119:
However, there is incentive to know better when a system would fail to better leverage the remaining useful life while at the same time avoiding unscheduled maintenance (unscheduled maintenance is typically more costly than scheduled maintenance and results in system downtime). Garga et al. describe conceptually a pre-estimate aggregation hybrid approach where domain knowledge is used to change the structure of a neural network, thus resulting in a more parsimonious representation of the network. Another way to accomplish the pre-estimate aggregation is by a combined off-line process and on-line process: In the off-line mode, one can use a physics-based simulation model to understand the relationships of sensor response to fault state; In the on-line mode, one can use data to identify current damage state, then track the data to characterize damage propagation, and finally apply an individualized data-driven propagation model for remaining life prediction. For example, Khorasgani et al. modeled the physics of failure in electrolytic capacitors. Then, they used a particle filter approach to derive the dynamic form of the degradation model and estimate the current state of capacitor health. This model is then used to get more accurate estimation of the
Remaining Useful Life (RUL) of the capacitors as they are subjected to the thermal stress conditions.
137:
accuracy and precision based where performance is evaluated against end of life, typically known a priori in an offline setting. More recently, efforts towards maturing prognostics technology has put a significant focus on standardizing prognostic methods, including those of performance assessment. A key aspect, missing from the conventional metrics, is the capability to track performance with time. This is important because prognostics is a dynamic process where predictions get updated with an appropriate frequency as more observation data become available from an operational system. Similarly, the performance of prediction changes with time that must be tracked and quantified. Another aspect that makes this process different in a PHM context is the time value of a RUL prediction. As a system approaches failure, the time window to take a corrective action gets shorter and consequently the accuracy of predictions becomes more critical for decision making. Finally, randomness and noise in the process, measurements, and prediction models are unavoidable and hence prognostics inevitably involves uncertainty in its estimates. A robust prognostics performance evaluation must incorporate the effects of this uncertainty.
88:
run-to-failure data, in particular for new systems, since running systems to failure can be a lengthy and rather costly process. When future usage is not the same as in the past (as with most non-stationary systems), collecting data that includes all possible future usages (both load and environmental conditions) becomes often nearly impossible. Even where data exist, the efficacy of data-driven approaches is not only dependent on the quantity but also on the quality of system operational data. These data sources may include temperature, pressure, oil debris, currents, voltages, power, vibration and acoustic signal, spectrometric data as well as calibration and calorimetric data. The data often needs to be pre-processed before it can be used. Typically two procedures are performed i) Denoising and ii) Feature extraction. Denoising refers to reducing or eliminating the influence of noise on data. Features extraction is important because in today's data hungry world, huge amount of data is collected using sensor measurement that may not be used readily. Therefore, domain knowledge and statistical signal processing is applied to extract important features from (more often than not) noisy, high-dimensional data.
80:
bilinear model, the projection pursuit, the multivariate adaptive regression splines, and the
Volterra series expansion. Since the last decade, more interests in data-driven system state forecasting have been focused on the use of flexible models such as various types of neural networks (NNs) and neural fuzzy (NF) systems. Data-driven approaches are appropriate when the understanding of first principles of system operation is not comprehensive or when the system is sufficiently complex such that developing an accurate model is prohibitively expensive. Therefore, the principal advantages to data driven approaches is that they can often be deployed quicker and cheaper compared to other approaches, and that they can provide system-wide coverage (cf. physics-based models, which can be quite narrow in scope). The main disadvantage is that data driven approaches may have wider confidence intervals than other approaches and that they require a substantial amount of data for training. Data-driven approaches can be further subcategorized into fleet-based statistics and sensor-based conditioning. In addition, data-driven techniques also subsume cycle-counting techniques that may include
97:
micro level (also called material level), physical models are embodied by series of dynamic equations that define relationships, at a given time or load cycle, between damage (or degradation) of a system/component and environmental and operational conditions under which the system/component are operated. The micro-level models are often referred as damage propagation model. For example, Yu and Harris's fatigue life model for ball bearings, which relates the fatigue life of a bearing to the induced stress, Paris and
Erdogan's crack growth model, and stochastic defect-propagation model are other examples of micro-level models. Since measurements of critical damage properties (such as stress or strain of a mechanical component) are rarely available, sensed system parameters have to be used to infer the stress/strain values. Micro-level models need to account in the uncertainty management the assumptions and simplifications, which may pose significant limitations of that approach.
