Advanced Finite Element Model Research Articles

Unraveling mechanisms and failures of stray current corrosion in urban Subway adjacent pipeline networks through statistical insights and 3D simulation

The escalating expansion of urban subway systems in recent years has accentuated the issue of stray current corrosion within pipeline networks, emerging as a critical concern for urban safety. This paper delves into the intricate interplay between these phenomena, employing data-driven statistical analyses to elucidate the coupling characteristics between subway lines and the occurrence of failures in adjacent buried pipelines. An advanced three-dimensional finite element model was developed for stray current corrosion in pipelines, seamlessly integrating empirical data and physics-based modeling, which is to uncover the spatial nuances and multifaceted impacts on subway pipeline corrosion from both macro and micro perspectives under varying influencing factors. The study unveils a pronounced geographical and functional affinity between urban subway networks and metallic pipeline networks. The coupling attributes between subway systems and pipelines, such as distance, angle, and pipeline-specific characteristics including material and age, assume pivotal roles. The results further emphasize the hierarchical order of influence, with stray current intensity holding the greatest sway, followed by the distance between subway and pipelines, the angle between them, and soil resistivity. This paper offers a comprehensive investigation of the interrelationships and influential factors between subway systems and adjacent pipelines. It contributes to the mitigation and management of stray current corrosion in pipelines induced by nearby rail transit, thereby enhancing the resilience of both subway and pipeline networks within urban areas.

Flexural performance of reinforced concrete beams strengthened using near-surface-mounted carbon-fiber-reinforced polymer bars: Effects of bonding patterns

The near-surface-mounted (NSM) fiber-reinforced polymer (FRP) technique is effective for enhancing the load carrying capacity of strengthened reinforced concrete (RC) beams. However, this technique significantly reduces the deformability of strengthened RC beams. To overcome this, the present study combines the NSM strengthening method with the partial bonding technique: the strengthening FRP reinforcement is partially bonded onto the concrete substrate). Six full-scale 4.3 m long RC beams strengthened with NSM FRPs using various bonding patterns were tested. The results indicated that mechanical interlocking grooves slightly improve the load carrying capacity, but have a negligible effect on the deformability of the strengthened beams. A decrease in the bond length increased deformability by 27.2 %, with only a 7.6 % drop in load capacity compared to that of the beam with full bonding. An advanced finite element model was developed and validated to confirm an optimal bond length of 930 mm for maximum flexural performance.

Fire Behavior of Concrete-Filled Hollow Section Columns with High Strength Bar-Bundle as Core

AbstractConcrete-filled hollow section (CFHS) columns with a solid steel core have gained popularity in the construction of tall buildings due to their robust load-bearing capacity, slender design, ease of prefabrication, and exceptional structural fire resistance. In this research paper, we introduce an innovative approach aimed at enhancing the structural performance of these columns. Our method involves replacing the solid steel core with high-strength bar bundles and substituting concrete with grout to achieve superior fire resistance. These modified columns are referred to as “bar-bundle columns.” The paper presents the results of extensive fire tests conducted on three bar-bundle columns, each with different bar-bundle sizes, quantities, and configurations. Additionally, we determine the temperature-dependent material properties of the high-strength steel used for reinforcing bars and the thermal properties of the grout used as a filler through standard experimental tests, which are crucial for numerical simulations. An advanced nonlinear finite element model is describe which is capable of predicting the fire behavior of bar-bundle columns. Finally, this numerical model is employed to conduct parametric analyses and propose a simplified design model for bar-bundle columns under fire conditions.Our findings indicate that the bar-bundle configuration and using grout as a filler significantly delays the heating of the steel core, resulting in enhanced fire resistance when compared to CFHS columns with a solid steel core. The simplified method proposed in this study can be used to estimate the fire resistance of slender bar bundles, but further experimental testing could further refine and improve its accuracy.

Modelling rapid non-destructive test using light weight deflectometer on granular soils across different degrees of saturation

This study introduces an advanced finite element model for the light weight deflectometer (LWD), which integrates contact mechanics with fully coupled models. By simulating LWD tests on granular soils at various saturation levels, the model accurately reflects the dependence of the LWD modulus on dry density, water content, and effective stress. This model addresses and overcomes the limitations of previous finite element models for this specific problem. Simultaneously, this research presents the first experimentally validated fully coupled contact impact model. Furthermore, the research provides a comparative assessment of elastoplastic and nonlinear elastic models and contrasts an enriched node-to-segment method (developed in this study) with the more precise mortar technique for contact mechanics. These comparisons reveal unique advantages and challenges for each method. Moreover, the study underscores the importance of careful application of the LWD modulus, emphasising the need for sophisticated tools to interpret soil behaviour accurately.

