Real-Time Estimation of Mean Remaining Life Using Sensor-Based Degradation Models

Advances in sensor technology have led to an increased interest in using degradation-based sensory information to predict the remaining lives of partially degraded components and systems. This paper presents a stochastic degradation modeling framework for computing and continuously updating remaining life distributions (RLDs) using in situ degradation signals acquired from individual components during their operational lives. Unfortunately, these sensory-updated RLDs cannot be characterized using parametric distributions and their moments do not exist. Such difficulties hinder the implementation of this sensor-based framework, especially from the standpoint of computational efficiency of embedded algorithms. In this paper, we identify an approximate procedure by which we can compute a conservative mean of the sensory-updated RLDs and express the mean and variance using closed form expressions that are easy to evaluate. To accomplish this, we use the first passage time of Brownian motion with positive drift, which follows an inverse Gaussian distribution, as an approximation of the remaining life. We then show that the mean of the inverse Caussian is a conservative lower bound of the mean remaining life using Jensen's inequality. The results are validated using real-world vibration-based degradation information.

[1]  T. A. Harris,et al.  Rolling Bearing Analysis , 1967 .

[2]  Richard A. Johnson,et al.  Applied Multivariate Statistical Analysis , 1983 .

[3]  Munir Ahmad,et al.  Bernstein reliability model: Derivation and estimation of parameters , 1984 .

[4]  J. Leroy Folks,et al.  The Inverse Gaussian Distribution: Theory: Methodology, and Applications , 1988 .

[5]  G. Casella,et al.  Statistical Inference , 2003, Encyclopedia of Social Network Analysis and Mining.

[6]  W. Meeker.  Accelerated Testing: Statistical Models, Test Plans, and Data Analyses , 1991 .

[7]  K. Doksum,et al.  Models for variable-stress accelerated life testing experiments based on Wiener processes and the inverse Gaussian distribution , 1992 .

[8]  C. Joseph Lu,et al.  Using Degradation Measures to Estimate a Time-to-Failure Distribution , 1993 .

[9]  A. Basu,et al.  The Inverse Gaussian Distribution , 1993 .

[10]  K. F. Martin,et al.  A review by discussion of condition monitoring and fault diagnosis in machine tools , 1994 .

[11]  Kai Yang,et al.  Statistical surface roughness checking procedure based on a cutting tool wear model , 1994 .

[12]  Loon Ching Tang,et al.  Reliability prediction using nondestructive accelerated-degradation data: case study on power supplies , 1995 .

[13]  G A Whitmore,et al.  Estimating degradation by a wiener diffusion process subject to measurement error , 1995, Lifetime data analysis.

[15]  G A Whitmore,et al.  Modelling Accelerated Degradation Data Using Wiener Diffusion With A Time Scale Transformation , 1997, Lifetime data analysis.

[16]  Kai Yang,et al.  Degradation Reliability Assessment Using Severe Critical Values , 1998 .

[17]  G. A. Whitmore,et al.  Failure Inference From a Marker Process Based on a Bivariate Wiener Model , 1998, Lifetime data analysis.

[18]  O.V. Thorsen,et al.  Failure identification and analysis for high voltage induction motors in petrochemical industry , 1998, Conference Record of 1998 IEEE Industry Applications Conference. Thirty-Third IAS Annual Meeting (Cat. No.98CH36242).

[19]  Loon Ching Tang,et al.  Mean residual life of lifetime distributions , 1999 .

[20]  Zhenlin Yang.  Maximum likelihood predictive densities for the inverse Gaussian distribution with applications to reliability and lifetime predictions , 1999 .

[21]  Y Shao,et al.  Prognosis of remaining bearing life using neural networks , 2000 .

[22]  M E Robinson,et al.  Bayesian Methods for a Growth-Curve Degradation Model with Repeated Measures , 2000, Lifetime data analysis.

[23]  Nagi Gebraeel,et al.  Residual life predictions from vibration-based degradation signals: a neural network approach , 2004, IEEE Transactions on Industrial Electronics.

[24]  W. J. Padgett,et al.  Inference from Accelerated Degradation and Failure Data Based on Gaussian Process Models , 2004, Lifetime data analysis.

[25]  Suk Joo Bae,et al.  A Nonlinear Random-Coefficients Model for Degradation Testing , 2004, Technometrics.

[26]  Rong Li,et al.  Residual-life distributions from component degradation signals: A Bayesian approach , 2005 .

[27]  Steven M. Cox,et al.  Stochastic models for degradation-based reliability , 2005 .

[28]  Chanseok Park,et al.  New cumulative damage models for failure using stochastic processes as initial damage , 2005, IEEE Transactions on Reliability.

[29]  Nagi Gebraeel,et al.  Sensory-Updated Residual Life Distributions for Components With Exponential Degradation Patterns , 2006, IEEE Transactions on Automation Science and Engineering.

[30]  Chanseok Park,et al.  Stochastic degradation models with several accelerating variables , 2006, IEEE Transactions on Reliability.

[31]  Suk Joo Bae,et al.  Degradation models and implied lifetime distributions , 2007, Reliab. Eng. Syst. Saf..