231:
gearboxes, batteries, and other machines that can be reused for developing custom predictive maintenance and condition monitoring algorithms. Other commercial software offerings focus on a few tools for anomaly detection and fault diagnosis, and are typically offered as a package solution instead of a toolkit offering. Example includes Smart
Signals anomaly detection analytical method, based on auto-associative type models (similarity based modeling) that look for changes in the nominal correlation relationship in the signals, calculates residuals between expected and actual performance, and then performs hypothesis testing on the residual signals (sequential probability ratio test). Similar types of analysis methods are also offered by Expert Microsystems, which uses a similar auto-associative kernel method for calculating residuals, and has other modules for diagnosis and prediction.
35:), which is an important concept in decision making for contingency mitigation. Prognostics predicts the future performance of a component by assessing the extent of deviation or degradation of a system from its expected normal operating conditions. The science of prognostics is based on the analysis of failure modes, detection of early signs of wear and aging, and fault conditions. An effective prognostics solution is implemented when there is sound knowledge of the failure mechanisms that are likely to cause the degradations leading to eventual failures in the system. It is therefore necessary to have initial information on the possible failures (including the site, mode, cause and mechanism) in a product. Such knowledge is important to identify the system parameters that are to be monitored. Potential uses for prognostics is in
110:
of data-driven approaches and data-driven approaches glean available information from models. An example for the former would be where model parameters are tuned using field data. An example for the latter is when the set-point, bias, or normalization factor for a data-driven approach is given by models. Hybrid approaches can be categorized broadly into two categories, 1) Pre-estimate fusion and 2.) Post-estimate fusion. Physics-based and empirical model may be combined in a simple additive fashion as well, such as an additive combination of a basic physics-based model with an artificial neural network. This could deliver superior diagnostic or prognostic performance than just one or the other. See, for example, such an additive hybrid model employed in jet engine performance analysis.
226:
toolkit, which is a collection of data-driven PHM algorithms that were developed by the Center for
Intelligent Maintenance Systems. This collection of over 20 tools allows one to configure and customize the algorithms for signature extraction, anomaly detection, health assessment, failure diagnosis, and failure prediction for a given application as needed. Customized predictive monitoring commercial solutions using the Watchdog Agent toolkit are now being offered by a recent start-up company called Predictronics Corporation in which the founders were instrumental in the development and application of this PHM technology at the Center for Intelligent Maintenance Systems. Another example is
128:
sources can help to improve performance of an estimator. This principle has been successfully applied within the context of classifier fusion where the output of multiple classifiers is used to arrive at a better result than any classifier alone. Within the context of prognostics, fusion can be accomplished by employing quality assessments that are assigned to the individual estimators based on a variety of inputs, for example heuristics, a priori known performance, prediction horizon, or robustness of the prediction.
222:
hardware can be customized or ruggedized as needed. Common sensor types for PHM applications include accelerometers, temperature, pressure, measurements of rotational speed using encoders or tachometers, electrical measurements of voltage and current, acoustic emission, load cells for force measurements, and displacement or position measurements. There are numerous sensor vendors for those measurement types, with some having a specific product line that is more suited for condition monitoring and PHM applications.
101:
The trade-off is increased coverage with possibly reducing accuracy of a particular degradation mode. Where this trade-off is permissible, faster prototyping may be the result. However, where systems are complex (e.g., a gas turbine engine), even a macro-level model may be a rather time-consuming and labor-intensive process. As a result, macro-level models may not be available in detail for all subsystems. The resulting simplifications need to be accounted for by the uncertainty management.