Hybrid CFRP-BFRP Rebars for Prestress Concrete Slabs: Abaqus Analysis and Experimental Validation

This study utilizes Abaqus finite element software to analyze concrete slabs reinforced with a novel hybrid of Carbon Fiber Reinforced Polymers (CFRP) and Basalt Fiber Reinforced Polymers (BFRP) rebars known as CBRP rebars. The investigation focuses on assessing the structural performance of these slabs under varying prestressing levels (PL) and FRP reduction factors (FR), along with the prestress index level (PI). The study defines the FR factor as the reduction in the transverse reinforcement area of FRP compared to the control slab. The PI is the ratio of prestressed CBRP rebars to the total FRP reinforcement in the transverse direction, aiming to determine a lower FRP reduction factor without compromising the FRP slabs' design requirements. Three FR values (0.45, 0.37, and 0.29) are considered, with the CBRP tendon tensioned at 35% and 50% of the ultimate strength, including 0.5 and 0.6 of the PI, for validation against experimental test data. The numerical analysis incorporates advanced finite element modeling techniques, integrating material properties, load conditions, and structural parameters to detail the behavior of CBR-reinforced concrete slabs. Various loading scenarios are simulated to evaluate the performance of CBRP rebars under diverse conditions. Comparison with experimental data validates the hybrid reinforcement approach's reliability. The study's findings contribute significantly to understanding the structural performance of concrete slabs reinforced with CBRP rebars. The hybridization of CFRP and BFRP rebars offers a promising alternative for enhancing concrete structure strength while reducing the carbon footprint associated with traditional reinforcement materials. The versatility of CBRP rebars allows customization based on specific structural requirements. Based on the Abaqus numerical analysis and its alignment with experimental data, the study recommends optimizing the CFRP-BFRP ratio to meet structural demands, evaluating long-term performance, and exploring environmental benefits associated with sustainable reinforcement materials. This research advances knowledge in the hybrid fiber-reinforced polymers field, offering insights for engineers and researchers seeking innovative solutions for concrete structure enhancement using Abaqus as a powerful analytical tool.

Prediction of the Response of Masonry Walls under Blast Loading Using Artificial Neural Networks

A methodology to predict key aspects of the structural response of masonry walls under blast loading using artificial neural networks (ANN) is presented in this paper. The failure patterns of masonry walls due to in and out-of-plane loading are complex due to the potential opening and sliding of the mortar joint interfaces between the masonry stones. To capture this response, advanced computational models can be developed requiring a significant amount of resources and computational effort. The article uses an advanced non-linear finite element model to capture the failure response of masonry walls under blast loads, introducing unilateral contact-friction laws between stones and damage mechanics laws for the stones. Parametric finite simulations are automatically conducted using commercial finite element software linked with MATLAB R2019a and Python. A dataset is then created and used to train an artificial neural network. The trained neural network is able to predict the out-of-plane response of the masonry wall for random properties of the blast load (standoff distance and weight). The results indicate that the accuracy of the proposed framework is satisfactory. A comparison of the computational time needed for a single finite element simulation and for a prediction of the out-of-plane response of the wall by the trained neural network highlights the benefits of the proposed machine learning approach in terms of computational time and resources. Therefore, the proposed approach can be used to substitute time consuming explicit dynamic finite element simulations and used as a reliable tool in the fast prediction of the masonry response under blast actions.

Artificial neural network surrogate modeling for uncertainty quantification and structural optimization of reinforced concrete structures

AbstractOptimization approaches are important to design sustainable structures. In structural mechanics, different design objectives can be defined, for example, to minimize the required construction material or to maximize the structural durability. In this paper, the durability of a reinforced concrete (RC) structure is assessed by advanced finite element (FE) models to simulate the cracking behavior and the chloride transport process. The corrosion initiation time is used as durability measure to be maximized within an optimization approach, where the concrete cover is defined as design variable. The variability of structural loads and material parameters and unavoidable construction imprecision leads to a probabilistic reliability and durability assessment, where aleatory as well as epistemic uncertainties are quantified by random variables, intervals and probability‐boxes. The FE simulation models cannot directly be applied to structural analyses and optimizations with polymorphic uncertain parameters and design variables because of the high computational demand of the multi‐loop algorithm (Monte Carlo simulation, interval analysis, global optimization). In this paper, a new surrogate modeling strategy is presented, where artificial neural networks are trained sequentially to speed‐up the coupled mechanical and transport simulation FE models. The new approach is applied to the uncertainty quantification and the structural durability optimization of a RC structure.

