System and method for predicting remaining useful life of device components
IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0957603
(2010-12-01)
|
등록번호 |
US-8712726
(2014-04-29)
|
발명자
/ 주소 |
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
1 인용 특허 :
7 |
초록
▼
A method and system for accurately predicting the remaining useful life of devices and components based on rigorous statistical analysis data to reduce service costs by implementing condition-based maintenance. One rigorous statistical model is the general degradation path model, which can be used t
A method and system for accurately predicting the remaining useful life of devices and components based on rigorous statistical analysis data to reduce service costs by implementing condition-based maintenance. One rigorous statistical model is the general degradation path model, which can be used to generate simulated data that shares similar data characteristics of historical field failure data. This generated data can be used in a reliability study based on, for example, Monte Carlo techniques for RUL prediction. The study can be used to investigate the effects of influential factors such as suspension percentage and heavy-tailed behavior. The remaining useful life prediction is based on both the fixed-time predictors and time-dependent covariates.
대표청구항
▼
1. A method for accurately predicting the remaining useful life of a device component, said method comprising: generating time-to-failure data that shares the same characteristics with use data of a device component;analyzing at least one operation condition and a device operation profile associated
1. A method for accurately predicting the remaining useful life of a device component, said method comprising: generating time-to-failure data that shares the same characteristics with use data of a device component;analyzing at least one operation condition and a device operation profile associated with said device component utilizing a statistical analysis based on a Monte Carlo technique;analyzing the effects of suspension percentage and heavy-tailed behavior on said remaining useful life prediction;analyzing a fixed-time predictor and a time-dependent covariate as indicators of remaining useful life of said device component; andgenerating prediction data indicative of a remaining useful life of said device component in order to thereafter perform a condition-based preventive maintenance on said device component based on said remaining useful life of said device component. 2. The method of claim 1 further comprising performing said condition-based preventative maintenance on said device component based on said remaining useful life of said device component. 3. The method of claim 1 further comprising: configuring said statistical analysis to comprise a General Path Model of analysis; andgenerating via said General Path Model of analysis, simulated data that shares similar data characteristics of historical field failure data. 4. The method of claim 3 further comprising: recording a degradation measurement taken at a specified time until said degradation measurement reaches a specified critical failure level, orrecording a degradation measurement taken at a specified time until said degradation measurement reaches a predetermined time. 5. The method of claim 1 wherein said statistical analysis further comprises a Random Forest Model of analysis, wherein said Random Forest Model comprises analyzing independent training decision trees on a set of data points to determine said remaining useful life of said device component. 6. The method of claim 1 further comprising: specifying a model form for a wear indicator behavior, time-dependent covariate and said covariate's parameters;generating said covariate using a regression model and a constant time predictor; andincorporating a remaining useful life predictor into said collected time-to-failure data, wherein said remaining useful life predictor comprises at least one of the following: printing rate, area coverage, inferred dark decay signal, and market segment. 7. The method of claim 1 further comprising: generating an actual degradation path for a device component; andrandomly generating a critical level from a specified distribution for said actual degradation path. 8. The method of claim 7 further comprising: adding noise into said actual degradation path, wherein said noise comprises at least one of the following: censoring, heavy-tail, and misclassification; andadding noise into time-to-failure data, wherein said noise comprises at least one of the following: censoring, heavy-tail, and misclassification. 9. The method of claim 1 further comprising: computing failure time for said device component when said actual degradation path reaches a threshold value; andpredicting a distribution of time-to-failure of said device component based on said calculated failure time. 10. A system for accurately predicting the remaining useful life of a device component, said system comprising: a processor;a data bus coupled to said processor; anda computer-usable medium embodying computer program code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for: generating time-to-failure data that shares the same characteristics with use data of a device component;analyzing at least one operation condition and a device operation profile associated with said device component utilizing a statistical analysis based on a Monte Carlo technique;analyzing the effects of suspension percentage and heavy-tailed behavior on said remaining useful life prediction;analyzing a fixed-time predictor and a time-dependent covariate as indicators of remaining useful life of said device component; andgenerating prediction data indicative of a remaining useful life of said device component in order to thereafter perform a condition-based preventive maintenance on said device component based on said remaining useful life of said device component. 