TY - JOUR
T1 - Full-scale experimental and numerical analysis of residential building and its masonry wall response under LPG explosion loading
AU - Zhang, Yue
AU - Qian, Xinming
AU - Li, Hongyu
AU - Li, Pengliang
AU - Feng, Jingchen
AU - Li, Fangzhou
AU - Fan, Wulong
N1 - Publisher Copyright:
© 2026
PY - 2026/7/1
Y1 - 2026/7/1
N2 - This study presents a combined experimental and numerical investigation: full-scale tests on a residential structure and component-level simulation of its masonry wall under liquefied petroleum gas (LPG) explosion loading. Controlled full-scale tests were conducted to comparatively examine the effects of architectural layout (rectangular vs. “gun-shaped”) and ignition method (black powder vs. dual electric ignition) on the blast environment and structural damage. Key parameters, including transient overpressure, dynamic temperature, wall collapse and glass fragmentation, were recorded. Under the two specific test configurations, the “gun-shaped” layout altered gas dispersion and shockwave propagation, an effect that may be a contributing factor to the reversal of the collapse direction of a critical masonry wall. Furthermore, compared with Test 2, Test 1 exhibited faster deflagration and a more severe blast, characterized by an earlier pressure rise (by 0.98 s), a steeper shockwave front, faster flame propagation (72.5 m/s vs. 6.5 m/s), and higher glass fragment velocities (53 m/s vs. 8.7 m/s). An important contributing factor to this difference is the ignition method (black powder vs. electric ignition). A numerical model was developed in LS-DYNA to simulate the component-level failure process of the masonry wall. Quantitative comparison shows that the model reproduces the X-shaped crack pattern and overall collapse, with a maximum displacement error within 10%. Wall fragment velocities range from 7.0 to 9.5 m/s. Furthermore, parametric analysis using a simplified triangular pressure pulse (20-100 kPa) reveals that average fragment velocity increases approximately linearly with peak overpressure (2.2-33.8 m/s), while fragment count rises nonlinearly (30-62), providing preliminary quantitative reference for fragment hazard assessment. These findings provide a reference database for typical residential LPG explosions and an initial technical reference for forensic analysis.
AB - This study presents a combined experimental and numerical investigation: full-scale tests on a residential structure and component-level simulation of its masonry wall under liquefied petroleum gas (LPG) explosion loading. Controlled full-scale tests were conducted to comparatively examine the effects of architectural layout (rectangular vs. “gun-shaped”) and ignition method (black powder vs. dual electric ignition) on the blast environment and structural damage. Key parameters, including transient overpressure, dynamic temperature, wall collapse and glass fragmentation, were recorded. Under the two specific test configurations, the “gun-shaped” layout altered gas dispersion and shockwave propagation, an effect that may be a contributing factor to the reversal of the collapse direction of a critical masonry wall. Furthermore, compared with Test 2, Test 1 exhibited faster deflagration and a more severe blast, characterized by an earlier pressure rise (by 0.98 s), a steeper shockwave front, faster flame propagation (72.5 m/s vs. 6.5 m/s), and higher glass fragment velocities (53 m/s vs. 8.7 m/s). An important contributing factor to this difference is the ignition method (black powder vs. electric ignition). A numerical model was developed in LS-DYNA to simulate the component-level failure process of the masonry wall. Quantitative comparison shows that the model reproduces the X-shaped crack pattern and overall collapse, with a maximum displacement error within 10%. Wall fragment velocities range from 7.0 to 9.5 m/s. Furthermore, parametric analysis using a simplified triangular pressure pulse (20-100 kPa) reveals that average fragment velocity increases approximately linearly with peak overpressure (2.2-33.8 m/s), while fragment count rises nonlinearly (30-62), providing preliminary quantitative reference for fragment hazard assessment. These findings provide a reference database for typical residential LPG explosions and an initial technical reference for forensic analysis.
KW - Blast loading
KW - Full-scale experiment
KW - LPG explosion
KW - Numerical simulation
KW - Structural response
UR - https://www.scopus.com/pages/publications/105045454725
U2 - 10.1016/j.jobe.2026.116972
DO - 10.1016/j.jobe.2026.116972
M3 - Article
AN - SCOPUS:105045454725
SN - 2352-7102
VL - 129
JO - Journal of Building Engineering
JF - Journal of Building Engineering
M1 - 116972
ER -