Advanced finite element modelling of stainless steel bolted T-stubs under large deformations

Following upon the recently published experimental and numerical studies on stainless steel bolted T-stubs in tension, this study reports the development and validation of an advanced FE model that reveals the effect of initial contact between the bolts and bolt holes and constructional imperfections on the response of stainless steel bolted T-stubs subjected to large deformations. Key simulation strategies regarding the modelling of bolts, initial position of the bolts, flange bowing effect due to the initial constructional imperfections and overcoming numerical instabilities are discussed. The developed FE model first validated against the available test data in the literature. The effect of pre-established contact between the bolts and bolt holes and constructional imperfections on the overall response of the bolted T-stubs are reported comprehensively by comparing the plastic and ultimate resistances with the experimental ones. Moreover, the membrane action developing in the flange influences the ductility reserve of the T-stub which has a vital role in case of progressive collapse and robustness scenarios. Therefore, by inspecting the stress distribution through the flange thickness at various locations along the T-stubs, effects of different modelling configurations on the flexural and the membrane component of the load transfer mechanism are revealed. A parametric study is performed to investigate the influence of all these aspects on resistance of T-stub connections. On the basis of the obtained results and the discussions, modelling recommendations for the simulation of stainless steel T-stubs are made.

Embedding strain-rate sensitivities of multiple deformation mechanisms to predict the behavior of a precipitate-hardened WE43 alloy under a wide range of strain rates

A rare earth Mg alloy, WE43, exhibits high strength, good ductility, low anisotropy, and moderately high strain rate sensitivity. As such, the alloy is a viable candidate for high strain rate applications. In this work, a comprehensive set of mechanical and microstructure data recorded during quasi-static, high strain rate split Hopkinson bar (SHB), and impact tests on specimens of WE43 Mg alloy reported in (Savage et al., 2020b) is simulated and interpreted using an advanced Taylor-type crystal plasticity finite element (T-CPFE) model. The T-CPFE model is formulated physically to embed two sources of strain-rate sensitivities inherent to each slip and twinning mode in WE43, one that occurs under constant structure and another that affects structure evolution. The model parameters are established for the alloy by achieving agreement in the stress-strain response and microstructure evolution under quasi-static and SHB tests. Density functional theory calculations of anti-phase boundary (APB) energy are carried out to explain origins of the unusually large initial slip resistance for basal dislocations, which shear precipitates in the alloy. The initial slip resistances of the prismatic and pyramidal dislocations are, instead, rationalized by Orowan looping around precipitates. After calibration and validation, the model is shown to successfully predict WE43 response at much larger strain rates than those used for model calibration. Specifically, mechanical response, specimen geometry changes, twin volume fractions, and texture evolution are predicted for different orientations of the Taylor cylinders. Details of the modeling framework, comparison between simulation and experimental results, and insights from the results are presented and discussed.

Reconsideration of shear strength expressions for CFSTs: Rectangular tubes

Concrete filled steel tubes (CFSTs) are typically circular or rectangular. Circular CFSTs are mostly used as vertical components of deep foundation systems; rectangular sections are more commonly used in building construction such as square CFSTs for columns. Prior work has focused on circular CFSTs with far fewer studies on square or rectangular sections. In addition to being more limited, prior research on square CFST sections has focused on flexural and axial response; there have been far fewerstudies on the shear response. As such, the accuracy of shear-strength equations is uncertain. A research study was undertaken to understand and improve design expressions for the shear design of rectangular CFSTs. First, the research team collectedexperimental data on shear resistance of rectangularCFSTs. Selected experimental data was used to develop and validate an advanced nonlinear finite element model (FEM) to predict the measured response and observed damage of rectangular CFSTs controlled by shear behavior. The validated model was then used to conduct parameter studies to investigate important design parameters of rectangular CFSTs responding primarily in shear including: 1) concrete and steel material properties, 2) cross-sectional aspect ratio, and 3) impact of enhanced bond capacity/composite action through the introduction of a single shear key at each end of the member. The results were used to better understand the importance of each parameter on the shear response and strength. A database was assembled, combining the experimental data with the strength predicted by each FEM, to evaluate existing design equations. The shortcomings of the existing equations included (1) only accounting for the steel or the concrete fill contribution, and (2) overestimating the total shear strength because the behavior of the CFST in shear is not well understood. This paper provides a new approach to assessing the shear strength as the individual contributions of the steel tube and the concrete fill are computed for each model to better understand their interaction. This also permits the development of a new design equation to remedy the shortcomings of existing codified models. In particular, the proposed model accurately estimates the individual contributions of steel tube and concrete fill to the shear resistance for salient design parameters.

