine |
| CN108958215A (en) * | 2018-06-01 | 2018-12-07 | 天泽信息产业股份有限公司 | A kind of engineering truck failure prediction system and its prediction technique based on data mining |
| CN109165108B (en) * | 2018-07-27 | 2022-04-05 | 同济大学 | Failure data reduction method and test method for software reliability accelerated test |
| CN110889083B (en) * | 2018-09-10 | 2020-12-22 | 湖南银杏可靠性技术研究所有限公司 | Degraded data consistency checking method based on window spectrum estimation |
| CN112444772B (en) * | 2020-11-11 | 2024-06-14 | 云南电网有限责任公司电力科学研究院 | Reliability prediction method and device for intelligent electric energy meter |
| CN113406421A (en) * | 2021-06-18 | 2021-09-17 | 上海华兴数字科技有限公司 | Durable aging loading test system and method |
| CN114267178B (en) * | 2021-12-30 | 2023-09-26 | 佳都科技集团股份有限公司 | Intelligent operation maintenance method and device for station |
| CN114896777B (en) * | 2022-05-05 | 2024-02-13 | 合肥工业大学 | Method for predicting service life of motor through temperature and load of motor based on gray theory |
| CN115830757B (en) * | 2022-12-02 | 2023-11-17 | 江苏锦花电子股份有限公司 | Display equipment performance monitoring system and method based on big data |
| CN117252040B (en) * | 2023-11-16 | 2024-02-06 | 杭州中安电子股份有限公司 | Multi-stress acceleration test analysis method, electronic device, and readable storage medium |
| CN117929906B (en) * | 2024-03-25 | 2024-06-04 | 天津蓝孚高能物理技术有限公司 | Test system of grid-control electron gun for irradiation accelerator |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629300A (en) * | 2012-03-15 | 2012-08-08 | 北京航空航天大学 | Step stress accelerated degradation data assessment method based on gray prediction models |
| CN102708306A (en) * | 2012-06-19 | 2012-10-03 | 华北电网有限公司计量中心 | Prediction method for q-precentile life of intelligent meter |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7006947B2 (en) * | 2001-01-08 | 2006-02-28 | Vextec Corporation | Method and apparatus for predicting failure in a system |
-
2013
- 2013-06-05 CN CN201310221921.6A patent/CN103310051B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629300A (en) * | 2012-03-15 | 2012-08-08 | 北京航空航天大学 | Step stress accelerated degradation data assessment method based on gray prediction models |
| CN102708306A (en) * | 2012-06-19 | 2012-10-03 | 华北电网有限公司计量中心 | Prediction method for q-precentile life of intelligent meter |
Non-Patent Citations (1)
| Title |
|---|
| "基于模糊聚类分析的特征识别方法及其应用";赵德滨等;《计算机集成制造***》;20091231;第15卷(第12期);第2417-2423、2486页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103310051A (en) | 2013-09-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103310051B (en) | Board information terminal Failure Rate Forecasting Method in a kind of life cycle management | |
| CN111443294B (en) | Method and device for indirectly predicting remaining life of lithium ion battery | |
| CN109033709B (en) | Component fatigue life evaluation method based on nonlinear fatigue damage accumulation theory | |
| CN107423414B (en) | Information transfer model-based process industry complex electromechanical system fault tracing method | |
| CN110555230B (en) | Rotary machine residual life prediction method based on integrated GMDH framework | |
| CN112149373B (en) | Complex analog circuit fault identification and estimation method and system | |
| CN111967688B (en) | Power load prediction method based on Kalman filter and convolutional neural network | |
| CN107730127B (en) | Relay storage degradation data prediction method based on output characteristic initial distribution | |
| CN111325403B (en) | Method for predicting residual life of electromechanical equipment of highway tunnel | |
| CN107944612B (en) | Bus net load prediction method based on ARIMA and phase space reconstruction SVR | |
| CN111639783A (en) | Line loss prediction method and system based on LSTM neural network | |
| CN105868557A (en) | Online prediction method for remaining life of electromechanical equipment under situation of two-stage degradation | |
| CN105468850A (en) | Multi-residual error regression prediction algorithm based electronic product degradation trend prediction method | |
| CN113449919B (en) | Power consumption prediction method and system based on feature and trend perception | |
| CN114035468B (en) | Method and system for predictively monitoring overhaul flow of fan based on XGBoost algorithm | |
| CN109598052B (en) | Intelligent ammeter life cycle prediction method and device based on correlation coefficient analysis | |
| CN109344967B (en) | Intelligent electric meter life cycle prediction method based on artificial neural network | |
| CN114066262A (en) | Method, system and device for estimating cause-tracing reasoning of abnormal indexes after power grid dispatching and storage medium | |
| CN108763250B (en) | Photovoltaic power station monitoring data restoration method | |
| CN103885867B (en) | Online evaluation method of performance of analog circuit | |
| Zhukovskiy et al. | The probability estimate of the defects of the asynchronous motors based on the complex method of diagnostics | |
| CN109921462B (en) | New energy consumption capability assessment method and system based on LSTM | |
| CN114862032A (en) | XGboost-LSTM-based power grid load prediction method and device | |
| Le | Contribution to deterioration modeling and residual life estimation based on condition monitoring data | |
| CN112232570A (en) | Forward active total electric quantity prediction method and device and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20130918 Assignee: JIANGSU SEA LEVEL DATA TECHNOLOGY Co.,Ltd. Assignor: TIANZE INFORMATION INDUSTRY Corp. Contract record no.: X2020320000015 Denomination of invention: Method for predicting failure rate of vehicle-mounted information terminal in entire life cycle Granted publication date: 20160302 License type: Exclusive License Record date: 20200518 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160302 Termination date: 20210605 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |