TY - GEN
T1 - Semi-supervised constrained hidden markov model using multiple sensors for remaining useful life prediction and optimal predictive maintenance
AU - Zhao, Xinyu
AU - Kang, Yunyi
AU - Yan, Hao
AU - Ju, Feng
N1 - Publisher Copyright:
© 2019 Prognostics and Health Management Society. All rights reserved.
PY - 2019/9/23
Y1 - 2019/9/23
N2 - Remaining Useful Life (RUL) estimation is critical in many engineering systems where proper predictive maintenance is needed to increase a unit's effectiveness and reduce time and cost of repairing. Typically for such systems, multiple sensors are normally used to monitor performance, which create difficulties for system state identification. In this paper, we propose a semi-supervised left-to-right constrained Hidden Markov Model (HMM) model, which is the first in the literature to simultaneously address the challenges of semisupervised setting, left-to-right constraint, and monotonicity constraint in a multiple-sensor setting. This proposed method is also effective in estimating the RUL while capturing the jumps among states in condition dynamics. In addition, based on the HMM model learned from multiple sensors, we build a Partial Observable Markov Decision Process (POMDP) to demonstrate how such RUL estimation can be effectively used for optimal preventative maintenance decision making. We apply this technique to the NASA Engine degradation data and demonstrate the effectiveness of the proposed method.
AB - Remaining Useful Life (RUL) estimation is critical in many engineering systems where proper predictive maintenance is needed to increase a unit's effectiveness and reduce time and cost of repairing. Typically for such systems, multiple sensors are normally used to monitor performance, which create difficulties for system state identification. In this paper, we propose a semi-supervised left-to-right constrained Hidden Markov Model (HMM) model, which is the first in the literature to simultaneously address the challenges of semisupervised setting, left-to-right constraint, and monotonicity constraint in a multiple-sensor setting. This proposed method is also effective in estimating the RUL while capturing the jumps among states in condition dynamics. In addition, based on the HMM model learned from multiple sensors, we build a Partial Observable Markov Decision Process (POMDP) to demonstrate how such RUL estimation can be effectively used for optimal preventative maintenance decision making. We apply this technique to the NASA Engine degradation data and demonstrate the effectiveness of the proposed method.
UR - http://www.scopus.com/inward/record.url?scp=85083996693&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85083996693&partnerID=8YFLogxK
U2 - 10.36001/phmconf.2019.v11i1.851
DO - 10.36001/phmconf.2019.v11i1.851
M3 - Conference contribution
AN - SCOPUS:85083996693
T3 - Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM
BT - Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM
A2 - Clements, N. Scott
A2 - Zhang, Bin
A2 - Saxena, Abhinav
PB - Prognostics and Health Management Society
T2 - 11th Annual Conference of the Prognostics and Health Management Society, PHM 2019
Y2 - 23 September 2019 through 26 September 2019
ER -