Regional biomechanical characterization of human ascending aortic aneurysms: Microstructure and biaxial mechanical response.

The ascending thoracic aortic aneurysm (ATAA) is a permanent dilatation of the vessel with a high risk of adverse events, and shows heterogeneous properties. To investigate regional differences in the biomechanical properties of ATAAs, tissue samples were collected from 10 patients with tricuspid aortic valve phenotype and specimens from minor, anterior, major, and posterior regions were subjected to multi-ratio planar biaxial extension tests and second-harmonic generation (SHG) imaging. Using the data, parameters of a microstructure-motivated constitutive model were obtained considering fiber dispersion. SHG imaging showed disruptions in the organization of the layers. Structural and material parameters did not differ significantly between regions. The non-symmetric fiber dispersion model proposed by Holzapfel etal. [25] was used to fit the data. The mean angle of collagen fibers was negatively correlated between minor and anterior regions, and the parameter associated with collagen fiber stiffness was positively correlated between minor and major regions. Furthermore, correlations were found between the stiffness of the ground matrix and the mean fiber angle, and between the parameter associated with the collagen fiber stiffness and the out-of-plane dispersion parameter in the posterior and minor regions, respectively. The experimental data collected in this study contribute to the biomechanical data available in the literature on human ATAAs. Region-specific parameters for the constitutive models are fundamental to improve the current risk stratification strategies, which are mainly based on aortic size. Such investigations can facilitate the development of more advanced finite element models capable of capturing the regional heterogeneity of pathological tissues. STATEMENT OF SIGNIFICANCE: Tissue samples of human ascending thoracic aortic aneurysms (ATAA) were collected. Samples from four regions underwent multi-ratio planar biaxial extension tests and second-harmonic generation imaging. Region-specific parameters of a microstructure-motivated model considering fiber dispersion were obtained. Structural and material parameters did not differ significantly between regions, however, the mean fiber angle was negatively correlated between minor and anterior regions, and the parameter associated with collagen fiber stiffness was positively correlated between minor and major regions. Furthermore, correlations were found between the stiffness of the ground matrix and the mean fiber angle, and between the parameter associated with the collagen fiber stiffness and the out-of-plane dispersion parameter in the posterior and minor regions, respectively. This study provides a unique set of mechanical and structural data, supporting the microstructural influence on the tissue response. It may facilitate the development of better finite element models capable of capturing the regional tissue heterogeneity.

Analysis of the performance of a new concept for automatic draping of wide reinforcement fabrics with pre-shear: A virtual prototyping study

The layup process of large composite structures made from dry reinforcement fabrics is considered. One such structure is a wind turbine blade, for which the current draping process is mostly manual. Automating the draping process will, therefore, lower the costs. Based on a literature review, a new concept is synthesized and analyzed using an advanced finite element model with rigid multi-body kinematics and a dedicated material model for the fabric. The material model is calibrated using experimental coupon tests, i.e. the bias-extension test (shear) and Peirce's cantilever test (out-of-plane bending). The concept is analyzed numerically by means of a simple parameter study and draping test cases on a flat mold as well as a general double-curved mold. The simulation results show that the concept is feasible for the draping operation and is thus qualified for the subsequent physical prototyping.