96:
Physics-based prognostics (sometimes called model-based prognostics) attempts to incorporate physical understanding (physical models) of the system into the estimation of remaining useful life (RUL). Modeling physics can be accomplished at different levels, for example, micro and macro levels. At the
225:
The data analysis algorithms and pattern recognition technology are now being offered in some commercial software platforms or as part of a packaged software solution. National
Instruments currently has a trial version (with a commercial release in the upcoming year) of the Watchdog Agent prognostic
100:
Macro-level models are the mathematical model at system level, which defines the relationship among system input variables, system state variables, and system measures variables/outputs where the model is often a somewhat simplified representation of the system, for example a lumped parameter model.
187:
this concerns the imprecisions in the mathematical models which is generated to represent the behavior of the system. These imprecisions (or uncertainties) can be the result of a set of assumptions used during the modeling process and which lead to models that don't fit exactly the real behavior of
109:
Hybrid approaches attempt to leverage the strength from both data-driven approaches as well as model-based approaches. In reality, it is rare that the fielded approaches are completely either purely data-driven or purely model-based. More often than not, model-based approaches include some aspects
208:
the failure threshold is important in any fault detection and prediction methods. It determines the time at which the system fails and consequently the remaining useful life. In practice, the value of the failure threshold is not constant and can change in time. It can also change according to the
201:
uncertainty is inherent to any prediction process. Any nominal and/or degradation model predictions are inaccurate which is impacted by several uncertainties such as uncertainty in the model parameters, the environmental conditions and the future mission profiles. The prediction uncertainty can be
26:
is an engineering discipline focused on predicting the time at which a system or a component will no longer perform its intended function. This lack of performance is most often a failure beyond which the system can no longer be used to meet desired performance. The predicted time then becomes the
239:
While most prognostics approaches focus on accurate computation of the degradation rate and the remaining useful life (RUL) of individual components, it is the rate at which the performance of subsystems and systems degrade that is of greater interest to the operators and maintenance personnel of
118:
The motivation for pre-estimate aggregation may be that no ground truth data are available. This may occur in situations where diagnostics does a good job in detecting faults that are resolved (through maintenance) before system failure occurs. Therefore, there are hardly any run-to-failure data.
230:
and its
Predictive Maintenance Toolbox which provides functions and an interactive app for exploring, extracting, and ranking features using data-based and model-based techniques, including statistical, spectral, and time-series analysis. This toolbox also includes reference examples for motors,
136:
Prognostic performance evaluation is of key importance for a successful PHM system deployment. The early lack of standardized methods for performance evaluation and benchmark data-sets resulted in reliance on conventional performance metrics borrowed from statistics. Those metrics were primarily
127:
Motivation for post-estimate fusion is often consideration of uncertainty management. That is, the post-estimate fusion helps to narrow the uncertainty intervals of data-driven or model-based approaches. At the same time, the accuracy improves. The underlying notion is that multiple information
221:
For most PHM industrial applications, commercial off the shelf data acquisition hardware and sensors are normally the most practical and common. Example commercial vendors for data acquisition hardware include
National Instruments and Advantech Webaccess; however, for certain applications, the
87:
The two basic data-driven strategies involve (1) modeling cumulative damage (or, equivalently, health) and then extrapolating out to a damage (or health) threshold, or (2) learning directly from data the remaining useful life. As mentioned, a principal bottleneck is the difficulty in obtaining
180:
this concerns the uncertainty in the values of the physical parameters of the system (resistance, inductance, stiffness, capacitance, etc.). This uncertainty is induced by the environmental and operational conditions where the system evolves. This can be tackled by using adequate methods such
79:
Data-driven prognostics usually use pattern recognition and machine learning techniques to detect changes in system states. The classical data-driven methods for nonlinear system prediction include the use of stochastic models such as the autoregressive (AR) model, the threshold AR model, the
194:
the degradation model can be obtained from accelerated life tests which are conducted on different data samples of a component. In practice, the data obtained by accelerated life tests performed under the same operating conditions may have different degradation trend. This difference in the
715:
1069:
Rodrigues, L. R.; Gomes, J. P. P.; Ferri, F. A. S.; Medeiros, I. P.; Galvão, R. K. H.; Júnior, C. L. Nascimento (December 2015). "Use of PHM Information and System
Architecture for Optimized Aircraft Maintenance Planning".
