Pipeline dents are a major threat to the integrity and safety of aging oil and gas transmission systems. While API 1183 provides methodologies for dent assessment, it has several limitations. Traditional strain assessment methods rely heavily on 2D cross-sections, assuming maximum strain or minimum radius of curvature occurs at the dent’s apex, and do not distinguish between restrained and un-restrained dents. Furthermore, fatigue evaluation based on regression models often leads to overly conservative and inconsistent outcomes. This paper presents a novel framework to address these limitations and enhance pipeline safety.
The process begins with ILI data collection, followed by an automated Quality Assurance/Quality Control (QA/QC) process utilizing machine learning to resolve data inconsistencies. The framework employs filtering and smoothing techniques to refine the ILI data. Determining the restraint condition is a key step, guiding the appropriate assessment workflow. Un-restrained dent shapes are scaled appropriately to simulate indentation depth which is always greater than ILI reported depth.
Strain analysis follows both API 1183 guidelines and recent recommendations published in the PRCI report “Improve Dent/Cracking Assessment Methods, PR-214-203806”. Fatigue life is estimated by transforming ILI caliper data into a 3D finite element analysis (FEA) mesh, enabling a rigorous assessment of dent severity.
A critical advancement in this framework is the use of non-linear material properties in the screening level FEA, calibrated to account for residual stresses, to improve stress estimation for restrained dents. By incorporating estimated approximate residual stresses from strain data, the resulting stress values align closely with Level 3 non-linear FEA results, enhancing both accuracy and efficiency in fatigue assessments. Other advancements include the 3D evaluation of dent curvature to identify potential high-strain regions beyond the dent's deepest point, and the integration of IMU data for identifying dents in non-straight pipeline sections.
This paper presents the technical development and real-world applications of the framework, demonstrating its effectiveness in managing pipeline integrity through risk-informed decision making.
Available after the event