Cold-formed stainless steel elliptical hollow sections under combined biaxial bending and axial compression - Numerical modeling and design

This paper aims to investigate the cross-section response of cold-formed stainless steel (CFSS) elliptical hollow sections (EHS) under combined biaxial bending and axial compression. In the present study, an advanced finite element (FE) model was adopted to perform a comprehensive parametric study for CFSS EHS under combined biaxial bending and axial compression, with different stainless steel alloys, as well as a broad range of cross-section sizes, aspect ratios, loading eccentricities and angles. Numerical results obtained from the parametric study were compared with the design strengths calculated by the current design rules in European code EN 1993-1-4, American specification SEI/ASCE 8-02 and Australian/New Zealand standard AS/NZS 4673. In addition, the Continuous Strength Method (CSM) for CFSS circular hollow sections (CHS) under combined actions, was also evaluated for the design of CFSS EHS under combined actions. The comparisons revealed that the current design codes EN 1993-1-4, SEI/ASCE 8-02 and AS/NZS 4673 provided conservative and very scattered predictions for CFSS EHS under combined actions. The existing CSM can provide more accurate predictions than the current design codes (i.e., AS/NZS 4673, EN 1993-1-4 and SEI/ASCE 8-02), but a further improvement is needed. In the present study, a modified CSM is proposed for CFSS EHS under combined actions, which can achieve reliable and accurate design predictions.

Evaluation and improvement of shear-strength expressions for circular concrete-filled steel tubes

The accuracy of the shear resistance design equation for concrete-filled steel tubes (CFSTs) is crucial to ensure the safety of CFST structures. However, the current design equations underestimate the test data, and cannot accurately predict the relative shear resistance of the steel and concrete fill. To solve this problem, this paper developed an advanced finite element model in the LS-Dyna program, which was validated using experimental results of CFST members failing in shear. A systematic parametric study was then conducted. The result showed that the contribution of the steel increased significantly with the strain-hardening ratio (F_u/F_y) while the contribution of concrete fill was greatly influenced by the internally reinforced ratio (ρ_int) and axial load level (P/P_0). These predictions were combined with the experimental results to develop a more accurate and reliable design expression for the shear strength of circular CFSTs.

Comparative study on fracture characteristics of carbon and stainless steel bolt material

Bolt fracture often limits the ultimate strength and deformation capacity of bolted connections and leads to overall joint failure. Therefore, the prediction of structural collapse under extreme loads necessitates the incorporation of a reliable fracture model that reflects the observed structural response of bolts. This paper reports a comprehensive experimental and numerical study into the structural behaviour, ductility and fracture characteristics of Grade A4–80 austenitic stainless steel bolts and carbon steel 8.8 bolts in tension, which are commonly used as fasteners in bolted connections. Tensile tests were performed on smooth material coupons machined from both bolt grades to obtain the material response. Thereafter, tensile tests were performed on notched specimens from both grades, allowing the fracture characteristics to be studied over a range of stress triaxialities and the development of an equation relating the plastic strain at fracture to the stress triaxiality. 2D axisymmetric and 3D advanced finite element models were calibrated against the experimental results and damage propagation parameters. The developed numerical models are shown to perfectly replicate the observed experimental behaviour of the tested bolt materials under predominantly tensile loading, including fracture. Both experimental and numerical results confirm the superior ductility of A4–80 bolts over their 8.8 counterparts.

Effects of stress ratio on plasticity-induced crack closure through three-dimensional advanced numerical finite element models

The plasticity induced crack closure (PICC) is a local phenomenon that substantially influences crack growth. PICC is most commonly determined by numerical models. Specifically, the vertical displacements of the nodes behind the crack front have been extensively used to obtain the PICC and the crack opening loads (Pop/Pmax). Although this procedure has been demonstrated to be adequate for positive stress ratios, this work evidences the uncertainty of such procedure to obtain PICC under negative stress ratios. For this reason, the main objective of this paper is to investigate the effect of stress ratio on PICC by 3D advanced numerical methodology. The purposed method consists of evaluating the contact pressures of all nodes behind the crack tip, from the first to the last released row of elements included on a crack propagation finite element model. Moreover, this method is tested at positive stress ratios and results show a good agreement with the CTOD method.

Finite Element Simulation and Sensitivity Analysis of the Cohesive Parameters for Delamination Modeling in Power Electronics Packages.

Delamination is a critical failure mode in power electronics packages that can significantly impact their reliability and performance, due to the large amounts of electrical power managed by the most recent devices which induce remarkable thermomechanical loads. The finite element (FE) simulation of this phenomenon is very challenging for the identification of the appropriate modeling tools and their subsequent calibration. In this study, we present an advanced FE modeling approach for delamination, together with fundamental guidelines to calibrate it. Considering a reference power electronics package subjected to thermomechanical loads, FE simulations with a global-local approach are proposed, also including the implementation of a bi-linear cohesive zone model (CZM) to simulate the complex interfacial behavior between the different layers of the package. A parametric study and sensitivity analysis is presented, exploring the effects of individual CZM variables on the delamination behavior, identifying the most crucial ones and accurately describing their underlying functioning. Then, this work gives valuable insights and guidelines related to advanced and aware FE simulations of delamination in power electronics packages, useful for the design and optimization of these devices to mitigate their vulnerability to thermomechanical loads.

