Mechanical Performance of Buried Continuous Steel Pipes Under Reverse Fault Action

Literature Overview

This paper by Zhao Xu, Cui Jianyang, Zhong Zilan, and Du Xiuli, published in the Journal of Beijing University of Technology (2022, Vol. 48, No. 7, pp. 729-738), investigates the mechanical behavior of buried continuous steel pipes crossing reverse faults in clay soil. The research, funded by the National Natural Science Foundation (51978020) and the Guangdong Provincial Key Laboratory of Earthquake Engineering and Applied Technology (2017B030314068), employs three-dimensional nonlinear finite element analysis to model pipe-soil interaction under reverse fault displacement. The study identifies three typical failure modes and determines the critical fault displacement quantities corresponding to each mode under varying fault dip angles, internal pressures, and pipe diameter-to-thickness ratios.

Numerical Modeling Approach

The finite element model captures the complex three-dimensional pipe-soil interaction under reverse fault displacement. Key modeling aspects include:

Modeling Component	Description	Element Type
Steel pipe	Cylindrical shell, elastic-plastic material (von Mises yield criterion)	4-node shell element (S4R)
Clay soil	Mohr-Coulomb plasticity model, undrained condition	8-node solid element (C3D8R)
Pipe-soil interface	Frictional contact with penalty formulation	Surface-to-surface contact
Fault plane	Rigid displacement boundary, prescribed displacement history	Displacement boundary condition

The model accounts for large deformations (geometric nonlinearity), material nonlinearity (plasticity in both pipe and soil), and contact nonlinearity (pipe-soil separation and friction). The soil is modeled with realistic clay properties: undrained shear strength of 50-200 kPa, Poisson's ratio of 0.49 (undrained), and a friction angle of 0-5° (undrained condition).

Failure Mode Analysis

The study identifies three distinct failure modes for buried steel pipes under reverse fault displacement:

Failure Mode 1: Tensile Failure (Rupture)

This mode occurs when the axial tensile strain at the fault crossing exceeds the material's ultimate tensile strain capacity. The critical fault displacement for tensile failure depends on:

Fault dip angle: Steeper dip angles (75°) produce smaller critical displacements because the fault movement is more directly aligned with the pipe axis.
Internal pressure: Higher internal pressure increases the critical displacement by reducing the effective axial strain through the Poisson effect and membrane stress interaction.
Diameter-to-thickness ratio (D/t): Larger D/t ratios reduce the critical displacement because thinner walls are more susceptible to strain concentration.

Typical critical fault displacements for tensile failure range from 0.5D to 2.0D, where D is the pipe outer diameter.

Failure Mode 2: Local Buckling

Local buckling occurs when the compressive hoop stress and axial compressive stress at the fault crossing exceed the local buckling capacity of the pipe wall. This mode is characterized by:

Wave-like deformation: Localized circumferential buckling waves with wavelength approximately 3-5 times the pipe wall thickness.
Critical displacement: Smallest at a fault dip angle of 75°, making this the most unfavorable fault geometry for local buckling.
Influence of internal pressure: Higher internal pressure provides a stabilizing membrane effect that increases the critical displacement for local buckling.

Failure Mode 3: Excessive Cross-Sectional Deformation

This mode occurs when the ovalization (cross-sectional distortion) of the pipe exceeds the allowable limit, typically defined as 20-30% of the original diameter. The critical displacement depends on:

Fault dip angle: Lower dip angles (30-45°) produce larger cross-sectional deformations because the fault movement has a larger component perpendicular to the pipe axis.
Internal pressure: Higher internal pressure effectively suppresses cross-sectional deformation by providing hoop tension that resists ovalization. As internal pressure increases, the governing failure mode shifts from local buckling to tensile failure.
Diameter-to-thickness ratio: Larger D/t ratios are more susceptible to cross-sectional deformation.

Parametric Study Results

Parameter	Effect on Tensile Failure	Effect on Local Buckling	Effect on Cross-Sectional Deformation
Fault dip angle (30° → 75°)	Critical displacement decreases	Critical displacement decreases (minimum at 75°)	Critical displacement increases
Internal pressure (0 → 10 MPa)	Critical displacement increases	Critical displacement increases	Critical displacement increases significantly
D/t ratio (200 → 500)	Critical displacement decreases	Critical displacement decreases	Critical displacement decreases significantly

Engineering Practice and Design Implications

The findings of this study have direct implications for the seismic design of buried pipelines crossing active faults. The following table summarizes key design recommendations:

Design Parameter	Recommended Approach	Reference Standard
Fault dip angle consideration	Design for 75° dip angle (most unfavorable for buckling)	ASCE 41, API RP 2201
Allowable strain capacity	Use 2-3% for carbon steel pipes; 4-6% for high-strain-capacity steels	API 5L, ASME B31.4
Internal pressure effect	Credit internal pressure for improved ductility, but limit credit to 50% of total capacity	ASCE 41
D/t ratio limit	Limit to 200-300 for seismic zones with active fault crossing	API RP 2201
Joint type	Use girth welds with full-penetration butt welds; avoid flanged joints at fault crossing	ASME B31.4

From a welding perspective, the girth welds at the fault crossing must be designed to accommodate the expected axial strain without cracking. This requires:

Weld metal selection: Use weld metals with elongation ≥ 25% and reduction of area ≥ 50% to match or exceed the pipe material ductility.
Weld procedure qualification: Qualify welding procedures for strain ranges up to 3-5% at the weld HAZ.
Post-weld heat treatment: Consider PWHT to reduce residual stresses and improve HAZ toughness, particularly for high-strength steels (API 5L X65 and above).
Non-destructive testing: Apply 100% ultrasonic testing (UT) or phased array UT (PAUT) to detect volumetric defects that could initiate fracture under cyclic loading.

The study also highlights the importance of considering the pipe's internal pressure condition during seismic events. In the event of a major earthquake, the pipeline may experience a transient pressure surge or pressure loss due to valve operation or pump shutdown. The design should consider the worst-case pressure condition for each failure mode: zero pressure for tensile failure (maximum axial strain) and full operating pressure for cross-sectional deformation (minimum resistance to ovalization).

Study Insights and Reflections

This paper provides valuable quantitative data on the failure mechanisms of buried steel pipes under reverse fault displacement. The identification of 75° as the most unfavorable fault dip angle for local buckling is a critical finding for seismic pipeline design. In many seismic hazard assessments, the fault dip angle is either assumed to be vertical (90°) or not explicitly considered, potentially underestimating the buckling risk.

The finding that higher internal pressure shifts the governing failure mode from local buckling to tensile failure has important implications for pipeline design philosophy. Traditionally, pipeline design focuses on preventing tensile rupture as the primary failure mode. However, for pipes crossing reverse faults, local buckling may be the governing mode under zero or low internal pressure conditions. This suggests that design codes should require checking both failure modes under different pressure scenarios.

The three-dimensional pipe-soil interaction modeling approach used in this study represents the state-of-the-art in numerical analysis of buried pipelines. However, the model's accuracy depends on the quality of the soil constitutive model and the contact parameters. Future research should incorporate more advanced soil models (e.g., Cam-Clay or modified Cam-Clay) and validate the numerical predictions against centrifuge testing or full-scale field tests.