TY - GEN
T1 - Probabilistic aging pipe strength estimation using multimodality information fusion
AU - Chen, Jie
AU - Liu, Yongming
N1 - Funding Information:
The work in this study was supported by DOT PHMSA through Gas Technology Institute (693JK31810003: Program Manager: Daniel Ersoy). The support is greatly acknowledged.
Publisher Copyright:
© 2019 Prognostics and Health Management Society. All rights reserved.
PY - 2019/9/23
Y1 - 2019/9/23
N2 - Accurate pipe material strength estimation is critical for the integrity and risk assessment of aging pipeline infrastructure systems. To predict the strength without interrupting the serviceability of the pipeline, inference methods are used through the relationship between the bulk yield tensile strength and surface material properties from nondestructive testing, such as chemical composition, microstructure images, and hardness testing. In order to make the best of information provided by multimodality surface measurements, Bayesian model averaging (BMA) method is used in this paper to integrate the information from various types of surface measurements for a more accurate bulk strength estimation. The models being considered are constructed by randomly combining the multimodality surface measurements and each case of linear combinations is included. The models considered are assessed by assigning different weights based on the posterior model probability. Markov Chain Monte Carlo sampling provides an effective way for numerically computing the marginal likelihoods, which are essential for obtaining the posterior model probabilities. To avoid the risk of overfitting, BMA is implemented to account for model uncertainty. The predictive performance of a single model and BMA are compared by logarithmic scoring rule. The data collected from industry are used for demonstration and model predictive performance assessment. It is shown that the Bayesian model averaging approach can provide more reliable results in predicting the strength of the aging pipelines.
AB - Accurate pipe material strength estimation is critical for the integrity and risk assessment of aging pipeline infrastructure systems. To predict the strength without interrupting the serviceability of the pipeline, inference methods are used through the relationship between the bulk yield tensile strength and surface material properties from nondestructive testing, such as chemical composition, microstructure images, and hardness testing. In order to make the best of information provided by multimodality surface measurements, Bayesian model averaging (BMA) method is used in this paper to integrate the information from various types of surface measurements for a more accurate bulk strength estimation. The models being considered are constructed by randomly combining the multimodality surface measurements and each case of linear combinations is included. The models considered are assessed by assigning different weights based on the posterior model probability. Markov Chain Monte Carlo sampling provides an effective way for numerically computing the marginal likelihoods, which are essential for obtaining the posterior model probabilities. To avoid the risk of overfitting, BMA is implemented to account for model uncertainty. The predictive performance of a single model and BMA are compared by logarithmic scoring rule. The data collected from industry are used for demonstration and model predictive performance assessment. It is shown that the Bayesian model averaging approach can provide more reliable results in predicting the strength of the aging pipelines.
UR - http://www.scopus.com/inward/record.url?scp=85083970608&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85083970608&partnerID=8YFLogxK
U2 - 10.36001/phmconf.2019.v11i1.817
DO - 10.36001/phmconf.2019.v11i1.817
M3 - Conference contribution
AN - SCOPUS:85083970608
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 -