Prediction of crack growth based on a hierarchical diffusion model

A general Bayesian approach for stochastic versions of deterministic growth models is presented to provide predictions for crack propagation in an early stage of the growth process. To improve the prediction, the information of other crack growth processes is used in a hierarchical (mixed-effects) model. Two stochastic versions of a deterministic growth model are considered. One is a nonlinear regression setup where the trajectory is assumed to be the solution of an ordinary differential equation with additive errors. The other is a diffusion model defined by a stochastic differential equation (SDE) where increments have additive errors. Six growth models in the two versions are compared with respect to their ability to predict the crack propagation in a large data example. Two of them are based on the classical Paris-Erdogan law for crack growth, and four are other widely used growth models. It turned out that the three-parameter Paris-Erdogan model and the Weibull model provide the best results followed by the logistic model. Suprisingly, the SDE approach has no advantage for the prediction compared with the nonlinear regression setup.

[1]  Christiane Fuchs,et al.  Inference for Diffusion Processes , 2013 .

[2]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[3]  P. Fearnhead,et al.  Exact and computationally efficient likelihood‐based estimation for discretely observed diffusion processes (with discussion) , 2006 .

[4]  Juan Antonio Cano,et al.  Approximation of the posterior density for diffusion processes , 2006 .

[5]  P. Protter.  Stochastic integration and differential equations , 1990 .

[6]  Francis Tuerlinckx,et al.  A Hierarchical Ornstein–Uhlenbeck Model for Continuous Repeated Measurement Data , 2009 .

[7]  P. Goel,et al.  The Statistical Nature of Fatigue Crack Propagation , 1979 .

[8]  M. Kenward,et al.  Parametric modelling of growth curve data: An overview , 2001 .

[9]  A. Raftery,et al.  Strictly Proper Scoring Rules, Prediction, and Estimation , 2007 .

[10]  Bradley P. Carlin,et al.  Bayesian Methods for Data Analysis , 2008 .

[11]  Sophie Donnet,et al.  Bayesian Analysis of Growth Curves Using Mixed Models Defined by Stochastic Differential Equations , 2010, Biometrics.

[12]  Wen-Fang Wu,et al.  Probabilistic models of fatigue crack propagation and their experimental verification , 2004 .

[13]  Asok Ray.  Stochastic modeling of fatigue crack dynamics for risk analysis and remaining life prediction , 1998, Proceedings of the 1998 American Control Conference. ACC (IEEE Cat. No.98CH36207).

[14]  F. Tuerlinckx,et al.  The linear mixed model and the hierarchical Ornstein-Uhlenbeck model: some equivalences and differences. , 2011, The British journal of mathematical and statistical psychology.

[15]  Boris A. Zárate,et al.  Bayesian model updating and prognosis of fatigue crack growth , 2012 .

[16]  J. Yang,et al.  On statistical moments of fatigue crack propagation , 1983 .

[17]  Eugene Demidenko,et al.  Mixed Models: Theory and Applications with R , 2013 .

[18]  Andrea De Gaetano,et al.  MIXED EFFECTS IN STOCHASTIC DIFFERENTIAL EQUATION MODELS , 2005 .

[19]  Guillermo Marshall,et al.  Non‐linear random effects models with continuous time autoregressive errors: a Bayesian approach , 2006, Statistics in medicine.

[20]  H. Sørensen.  Parametric Inference for Diffusion Processes Observed at Discrete Points in Time: a Survey , 2004 .

[21]  Anne S. Kiremidjian,et al.  Stochastic modeling of fatigue crack growth , 1988 .

[22]  A. Pedersen.  A new approach to maximum likelihood estimation for stochastic differential equations based on discrete observations , 1995 .

[23]  Umberto Picchini,et al.  Stochastic Differential Mixed‐Effects Models , 2010 .

[24]  N. Limnios,et al.  Piecewise deterministic Markov processes applied to fatigue crack growth modelling , 2009 .

[25]  S. Winterstein,et al.  Random Fatigue: From Data to Theory , 1992 .

[26]  Asok Ray,et al.  Stochastic modeling of fatigue crack dynamics for on-line failure prognostics , 1996, IEEE Trans. Control. Syst. Technol..