Intelligent Prediction Method for Pipeline Structural Health State Under Fault Movement

The rapid development of the oil and gas industry has led to increasingly severe challenges for buried pipelines when crossing complex geological environments. Especially in fault zones induced by seismic action, the pipe–soil interaction mechanism and the rapid judgment of pipeline mechanical response urgently require in-depth research. This study conducted pipe–soil interaction tests on pipeline uplift under seismic-frequency loading, and for the first time, proposed a modified soil-spring method suitable for typical soft clay under seismic wave frequencies of 1–5 Hz. Through numerical simulation, the axial strain response of pipelines under normal fault movement was systematically analyzed. Considering comprehensively various variables such as fault dip angle, seismic wave frequency, internal pipeline pressure and wall thickness variation, this study extracted the maximum and minimum strain characteristics of the pipe top and pipe bottom, established a diversified intelligent prediction system for fault geological hazards, constructed the optimal machine learning model matching the type of normal fault geological hazards, and realized full-process intelligent modeling from model selection to parameter optimization. The research results can provide technical support for the seismic design and safety status prediction of pipelines under normal faulting conditions.