11. The system of claim 10 wherein said instructions are further configured for: performing said condition-based preventative maintenance on said device component based on said remaining useful life of said device component. 12. The system of claim 10 wherein said instructions are further configured for: configuring said statistical analysis to comprise a General Path Model of analysis; andgenerating via said General Path Model of analysis, simulated data that shares similar data characteristics of historical field failure data. 13. The system of claim 12 wherein said instructions are further configured for: recording a degradation measurement taken at a specified time until said degradation measurement reaches a specified critical failure level, orrecording a degradation measurement taken at a specified time until said degradation measurement reaches a predetermined time. 14. The system of claim 10 wherein said statistical analysis further comprises a Random Forest Model of analysis, wherein said Random Forest Model comprises analyzing independent training decision trees on a set of data points to determine said remaining useful life of said device component. 15. The system of claim 10 wherein said instructions are further configured for: specifying a model form for a wear indicator behavior, time-dependent covariate and said covariate's parameters;generating said covariate using a regression model and a constant time predictor; andincorporating a remaining useful life predictor into said collected time-to-failure data, wherein said remaining useful life predictor comprises at least one of the following: printing rate, area coverage, inferred dark decay signal, and market segment. 16. The system of claim 10 wherein said instructions are further configured for: generating an actual degradation path for a device component; andrandomly generating a critical level from a specified distribution for said actual degradation path. 17. The system of claim 16 wherein said instructions are further configured for: adding noise into said actual degradation path, wherein said noise comprises at least one of the following: censoring, heavy-tail, and misclassification; andadding noise into time-to-failure data, wherein said noise comprises at least one of the following: censoring, heavy-tail, and misclassification. 18. The system of claim 10 wherein said instructions are further configured for: computing failure time for said device component when said actual degradation path reaches a threshold value; andpredicting a distribution of time-to-failure of said device component based on said calculated failure time. 19. A system for accurately predicting the remaining useful life of a device component, said system comprising: a processor;a data bus coupled to said processor; anda computer-usable medium embodying computer program code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for: generating time-to-failure data that shares the same characteristics with use data of a device component;analyzing at least one operation condition and a device operation profile associated with said device component utilizing a statistical analysis based on a Monte Carlo technique, said statistical analysis further comprising a Random Forest Model of analysis, wherein said Random Forest Model comprises analyzing independent training decision trees on a set of data points to determine said remaining useful life of said device component; andgenerating prediction data indicative of a remaining useful life of said device component in order to thereafter perform a condition-based preventive maintenance on said device component based on said remaining useful life of said device component. 20. The system of claim 19 wherein said instructions are further configured for: performing said condition-based preventative maintenance on said device component based on said remaining useful life of said device component.
이 특허에 인용된 특허 (7)
-
Gross, Kenny C.; Whisnant, Keith A.; Urmanov, Aleksey M., Method and apparatus for predicting remaining useful life for a computer system.
-
Goebel,Kai Frank; Graichen,Catherine Mary; Dometita,Michael Robert, Method, system and computer product for estimating a remaining equipment life.
-
Goebel,Kai Frank; Bonissone,Piero Patrone; Yan,Weizhong; Eklund,Neil Holger White; Xue,Feng; Qiu,Hai, System and method for equipment life estimation.
-
Bonissone, Piero Patrone; Xue, Feng; Varma, Anil; Goebel, Kai Frank; Yan, Weizhong; Eklund, Neil Holger White, System and method for equipment remaining life estimation.
-
Goebel, Kai Frank; Bonissone, Piero Patrone; Yan, Weizhong; Eklund, Neil Holger White; Xue, Feng, System and method for equipment remaining life estimation.
-
Wachter,Roman T.; Kumar,Jeetendra; Gobbak,Nataraj Kumar, Tracking component usage in a printing device.
-
Bradski, Gary; Chuang, Jason C., Training and using classification components on multiple processing units.
이 특허를 인용한 특허 (1)
-
Schramm, Gerrit; Schirrmann, Arnd; Zoller, Klaus, Maintenance information device, condition sensor for use therein and method which can be carried out therewith for arriving at a decision whether or not to perform servicing or maintenance.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.