A Bayesian approach has been applied to inline inspection (ILI) and non-destructive examination (NDE) field measurement data to increase confidence in estimated feature depths. Multiple crack depth measurements were used to improve the characterization of ILI sizing error bias and to update the maximum depth distribution of individual crack features.
The paper—A Bayesian Approach for Effective Use of Multiple Measurements of Crack Depths—describes the application of Bayesian methodology to three datasets, two of which collected over a series of Pipeline Research Council International (PRCI) projects. The results show that Bayesian updating, when compared to simple averaging of the combined data from multiple ILI runs:
- Resulted in a higher confidence in the estimated tool sizing bias; and
- Increased confidence in the individual feature depth estimates.
Dr. Smitha Koduru is presenting the results of this project at the upcoming virtual International Pipeline Conference. This presentation is available on-demand during the conference. You will learn about the application of a Bayesian approach and how it was used to increase size estimates.
The entire paper—IPC2020-9307 A Bayesian Approach for Effective Use of Multiple Measurements of Crack Depths —is available through the 2020 International Pipeline Conference.
Dr. Smitha Koduru is the Manager of C-FER’s Pipeline Integrity and Operations department. Her focus is on risk assessment and detailed analyses of structural systems.