144:
Prognostic horizon (PH) quantifies how much in advance an algorithm can predict with a desired accuracy before a failure occurs. A longer prognostic horizon is preferred as more time is then available for a corrective
209:
nature of the system, operating conditions and in the environment which it evolves. All these parameters induce uncertainty which should be considered in the definition of the failure threshold.
791:
Sun, Jianzhong; Zuo, Hongfu; Wang, Wenbin; Pecht, Michael G. (2014). "Prognostics uncertainty reduction by fusing on-line monitoring data based on a state-space-based degradation model".
1128:
is an open access journal that provides an international forum for the electronic publication of original research and industrial experience articles in all areas of systems prognostics.
1151:
is a non-profit organization established to address the key needs of the
Chinese community relating to quality, reliability, maintainability, safety and sustainability.
71:). Technical approaches to building models in prognostics can be categorized broadly into data-driven approaches, model-based approaches, and hybrid approaches.
151:
accuracy further tightens the desired accuracy levels using a shrinking cone of desired accuracy as end of life approaches. In order to comply with desired
752:
Sankararaman, Shankar (2015). "Significance, interpretation, and quantification of uncertainty in prognostics and remaining useful life prediction".
900:
Rocchetta, Roberto; Broggi, Matteo; Huchet, Quentin; Patelli, Edoardo (2018). "On-line Bayesian model updating for structural health monitoring".
195:
degradation trends can then be considered as an uncertainty in the degradation models derived from the data related to the accelerated life tests.
1200:
427:
Mosallam, A.; Medjaher, K.; Zerhouni, N. (2015). "Component based data-driven prognostics for complex systems: Methodology and applications".
1177:
has a group dedicated to providing a research and knowledge base to support the advancement of health management with a focus on electronics.
1054:
Clarkson, S.A.; Bickford, R.L. (2013). "Path Classification and Remaining Life Estimation for Systems Having Complex Modes of Failure".
428:
386:
834:
Duong, Pham L.T.; Raghavan, Nagarajan (2017). "Uncertainty quantification in prognostics: A data driven polynomial chaos approach".
1145:
is dedicated to developing PHM technologies that can benefit the industries by increasing their competitiveness and profitability.
884:
851:
449:
334:
307:
1206:
1174:
1165:
1139:
is an international non-profit professional organization dedicated to the advancement of PHM as an engineering discipline.
623:
Pecht, Michael; Jaai (2010). "A prognostics and health management roadmap for information and electronics-rich systems".
549:"Closure to "Discussions of 'A Critical Analysis of Crack Propagation Laws'" (1963, ASME J. Basic Eng., 85, pp. 533–534)"
350:
Liu, Jie; Wang, Golnaraghi (2009). "A multi-step predictor with a variable input pattern for system state forecasting".
202:
tackled by using Bayesian and online estimation and prediction tools (e.g. Particle Filters and Kalman filter etc.).
1199:
provides an umbrella for prognostic technology development applied to aerospace applications. It also maintains a
39:. The discipline that links studies of failure mechanisms to system lifecycle management is often referred to as
867:
Datong Liu; Yue Luo; Yu Peng (2012). "Uncertainty processing in prognostics and health management: An overview".
860:
173:
There are many uncertainty parameters that can influence the prediction accuracy. These can be categorized as:
165:
A visual representation of these metrics can be used to depict prognostic performance over a long time horizon.
1027:
Wegerich, S. (2005). "Similarity-based Modeling of Vibration Features for Fault Detection and Identification".
