Journal of Earthquake and Tsunami

With the progress of technologies in data collection systems, structural engineers are faced with a large amount of unlabeled data gathered from various states of different structures, directing studies toward using improved classifiers in machine learning approaches of structural health monitoring. In this research, Laplacian Support Vector Machines (LapSVMs) are used as a semi-supervised learning algorithm to classify mentioned data. Using vibrational data of structure, dynamic properties of the structure are extracted. Modal strain energy is used as damage sensitive features (DSF) to perform damage assessment in a statistical pattern recognition framework by a semi-supervised learning algorithm using LapSVMs to classify unlabeled and labeled data. Also, Support Vector Machines (SVMs) and Regularized Least Square Classifier (RLSC) are used as classifiers to compare results. To compare the effectiveness of the proposed algorithm, different states of structural response are determined by labeled and unlabeled data. These results show high accuracy of LapSVM methods compared to others in cases where the labeled dataset is small.

Continuous shear wave velocity measurements are conducted on cylindrical triaxial sandy specimens consolidated with isotropic and anisotropic stress histories to characterize the evolution of the elastic shear modulus under different stress paths. The shear wave velocity is measured using an automatic device capable of consecutive measurements of travel time. All specimens were prepared using the Water Sedimentation (WS) method, and after undergoing a stress-controlled path representing the stress history of the specimens, they were subjected to stress-controlled drained shearing along eight different paths. The stress components used in the existing correlations estimating the shear moduli are specified under different stress paths. The results demonstrate that the existing empirical correlations can appropriately be used within the elastic region or the loading surface. However, the accuracy of such models estimating shear moduli might be depreciated as shearing commences, particularly if the stress path involves deviatoric stress, diverging from the expected trends at a specific stress ratio, which is strongly linked to the consolidation stress state. A criterion at which the shear modulus is diverted is also suggested, found to be directly related to the initial stress anisotropy.

Seismic isolation structures have excellent seismic mitigation effects, but implementing isolation measures may increase the initial cost of the structure. This paper presents a method to calculate the seismic loss and assess the life-cycle cost of isolated structures, which is convenient for owners to make decisions. The life-cycle cost assessment model is divided into two parts: initial cost assessment and seismic loss assessment over life-cycle period. Among them, the assessment method of the isolation layer is proposed, including the cost assessment of isolation bearings and dampers. The assessment method of seismic loss considers direct loss, indirect loss, casualties and integrating casualties into a unified assessment system. Direct loss includes building loss, indirect loss includes rental loss, relocation loss, income loss, personal property loss and business inventory loss. The seismic loss model is associated with seismic damage probability, which is obtained through seismic vulnerability and seismic probability risk assessment. The 8-story base-isolated-reinforced concrete structure illustrates the feasibility of the life-cycle-cost assessment model. The example shows that the initial cost of the isolated structure exceeds that of the nonisolated structure, however, the rental loss, income loss, personal property loss and casualties are much lower than those of nonisolated structures. The income loss of the isolated structure is the maximum proportion in life-cycle cost, and the casualty loss is the minimum proportion in the seismic loss, due to its excellent seismic mitigation effects.

A performance-based seismic design method using the elastic displacement design spectrum based on equal displacement approximation is proposed for the Hinged Wall with BRBs at Base (HWBB) structural system. HWBB is simplified as an Equivalent Single Degree of Freedom (ESDOF) system. Design flowcharts are given. The proposed design approach aims to limit the lateral displacements to an allowable target displacement. Design examples are presented and target displacements are compared with numerical results. The results indicate that the plastic response of the HWBB structural system is accurately predicted by the elastic displacement spectrum, which verified that the performance-based design method of HWBB is accurate.

In this paper, a series of 1-g shaking table model tests were carried out to investigate the seismic behavior of a relatively stiff rectangular tunnel structure installed in soft clay bed, accounting for different ground motions with varying peak accelerations. Using a validated numerical analysis procedure, a suite of three-dimensional (3D) finite element (FE) analyses was performed to systematically study the factors of tunnel burial depth, seismic intensity, flexural rigidity of the middle column and tunnel wall thickness on the seismic response of rectangular tunnel structures installed in clayey ground. It was found that, with the increasing burial depth, the seismic response of rectangular tunnel structure became more intense due to the increased inertial force arising from the overlying clay; in terms of improving the seismic performance of tunnel structure, increasing tunnel wall thickness seemed more effective than increasing flexural rigidity of the middle column. Furthermore, it was found that the existing simplified approaches generally tended to overestimate the earthquake-induced racking distortions of rectangular tunnels installed in clayey ground. A new semi-empirical relationship was derived for better correlating the racking ratio with flexibility ratio for rectangular tunnels embedded in soft clays, which could provide a useful reference for the seismic design and risk assessment of similar clay-tunnel systems.

Studies on depth spatial variations are rare because of the limited number of measured recordings of the subsurface along the depth direction from the ground surface. However, these depth variations can directly influence the seismic dynamic response of underground structures or the foundations of space structures through multi-support excitation. In this study, depth spatial variations, including phase and amplitude, are analyzed to describe ground motions along different depths based on surface-earthquake time histories. The observed recordings from the Center for Engineering Strong Motion Data are used to investigate the spatial variation characteristics of the phase and amplitude in the depth direction, represented by the lagged coherence and Fourier amplitude spectrum ratio, respectively. A corresponding model is established under various site conditions to simulate multipoint correlated ground motions along the depth from the surface. Finally, a numerical simulation of multipoint ground motions is conducted to verify the application of the established models. This study provides an effective tool for describing spatial variations in both phase and amplitude, which are crucial for simulating ground motions at various depths.