Session: DA-08-02 FFS Involving Piping and Pipelines

Paper Number: 101256

101256 - Validation of Fatigue Crack Growth Modeling Solutions Using Measurements Collected on Api X65 Piping Specimens 

Building confidence in different modeling techniques to assess damage tolerance capabilities of components under service conditions relies on a continuous verification and validation effort. The verification benchmarking (usually conducted for mode I stress intensity factors, K_I) is accomplished by comparison against analytical or high-resolution numerical solutions while for the validation task, lab-controlled experimental measurements (i.e., crack size vs. cycles) are used as a reference for a complete remaining useful life solution. Different mechanical testing conditions and multiple specimens are often considered to capture experimental variability and provide a large set of measurements that can be used for numerical solution validation and uncertainty quantification.

In this study, four fatigue crack growth experimental measurements, presented by Li et. al.**, were used as a validation reference set for an incremental 3D FE modeling procedure and for a reduced order model-based application (AFGROW). The test procedure consists of a four-point bend X65 pipe specimen subjected to two repeating R-ratio constant amplitude loading blocks to allow formation of beach marks that can be used post-failure to identify crack size and shape during the experiment. The two modeling processes capture all details related to experimental procedure: overall geometry of the specimen, crack location, and an R=0.1 loading block of 10000 cycles followed by R=0.5 for 5000 cycles. Fatigue crack growth rate data provided by BS 7910 was used to obtain the 3D FEM based explicit crack growth solution as well as the equivalent solution using the reduced-order model. Digitized post-failure beach mark data was used as a reference to validate against the crack fronts generated in the numerical solution for each loading block. Fatigue crack growth rate uncertainty within the bounds provided by BS 7910 is considered in the validation benchmark.

The solution difference between the reduced order model and the explicit 3D crack propagation modeling procedure is also investigated by enforcing the crack front increments to be elliptical shapes in the 3D FE modeling procedure to be consistent with the 2-parameter reduced order modeling implementation. This assessment serves as a verification benchmark for the two modeling processes.

 

**Li, Z, Jiang, X., Hopman, H, Liu, Z., An investigation on the circumferential surface crack growth in steel pipes subjected to fatigue bending, Theoretical and Applied Fracture Mechanics, pp. 205, 2020.

Presenting Author: Adrian Loghin Simmetrix Inc.

Presenting Author Biography: Dr. Adrian Loghin received his Ph.D. degree in the area of Engineering Mechanics from Clemson University in 2001. After graduation, he continued his career at General Electric Global Research Center in Niskayuna, NY for 16 years mostly in the Lifing Technologies Lab as an expert in Fracture Mechanics and Fatigue. In 2017 he joined Simmetrix as a senior application engineer responsible with development of SimModeler Crack, an application designed for fatigue crack growth simulations. Adrian holds two US patents and coauthored more than 15 publications.

Authors:

Adrian Loghin Simmetrix Inc.
James Harter LexTech Inc.