A corrugated steel fender for bridge pier protection against truck collision

This study proposes a novel corrugated steel fender system to protect bridge piers against truck collisions. In particular, the impact energy is mainly absorbed through the deformation of corrugated steel webs in the fender. Pendulum impact tests are conducted to demonstrate the benefits of installing the corrugated steel web to increase the fender stiffness, engage more lateral displacement, and dissipate considerable impact energy. Advanced finite element (FE) models are established using the software LS-DYNA and verified against experimental results of the pendulum impact test. Modeling considerations are implemented into developing a high-resolution FE model to simulate full-scale truck–fender–bridge collisions, which discloses the fender’s effectiveness in protecting the bridge pier. The proposed fender system can mitigate the post-impact damage to the bridge pier by reducing its peak impact force, lateral displacement, and bending moment at the base. Moreover, parametric analyses are performed to investigate the effects of different fender design variables on the peak impact force and absorbed impact energy. Among various design parameters, the thicknesses of the corrugated steel web and surface steel plate are the predominant ones. Finally, an optimal fender design criteria is proposed to simultaneously minimize the impact damage to the bridge pier and the truck.

Fracture Toughness and Fatigue Crack Growth Analyses on a Biomedical Ti-27Nb Alloy under Constant Amplitude Loading Using Extended Finite Element Modelling.

The human body normally uses alternative materials such as implants to replace injured or damaged bone. Fatigue fracture is a common and serious type of damage in implant materials. Therefore, a deep understanding and estimation or prediction of such loading modes, which are influenced by many factors, is of great importance and attractiveness. In this study, the fracture toughness of Ti-27Nb, a well-known implant titanium alloy biomaterial, was simulated using an advanced finite element subroutine. Furthermore, a robust direct cyclic finite element fatigue model based on a fatigue failure criterion derived from Paris' law is used in conjunction with an advanced finite element model to estimate the initiation of fatigue crack growth in such materials under ambient conditions. The R-curve was fully predicted, yielding a minimum percent error of less than 2% for fracture toughness and less than 5% for fracture separation energy. This provides a valuable technique and data for fracture and fatigue performance of such bio-implant materials. Fatigue crack growth was predicted with a minimum percent difference of less than nine for compact tensile test standard specimens. The shape and mode of material behaviour have a significant effect on the Paris law constant. The fracture modes showed that the crack path is in two directions. The finite element direct cycle fatigue method was recommended to determine the fatigue crack growth of biomaterials.

Comparative study on seismic performance of prefabricated and cast-in-place underground structures under different site conditions

Since the seismic performance of prefabricated underground structures is not yet clear, they not have been constructed in high seismic intensity areas or poor sites. This paper aims to study the applicability of rectangular prefabricated (semi-rigid prefabricated, ASF, and rigid prefabricated, AMT) underground structures in soft soil sites located in high seismic intensity areas. The structures were compared with cast-in-place (CIP) underground structures. Three-dimensional soil-structure interaction advanced finite element models were established to simulate the seismic behavior of the three types of underground structures (ASF, AMT and CIP) constructed in two different sites (soft rock site and soft soil site). The pre-earthquake stress states, dynamic responses, and post-earthquake performance states of the three underground structures were systematically analyzed, considering various static and dynamic conditions. Additionally, the results of dynamic time history were compared with those of the standard procedures available in the literature. The results demonstrated that: i) site conditions have a significant effect on the structures’ static and dynamic responses, ii) the seismic response in soft soil site is much larger than that in soft rock site, but the ratio of response of prefabricated underground structures to that of CIP is almost the same for both soft soil and soft rock sites. Furthermore, as the peak ground acceleration increases, the percentage difference of inter-story displacement ratio (IDR) between ASF, AMT and CIP underground structures in soft soil site decreases, and iii) prefabricated underground structures can be used in both soft rock and soft soil sites. In addition, the seismic performance of ASF is better than that of the AMT structure, which is better than that of the CIP structure. However, when the ASF structure is built in a site with high seismic intensity and poor soil conditions, special consideration must be given to waterproofing measures for wall-slab joints or setting up tongue and groove to improve the connection stiffness and reduce the deformation.