@article{LI2026, 
author = {Chunbao LI and Tianqi ZHENG and Feng GUO and Pengju QIN and Deqiang ZHAO and Jing LI and Zonglu LIU and Weiqiang CHU and Wenbei WU},
title = {Simulation of welding repair for low-alloy high-strength steel storage tanks and microstructure–property control in the heat-affected zone},
year = {2026},
journal = {Experimental Technology and Management},
volume = {43},
number = {7},
pages = {157-165},
keywords = {GTAW welding, FEM, residual stress, t8/5 prediction formula},
url = {https://www.sciopen.com/article/10.16791/j.cnki.sjg.2026.07.018},
doi = {10.16791/j.cnki.sjg.2026.07.018},
abstract = {ObjectiveIn-service oil and gas storage tanks often require local shell-plate replacement after prolonged operation. Repair welding imposes complex, non-uniform thermal cycles that can generate high residual stresses and degrade the microstructure and toughness of the coarse-grained heat-affected zone (CGHAZ). This study aims to (ⅰ) reveal the thermal–mechanical evolution during multi-layer, multi-pass repair welding of a low-alloy high-strength steel tank wall, (ⅱ) quantify the spatial distribution of residual stresses in the circumferential and axial directions, and (ⅲ) establish a process–cooling–microstructure relationship based on the cooling time t8/5, so that repair parameters can be optimized for safer service performance.MethodsA three-dimensional finite element (FE) model representing a sector of a thin-walled storage tank was constructed for a representative repair scenario involving butt and T-joint configurations. The weld build-up was reproduced as a three-layer, single-pass sequence using element activation to simulate progressive filler deposition. A double-ellipsoidal moving volumetric heat source was adopted to represent gas tungsten arc welding (GTAW) heat input, and temperature-dependent convection and radiation were applied as boundary conditions. A sequentially coupled thermo-mechanical strategy was implemented: transient temperature fields were solved first and then mapped into an elastic–plastic mechanical analysis to predict residual stress after cooling. To evaluate the credibility of the model for field application, predicted stress profiles were compared with on-site measurements obtained by a coercive force–based method. Thermal cycles were extracted at CGHAZ-relevant locations to calculate t8/5, and parametric simulations were performed by varying current and travel speed. Continuous cooling transformation (CCT) behavior was assessed with JMatPro to interpret the tendency toward different transformation products under various t8/5 values. In addition, commonly used empirical formulas for t8/5 were compared with FE predictions to identify a suitable approach for thin-wall tank GTAW repair.ResultsThe thermal model produced a stable molten-pool response under the selected parameters, with peak temperatures on the order of 2100 ℃ and continuous fusion along the modeled passes. The residual stress field was strongly localized: tensile stresses concentrated within the weld metal and adjacent HAZ, while balancing compressive stresses formed in the surrounding plate. Peak tensile residual stresses occurred near the HAZ, reaching approximately 428 MPa in the circumferential direction and approximately 408 MPa in the axial direction, indicating that near-weld regions dominate the structural risk during service. Along the weld path, stresses were comparatively smooth in the stable central segment, whereas across the weld, the gradients were steep, consistent with sharp thermal gradients and constrained shrinkage. Agreement between simulation and field measurements showed consistent trends and comparable magnitudes, with deviations within roughly 20%. The FE-computed t8/5 increased monotonically with heat input: for the studied window (approximately 0.80–1.28 kJ/mm), t8/5 rose from about 5 s to nearly 13 s, demonstrating that practical parameter adjustment can substantially alter the cooling severity. Among the examined empirical approaches, a thin-plate formulation provided the closest match to FE trends, whereas other formulas showed larger deviations for the present geometry. CCT-based interpretation suggested that shorter t8/5 (faster cooling) increases the tendency toward bainitic/martensitic products and martensite–austenite (M–A) constituent formation, which may reduce impact toughness and raise cracking susceptibility in the CGHAZ, whereas excessively prolonged t8/5 may increase the risk of softening depending on service requirements.ConclusionsA validated FE framework for multi-pass GTAW repair welding of low-alloy high-strength steel storage tank walls was developed to jointly evaluate temperature history, residual stress, and t8/5-controlled microstructure tendencies. The study confirms that tensile residual stress risk is concentrated in the weld and HAZ and that t8/5 can be systematically regulated through current and travel speed to support CGHAZ microstructure control. The combined FE–CCT workflow provides quantitative guidance for selecting balanced repair parameters that mitigate residual stress concentration while maintaining acceptable microstructure and mechanical performance.}
}