728:"Prognostics and Health Management for Maintenance Practitioners - Review, Implementation and Tools Evaluation"
36:
387:"Data-driven prognostic method based on Bayesian approaches for direct remaining useful life prediction"
1237:
1196:
697:
Volponi, Allan (May 2014). "Gas Turbine Engine Health Management: Past, Present, and Future Trends".
324:
297:
161:
Convergence quantifies how fast the performance converges for an algorithm as end of life approaches.
1189:
has an annual meeting focusing on latest developments in practical applications in the field of PHM.
869:
Proceedings of the IEEE 2012 Prognostics and System Health Management Conference (PHM-2012 Beijing)
275:
580:
Li, Y.; Kurfess, T.R.; Liang, S.Y. (2000). "Stochastic Prognostics for Rolling Element Bearings".
1003:
270:
265:
158:
Relative accuracy quantifies the accuracy relative to the actual time remaining before failure.
122:
1232:
1161:
1079:
909:
800:
757:
632:
589:
359:
255:
8:
1227:
474:
140:
Several prognostics performance metrics have evolved with consideration of these issues:
1210:
1083:
974:
913:
804:
761:
636:
593:
512:
Yu, Wei Kufi; Harris (2001). "A new stress-based fatigue life model for ball bearings".
363:
155:
specifications at all times an algorithm must improve with time to stay within the cone.
1103:
959:
679:
529:
494:
455:
409:
475:"Nonparametric time series modelling for industrial prognostics and health management"
1095:
925:
880:
847:
816:
773:
605:
533:
445:
330:
303:
1107:
683:
1143:
Centre for Prognostics and System Health Management at City University of Hong Kong
1087:
1036:
917:
872:
839:
808:
765:
675:
671:
644:
640:
597:
560:
521:
498:
486:
459:
437:
413:
401:
367:
81:
250:
921:
812:
769:
371:
1091:
843:
836:
2017 IEEE International Conference on Prognostics and Health Management (ICPHM)
660:"A data-model-fusion prognostic framework for dynamic system state forecasting"
659:
441:
1040:
525:
490:
430:
2015 First International Conference on Reliability Systems Engineering (ICRSE)
405:
1221:
1099:
929:
876:
820:
777:
609:
16:
This article is about the engineering discipline. For the medical term, see
1213:
develops methods and experiments for the prognostics of industrial systems.
601:
123:
Post-estimate fusion of model-based approaches with data-driven approaches
1183:
features one of the oldest tracks on prognostics and health management.
727:
260:
565:
548:
1164:
Program made substantial investments in PHM and related technologies
295:
17:
1125:
784:
472:
384:
1157:
is an annual international conference focusing exclusively on PHM.
1170:
893:
299:
Intelligent fault Diagnosis and Prognosis for Engineering Systems
1154:
1142:
989:
227:
1192:
1187:
The Society for Machinery Failure Prevention Technology (MFPT)
479:
The International Journal of Advanced Manufacturing Technology
745:
216:
1136:
1053:
899:
1203:
with run-to-failure data sets that are publicly accessible.
1137:
The Prognostics and Health Management Society (PHM Society)
1068:
1180:
1132:
International Journal of Prognostics and Health Management
944:
113:
866:
426:
213:
Examples of uncertainty quantification can be found in.
1131:
1186:
1171:
The Center for Advanced Life Cycle Engineering (CALCE)
1148:
827:
657:
131:
664:
Engineering Applications of Artificial Intelligence
349:
296:Vachtsevanos; Lewis, Roemer; Hess, and Wu (2006).
699:Journal of Engineering for Gas Turbines and Power
1219:
972:
942:
720:
473:Mosallam, A.; Medjaher, K; Zerhouni, N. (2013).
385:Mosallam, A.; Medjaher, K; Zerhouni, N. (2014).
