Multi-sensor prognostics modeling for applications with highly incomplete signals

Abstract

Multi-stream degradation signals have been widely used to predict the residual useful lifetime of partially degraded systems. To achieve this goal, most of the existing prognostics models assume that degradation signals are complete, i.e., they are observed continuously and frequently at regular time grids. In reality, however, degradation signals are often (highly) incomplete, i.e., containing missing and corrupt observations. Such signal incompleteness poses a significant challenge for the parameter estimation of prognostics models. To address this challenge, this article proposes a prognostics methodology that is capable of using highly incomplete multi-stream degradation signals to predict the residual useful lifetime of partially degraded systems. The method first employs multivariate functional principal components analysis to fuse multi-stream signals. Next, the fused features are regressed against time-to-failure using (log)-location-scale regression. To estimate the fused features using incomplete multi-stream degradation signals, we develop two computationally efficient algorithms: subspace detection and signal recovery. The performance of the proposed prognostics methodology is evaluated using simulated datasets and a degradation dataset of aircraft turbofan engines from the NASA repository.

Keywords:

• RUL
• degradation modelling
• multi-stream signal fusion
• missing data

Acknowledgments

The authors thank the Editor, AE and referees for their valuable comments. The research of Paynabar is supported by the NSF grants CMMI-1839591.

Additional information

Notes on contributors

Xiaolei Fang

Xiaolei Fang received his B.S. degree in Mechanical Engineering from the University of Science and Technology Beijing, China, in 2008 and an M.S. degree in Statistics and a Ph.D. in Industrial Engineering from the Georgia Institute of Technology, Atlanta, GA, in 2016 and 2018, respectively. Currently, he is an assistant professor in the Edward P. Fitts Department of Industrial and Systems Engineering, North Carolina State University, Raleigh, NC. His research interests lie in the field of industrial data analytics for High-Dimensional and Big Data applications in the energy, manufacturing, and service sectors. Specifically, he focuses on addressing analytical, computational, and scalability challenges associated with the development of statistical and optimization methodologies for analyzing massive amounts of complex data structures for real-time asset management and optimization. He is a member of IISE and INFORMS.

Hao Yan

Hao Yan received a B.S. degree in physics from Peking University, Beijing, China, in 2011. He also received an M.S. degree in Statistics, an M.S. degree in computational science and engineering, and a Ph.D. degree in industrial engineering from Georgia Institute of Technology, Atlanta, in 2015, 2016, 2017, respectively. Currently, he is an assistant professor in the School of Computing, Informatics, & Decision Systems Engineering at Arizona State University. His research interests focus on developing scalable statistical learning algorithms for large-scale highdimensional data with complex heterogeneous structures to extract useful information for the purpose of system performance assessment, anomaly detection, intelligent sampling and decision making. He is a member of INFORMS and IISE.

Nagi Gebraeel

Nagi Gebraeel is the Georgia Power Early Career professor and Professor in the H. Milton Stewart School of Industrial and Systems Engineering at the Georgia Institute of Technology. His research interests lie at the intersection of industrial predictive analytics and decision optimization models for large scale power generation applications. Dr. Gebraeel serves as an associate director at Georgia Tech's Strategic Energy Institute and the director of the Analytics and Prognostics Systems laboratory at Georgia Tech's Manufacturing Institute. Dr. Gebraeel was the former president of the IIE Quality and Reliability Engineering Division, and is currently a member of INFORMS.

Kamran Paynabar

Kamran Paynabar received his B.Sc. and M.Sc. in industrial engineering from Iran University of Science and Technology and Azad University in 2002 and 2004, respectively, and his Ph.D. in industrial and operations engineering from The University of Michigan in 2012. He also holds an M.A. in statistics from The University of Michigan. Currently, he is a Fouts Family Career Professor and associate professor at the H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta. His research interests include data fusion for multi-stream waveform signals and functional data, engineering-driven statistical modeling, sensor selection in distributed sensing networks, probabilistic graphical models, and statistical learning with applications in manufacturing and healthcare systems. Dr. Paynabar is a member of Institute of Industrial and Systems Engineers, The Institute for Operations Research and the Management Sciences, and IEEE Robotics and Automation Society.