Cracking the Code: Modeling Ultrasonic Wall Loss Data to Predict Cast Iron Pipe Failure
This study introduces an innovative application of ultrasonic technology coupled with finite element modeling for the early detection of damage mechanisms in cast iron pipes that precede catastrophic failure due to cracking. Spun Cast Iron Pipe emerged as the predominant choice for water conveyance in the United States during the period spanning from 1920 to 1950, consequently establishing itself as a pivotal component of the nation’s water infrastructure. However, the deterioration of these pipes over time has presented substantial challenges. It is well-established that corrosion is the principal means of deterioration of cast iron pipes and that stress concentrations near corrosion anomalies reduce structural capacity of the pipe wall and ultimately lead to splitting and failure. Prior research has studied pipe deterioration by creating artificial anomalies on the pipe experimentally or by means of simulation. However, there is limited study for real-world corrosion scenarios and the ensuing crack propagation phenomena. On June 28, 2023, a 20-in. cast iron pipe, previously inspected by ultrasonic technology, failed in Johnson County, Kansas. The inspection had identified 15 distinct regions of wall thinning and anomalies (most likely graphitic corrosion) distributed across the pipe’s length. This information was used to generate a finite element model of the pipe wall, and a study was conducted to determine the model’s capacity to predict the origin and trajectory of cracks witnessed in the forensic evaluation of the failed pipe stick. Accurate understanding and prediction of such failures in cast iron pipe will enable timely maintenance and replacement decisions to ensure the reliability and safety of water transmission systems. This research contributes to the advancement of infrastructure health monitoring and the sustainable pipeline management of urban utilities.
Authors
- Masood Hajali (ORCID: https://orcid.org/0000-0002-3585-8125)
- Ikram Efaz
- Ashan McNealy
- Peter Gaskamp
- Leo Liu
Institutions
- Xylem (United States) (US)
Publication Details
- Published
- 2024-08-30
- DOI
- https://doi.org/10.1061/9780784485583.027
- Citations
- 1
- Primary Topic
- Non-Destructive Testing Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.21