326:Prognostics and Health Management of Electronics
790:
546:
579:
833:
168:
1026:
751:
573:
91:
1193:The Prognostics Center of Excellence (PCoE)
234:
217:Commercial hardware and software platforms
74:
987:
622:
564:
1155:The Annual Conference of the PHM Society
902:Mechanical Systems and Signal Processing
793:Mechanical Systems and Signal Processing
754:Mechanical Systems and Signal Processing
582:Mechanical Systems and Signal Processing
511:
352:Mechanical Systems and Signal Processing
192:Uncertainty in system degradation model:
696:
658:Liu, Jie; Wang, Ma; Yang, Yang (2012).
1220:
114:Pre-estimate fusion of models and data
957:
322:
1175:University of Maryland, College Park
394:Journal of Intelligent Manufacturing
185:Uncertainty in nominal system model:
104:
55:) or—in transportation applications—
13:
206:Uncertainty in failure thresholds:
14:
1249:
1118:
178:Uncertainty in system parameters:
132:Prognostic performance evaluation
41:prognostics and health management
1004:"Predictive Maintenance Toolbox"
547:Paris, P.C.; F. Erdogan (1963).
1062:
1047:
1020:
996:
981:
966:
951:
936:
756:. 52–53. Elsevier BV: 228–247.
709:
690:
651:
616:
676:10.1016/j.engappai.2012.02.015
645:10.1016/j.microrel.2010.01.006
540:
505:
466:
420:
378:
343:
316:
289:
1:
625:Microelectronics Reliability
553:Journal of Basic Engineering
7:
922:10.1016/j.ymssp.2017.10.015
813:10.1016/j.ymssp.2013.08.022
799:(2). Elsevier BV: 396–407.
770:10.1016/j.ymssp.2014.05.029
372:10.1016/j.ymssp.2008.09.006
243:
37:condition-based maintenance
10:
1254:
1092:10.1109/jsyst.2014.2343752
844:10.1109/icphm.2017.7998318
838:. IEEE. pp. 135–142.
442:10.1109/ICRSE.2015.7366504
323:Pecht, Michael G. (2008).
199:Uncertainty in prediction:
169:Uncertainty in prognostics
15:
1197:NASA Ames Research Center
1181:IEEE Aerospace Conference
1041:10.1108/02602280510585691
975:"Watchdog Agent® Toolkit"
526:10.1080/10402000108982420
491:10.1007/s00170-013-5065-z
406:10.1007/s10845-014-0933-4
92:Physics-based prognostics
57:vehicle health management
908:. Elsevier BV: 174–195.
877:10.1109/phm.2012.6228860
282:
276:Wind turbine prognostics
235:System-level prognostics
65:engine health management
49:system health management
75:Data-driven prognostics
1207:The PHM Research Group
973:National Instruments.
945:"Condition Monitoring"
943:National Instruments.
871:. IEEE. pp. 1–6.
602:10.1006/mssp.2000.1301
514:Tribology Transactions
271:Preventive maintenance
266:Predictive maintenance
29:remaining useful life
1162:Joint Strike Fighter
1072:IEEE Systems Journal
256:Planned obsolescence
1126:Prognostics Journal
1084:2015ISysJ...9.1197R
914:2018MSSP..103..174R
805:2014MSSP...45..396S
762:2015MSSP...52..228S
637:2010ESSFR...3.4.25P
594:2000MSSP...14..747L
364:2009MSSP...23.1586L
1211:FEMTO-ST Institute
47:), sometimes also
1238:Survival analysis
1149:China PHM Society
1008:www.mathworks.com
886:978-1-4577-1911-0
853:978-1-5090-5710-8
566:10.1115/1.3656903
451:978-1-4673-8557-2
336:978-0-470-27802-4
309:978-0-471-72999-0
105:Hybrid approaches
1245:
1112:
1111:
1078:(4): 1197–1207.
1066:
1060:
1059:
1051:
1045:
1044:
1024:
1018:
1017:
1015:
1014:
1000:
994:
993:
985:
979:
978:
970:
964:
963:
955:
949:
948:
940:
934:
933:
897:
891:
890:
864:
858:
857:
831:
825:
824:
788:
782:
781:
749:
743:
742:
740:
739:
724:
718:
716:researchgate.net
713:
707:
706:
694:
688:
687:
655:
649:
648:
620:
614:
613:
577:
571:
570:
568:
544:
538:
537:
509:
503:
502:
485:(5): 1685–1699.
