Seismic fragility and resilience assessment of shallowly buried large-section underground civil defense structure in soft soils: Framework and application

Abstract

The assessment of fragility of underground civil defense structures is of paramount importance in ensuring their functionality and enhancing safety in seismic-prone regions. This study aims to explore the evolution law of seismic fragility for a shallowly buried large-section underground civil defense structure in soft soils, employing advanced data-based approaches. To achieve this, a two-dimensional (2D) nonlinear finite element model of the soil-civil defense structure interaction system was established, integrating data-based parameters and insights. Subsequently, the cloud and incremental dynamical analysis (IDA) methods were employed to develop a probabilistic seismic demand model for the shallowly buried large-section civil defense structure, effectively utilizing data-based approaches to account for the nonlinear characteristics of the soil and the uncertainties of input seismic motions. Based on this model, seismic fragility curves were generated using different seismic intensity measures (IMs), and the evaluation of IMs was carried out by the efficiency and practicality analyses of IMs. These curves were then systematically compared with other existing empirical and analytical fragility curves. The developed fragility curves were subsequently applied to assess seismic risk and resilience for a typical underground civil defense structure in soft soils, offering valuable data-driven insights into the structure's performance under various scenarios. The results indicated that the seismic fragility of the civil defense structure increases with the softening of the soil. Moreover, this study highlighted the substantial influence of structural typologies and local soil conditions on the fragility of civil defense structures, particularly the seismic fragility of underground structures characterized by shallowly buried and large-section, and demonstrating the utility of data-based approaches in comprehending complex interactions. The efficiency and practicality analyses highlighted that ground surface IMs outperform bedrock IMs and acceleration-type IMs over velocity-type IMs. The proposed seismic fragility curves indicated that the exceedance probability for each damage state gradually approaches 100% with the increase of seismic intensity. Finally, a case study was presented on the use of seismic fragility curves to assessment of seismic risk and seismic resilience of civil defense structure. The outcomes of this research, enriched by data-based approaches, can serve as invaluable references for seismic performance assessment, risk analysis, and resilience evaluation of typical civil defense structures in similar soft soil sites.