Zhuojin Pipe Fitting Co., LtdSeismic Vulnerability Analysis of Buried Steel Pipes in Liquefiable Sites Using Advanced Numerical Modeling
Literature Overview
The research by Liang Zetian, Zheng Shansuo, Liu Xiaohang, Wu Xingxia, and Yang Feng from Xi'an University of Architecture and Technology addresses a complex geotechnical-structural interaction problem: the seismic vulnerability of buried steel pipes in liquefiable soil conditions. Supported by multiple funding sources including the National Key R&D Program (2019YFC1509302), the National Natural Science Foundation (52278530), and Shaanxi Provincial Key R&D Program (2021ZDLSF06-10), this study develops a sophisticated numerical framework that integrates soil liquefaction mechanics with pipe-soil interaction modeling.
Technical Framework and Methodology
Soil Constitutive Modeling
The study employs the Matasović nonlinear constitutive model to describe the nonlinear mechanical behavior of liquefiable sandy soils. This model is particularly well-suited for liquefaction analysis because it captures the key features of cyclic soil behavior, including stiffness degradation, plastic strain accumulation, and pore pressure generation under repeated loading. The Matasović model parameters are calibrated against laboratory cyclic triaxial test data, ensuring that the numerical model accurately represents the actual soil behavior.
In addition to the stress-strain relationship, the study incorporates a pore pressure increment model based on the Martin method, modified by Byrne. This modification is significant because the original Martin model was developed for specific soil types and loading conditions, and the Byrne correction extends its applicability to a broader range of liquefiable soils encountered in practice. The pore pressure model is essential because pore pressure accumulation is the primary mechanism by which liquefaction develops—the progressive buildup of excess pore water pressure reduces the effective stress in the soil, leading to loss of shear strength and potentially catastrophic structural failure.
UMAT Implementation
A user-defined material subroutine (UMAT) was developed for the ABAQUS finite element analysis platform to implement the combined Matasović constitutive model and Byrne-corrected Martin pore pressure model. This custom UMAT allows for the simultaneous computation of soil stress-strain response and pore pressure evolution within a unified numerical framework. The reliability of this implementation was validated by reproducing results from existing shaking table tests, which provides confidence in the model's predictive capability for practical engineering applications.
Pipe-Soil Interaction Model
The study develops a hybrid numerical model that combines the entity nonlinear contact model with a shell-equivalent soil spring model. This approach is a practical compromise between computational efficiency and modeling accuracy:
| Model Component | Description | Purpose |
|---|---|---|
| Entity nonlinear contact | 3D contact elements between pipe and soil | Captures localized pipe-soil interaction |
| Shell element | Pipe wall discretized as shell elements | Efficiently models pipe structural response |
| Equivalent soil spring | Spring elements representing soil resistance | Models distributed soil-pipe interaction |
This hybrid approach allows for efficient incremental dynamic time history analysis of buried steel pipes under various liquefaction scenarios, which would be computationally prohibitive using a fully coupled 3D soil-structure interaction model.
Seismic Vulnerability Analysis
Demand Models
The study establishes demand models for buried steel pipes under different liquefaction zone lengths and embedment depths. Incremental dynamic analysis (IDA) was performed by subjecting the numerical models to progressively increasing ground motion intensities, recording the structural response at each intensity level. The resulting demand models express the relationship between seismic demand parameters (such as maximum pipe strain, displacement, or curvature) and seismic intensity measures (such as peak ground acceleration or spectral acceleration).
Vulnerability Models
By combining the demand models with limit state definitions for different damage levels, the study constructs fragility curves that express the probability of exceeding each damage state as a function of seismic intensity. Three different liquefaction zone lengths and three different embedment depths were analyzed, providing a parametric basis for understanding the influence of these key variables on seismic vulnerability.
Key Findings
| Parameter Variation | Effect on Vulnerability |
|---|---|
| Increasing liquefaction zone length (same PGA, same depth) | Higher exceedance probability at all damage states |
| Increasing embedment depth (same PGA, same liquefaction length) | Higher exceedance probability at all damage states |
The finding that greater embedment depth leads to higher vulnerability is counterintuitive at first glance but can be explained by the increased soil displacement demands imposed on deeper pipes during liquefaction-induced lateral spreading. Deeper pipes experience larger lateral soil displacements due to the geometry of the liquefaction-induced deformation field, resulting in greater pipe strain and displacement demands.
Engineering Practice Implications
From the perspective of steel pipe manufacturing and installation, several important implications emerge from this research:
Material Selection and Grade Requirements
The study's vulnerability analysis implicitly highlights the importance of material selection for buried steel pipes in liquefiable sites. Pipes subjected to significant seismic demands in liquefiable soils require materials with adequate ductility to accommodate large deformations without fracture. From a metallurgical standpoint, this means that pipes fabricated from high-strength low-alloy (HSLA) steels or controlled-rolled steels with good Charpy V-notch (CVN) toughness properties are preferable to conventional carbon-manganese steels. The welding quality of pipe girth welds and longitudinal seams becomes particularly critical, as weld defects can initiate fracture under the large cyclic strains experienced during seismic events.
Pipe Wall Thickness and Corrosion Protection
The study does not explicitly address corrosion effects, but in practice, corrosion-induced wall thinning can significantly reduce the seismic capacity of buried steel pipes. For pipes installed in aggressive soil environments (high pH, high sulfate content, or poor cathodic protection), the design should incorporate corrosion allowance in the wall thickness calculation. The pipe specifications should comply with standards such as API 5L for line pipe or ASTM A53 for general structural pipe, with appropriate grade and wall thickness selected based on the combined requirements of design pressure, seismic demand, and corrosion allowance.
Installation and Backfill Considerations
The embedment depth finding has direct implications for pipe installation practice. In liquefiable sites, deeper burial does not necessarily provide better seismic protection and may actually increase vulnerability. Engineers should consider shallower burial depths where possible, combined with appropriate soil improvement measures (such as densification, grouting, or installation of stone columns) to reduce liquefaction potential. The backfill soil properties also play a critical role in pipe-soil interaction, and the study's numerical model should be calibrated with site-specific backfill properties for accurate vulnerability assessment.
Key Technical Challenges and Limitations
The study's approach, while sophisticated, has several limitations that should be acknowledged. First, the UMAT implementation assumes a specific soil type and loading history, and the model parameters may not be directly transferable to other soil conditions without recalibration. Second, the hybrid pipe-soil interaction model, while computationally efficient, may not accurately capture all aspects of the three-dimensional pipe-soil interaction, particularly in cases of large relative displacement between the pipe and the surrounding soil. Third, the study does not address the effect of pipe joints (welded or flanged connections) on seismic vulnerability, which is a critical practical consideration since pipe joints are often the weakest links in buried pipeline systems.
Study Insights and Implications
This research represents a significant advancement in the numerical modeling of buried steel pipe seismic behavior in liquefiable soils. The development of a validated UMAT subroutine combining advanced soil constitutive models with pore pressure generation is a valuable contribution to the geotechnical engineering community. The resulting fragility curves provide a quantitative basis for seismic risk assessment and performance-based design of buried pipeline systems. Future research should extend this framework to include the effects of pipe material properties, welding quality, corrosion degradation, and joint design on seismic vulnerability, creating a more comprehensive tool for pipeline seismic safety assessment.