470:
464:
463:
436:. pp. 1–7.
435:
424:
418:
417:
400:(5): 1037–1048.
391:
382:
376:
375:
358:(5): 1586–1599.
347:
341:
340:
320:
314:
313:
293:
82:domain knowledge
1253:
1252:
1248:
1247:
1246:
1244:
1243:
1242:
1218:
1217:
1216:
1201:data repository
1121:
1116:
1115:
1067:
1063:
1056:MFPT Conference
1052:
1048:
1025:
1021:
1012:
1010:
1002:
1001:
997:
990:"Predictronics"
988:Predictronics.
986:
982:
971:
967:
956:
952:
941:
937:
898:
894:
887:
865:
861:
854:
832:
828:
789:
785:
750:
746:
737:
735:
726:
725:
721:
714:
710:
695:
691:
656:
652:
621:
617:
578:
574:
545:
541:
510:
506:
471:
467:
452:
433:
425:
421:
389:
383:
379:
348:
344:
337:
321:
317:
310:
294:
290:
285:
280:
251:Futures studies
246:
240:these systems.
237:
219:
171:
134:
125:
116:
107:
94:
77:
21:
12:
11:
5:
1251:
1241:
1240:
1235:
1230:
1215:
1214:
1209:of the French
1204:
1190:
1184:
1178:
1168:
1158:
1152:
1146:
1140:
1134:
1129:
1122:
1120:
1119:External links
1117:
1114:
1113:
1061:
1046:
1035:(2): 114–122.
1019:
995:
980:
965:
950:
935:
892:
885:
859:
852:
826:
783:
744:
719:
708:
689:
670:(4): 814–823.
650:
631:(3): 317–323.
615:
588:(5): 747–762.
572:
559:(4): 528–534.
539:
504:
465:
450:
419:
377:
342:
335:
315:
308:
287:
286:
284:
281:
279:
278:
273:
268:
263:
258:
253:
247:
245:
242:
236:
233:
218:
215:
211:
210:
203:
196:
189:
182:
181:interval ones.
170:
167:
163:
162:
159:
156:
146:
133:
130:
124:
121:
115:
112:
106:
103:
93:
90:
76:
73:
9:
6:
4:
3:
2:
1250:
1239:
1236:
1234:
1231:
1229:
1226:
1225:
1223:
1212:
1208:
1205:
1202:
1198:
1194:
1191:
1188:
1185:
1182:
1179:
1176:
1172:
1169:
1166:
1163:
1159:
1156:
1153:
1150:
1147:
1144:
1141:
1138:
1135:
1133:
1130:
1127:
1124:
1123:
1109:
1105:
1101:
1097:
1093:
1089:
1085:
1081:
1077:
1073:
1065:
1057:
1050:
1042:
1038:
1034:
1030:
1029:Sensor Review
1023:
1009:
1005:
999:
991:
984:
976:
969:
961:
954:
946:
939:
931:
927:
923:
919:
915:
911:
907:
903:
896:
888:
882:
878:
874:
870:
863:
855:
849:
845:
841:
837:
830:
822:
818:
814:
810:
806:
802:
798:
794:
787:
779:
775:
771:
767:
763:
759:
755:
748:
733:
729:
723:
717:
712:
704:
700:
693:
685:
681:
677:
673:
669:
665:
661:
654:
646:
642:
638:
634:
630:
626:
619:
611:
607:
603:
599:
595:
591:
587:
583:
576:
567:
562:
558:
554:
550:
543:
535:
531:
527:
523:
519:
515:
508:
500:
496:
492:
488:
484:
480:
476:
469:
461:
457:
453:
447:
443:
439:
432:
431:
423:
415:
411:
407:
403:
399:
395:
388:
381:
373:
369:
365:
361:
357:
353:
346:
338:
332:
328:
327:
319:
311:
305:
301:
300:
292:
288:
277:
274:
272:
269:
267:
264:
262:
259:
257:
254:
252:
249:
248:
241:
232:
229:
223:
214:
207:
204:
200:
197:
193:
190:
186:
183:
179:
176:
175:
174:
166:
160:
157:
154:
150:
147:
143:
142:
141:
138:
129:
120:
111:
102:
98:
89:
85:
83:
72:
70:
66:
62:
58:
54:
50:
46:
42:
38:
34:
30:
25:
19:
1075:
1071:
1064:
1055:
1049:
1032:
1028:
1022:
1011:. Retrieved
1007:
998:
983:
968:
953:
938:
905:
901:
895:
868:
862:
835:
829:
796:
792:
786:
753:
747:
736:. Retrieved
734:. 2017-12-11
731:
722:
711:
702:
698:
692:
667:
663:
653:
628:
624:
618:
585:
581:
575:
556:
552:
542:
520:(1): 11–18.
