Application of Bayesian estimation to structural health monitoring of fatigue cracks in welded steel pipe

Abstract Vibration induced fatigue is a well-known problem in oil and gas piping systems. However the use of vibration data to detect damage is not an easy task without a priori knowledge of the undamaged condition. In this paper, level 1: Detection and level 2: Localization of damage from structural health monitoring strategies is adapted for damage identification. An experimental trial of acoustic emission monitoring was run to monitor fatigue damage during a full-scale resonance fatigue test of a girth-welded steel pipe from healthy condition until failure. The pipe was excited into the first mode of vibration using a resonance fatigue testing machine in order to determine the high-cycle fatigue strength of the weld. The information provided by acoustic emission monitoring is useful in evaluating the condition of the pipe during the test and the occurrence of cracking before failure. However, the acoustic emission signals are embedded in noise. To overcome this problem, the signals from different combinations of sensors were recursively cross-correlated, which provides for the derivation of a new effective coefficient (EC) parameter for Bayesian estimation. This parameter is useful for evaluating uncertainty arising from the signals that contribute to source localization errors. The estimation finds the most probable parameters corresponding to cracks using prior knowledge derived from standard pencil lead break tests. The proposed method demonstrates a strong relationship between the acoustic emission energy and the estimated coefficients. A high correlation between signals was found to be associated with cracking, and a low correlation between signals was found to be associated with random signals or noise. The method will be useful for monitoring the condition of piping to manage the risk of vibration induced fatigue failure.

[1]  Hoon Sohn,et al.  Statistical Damage Classification Under Changing Environmental and Operational Conditions , 2002 .

[2]  Hoon Sohn,et al.  Damage diagnosis using time series analysis of vibration signals , 2001 .

[3]  N. Metropolis,et al.  The Monte Carlo method. , 1949 .

[4]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Hoon Sohn,et al.  A review of structural health monitoring literature 1996-2001 , 2002 .

[6]  J. Beck,et al.  Updating Models and Their Uncertainties. I: Bayesian Statistical Framework , 1998 .

[7]  A. K. Rao.  Acoustic Emission and Signal Analysis , 1990 .

[8]  E. Peter Carden,et al.  Vibration Based Condition Monitoring: A Review , 2004 .

[9]  Charles R. Farrar,et al.  A summary review of vibration-based damage identification methods , 1998 .

[10]  Andrew Thomas,et al.  WinBUGS - A Bayesian modelling framework: Concepts, structure, and extensibility , 2000, Stat. Comput..

[11]  Hoon Sohn,et al.  Structural Health Monitoring Using Statistical Process Control , 2000 .

[12]  Shih-Lin Hung,et al.  A Wavelet‐Based Approach to Identifying Structural Modal Parameters from Seismic Response and Free Vibration Data , 2005 .

[13]  Norris Stubbs,et al.  Damage identification in beam-type structures: frequency-based method vs mode-shape-based method , 2003 .

[14]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[15]  John Alexander Steel,et al.  Acoustic Emission Source Location for Steel Pipe and Pipeline Applications: The Role of Arrival Time Estimation , 2006 .

[16]  H Prosser William,et al.  AE Source Orientation by Plate Wave Analysis , 1994 .

[17]  Hani G. Melhem,et al.  Signal Pattern Recognition for Damage Diagnosis in Structures , 2012, Comput. Aided Civ. Infrastructure Eng..

[18]  James M. Ricles,et al.  Damage Detection in Structures by Modal Vibration Characterization , 1999 .

[19]  E. George,et al.  Journal of the American Statistical Association is currently published by American Statistical Association. , 2007 .

[20]  Pizhong Qiao,et al.  Vibration-based Damage Identification Methods: A Review and Comparative Study , 2011 .

[21]  James M. W. Brownjohn,et al.  ARMA modelled time-series classification for structural health monitoring of civil infrastructure , 2008 .

[22]  J. Beck,et al.  Bayesian Updating of Structural Models and Reliability using Markov Chain Monte Carlo Simulation , 2002 .

[23]  M. Gorman,et al.  Source location in thin plates using cross-correlation , 1991 .