517:
513:
507:
482:
478:
468:
429:
422:
397:
393:
380:
355:
351:
345:
325:
318:
298:
291:
238:
224:
220:
212:
205:
198:
191:
184:
177:
172:
164:
152:
148:
139:
135:
126:
117:
108:
99:
95:
86:
78:
68:
64:
60:
56:
52:
48:
44:
40:
32:
28:
23:
22:
1233:Maintenance
960:"Webaccess"
958:Advantech.
732:PHM Society
188:the system.
24:Prognostics
1228:Prediction
1222:Categories
1013:2019-07-11
738:2020-06-13
261:Prediction
1100:1932-8184
930:0888-3270
821:0888-3270
778:0888-3270
610:0888-3270
534:110685617
329:. Wiley.
302:. Wiley.
18:prognosis
1108:22285080
684:33838225
244:See also
1195:at the
1173:at the
1080:Bibcode
910:Bibcode
801:Bibcode
758:Bibcode
633:Bibcode
590:Bibcode
499:3182648
460:2931906
414:1978502
360:Bibcode
145:action.
1106:
1098:
928:
883:
850:
819:
776:
682:
608:
532:
497:
458:
448:
412:
333:
306:
228:MATLAB
1104:S2CID
680:S2CID
530:S2CID
495:S2CID
456:S2CID
434:(PDF)
410:S2CID
390:(PDF)
283:Notes
63:) or
1160:The
1096:ISSN
926:ISSN
881:ISBN
848:ISBN
817:ISSN
774:ISSN
606:ISSN
446:ISBN
331:ISBN
304:ISBN
1088:doi
1037:doi
918:doi
906:103
873:doi
840:doi
809:doi
766:doi
703:136
672:doi
641:doi
598:doi
561:doi
522:doi
487:doi
438:doi
402:doi
368:doi
153:α-λ
149:α-λ
69:EHM
61:VHM
53:SHM
45:PHM
33:RUL
1224::
1102:.
1094:.
1086:.
1074:.
1033:25
1031:.
1006:.
924:.
916:.
904:.
879:.
846:.
815:.
807:.
797:45
795:.
772:.
764:.
730:.
701:.
678:.
668:25
666:.
662:.
639:.
629:50
627:.
604:.
596:.
586:14
584:.
557:85
555:.
551:.
528:.
518:44
516:.
493:.
483:69
481:.
477:.
454:.
444:.
408:.
398:27
396:.
392:.
366:.
356:23
354:.
84:.
1167:.
1110:.
1090::
1082::
1076:9
1058:.
1043:.
1039::
1016:.
992:.
977:.
962:.
947:.
932:.
920::
912::
889:.
875::
856:.
842::
823:.
811::
803::
780:.
768::
760::
741:.
705:.
686:.
674::
647:.
643::
635::
612:.
600::
592::
569:.
563::
536:.
524::
501:.
489::
462:.
440::
416:.
404::
374:.
370::
362::
339:.
312:.
67:(
59:(
51:(
43:(
31:(
20:.
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.