간행물
Journal of the Earthquake Engineering Society of Korea KCI 등재 한국지진공학회논문집
- 발행기관 한국지진공학회
- 자료유형 학술지
- 간기 격월간
- ISSN 1226-525x (Print)2234-1099 (Online)
- 수록기간 1997 ~ 2026
- 주제분류 공학 > 토목공학 공학 분류의 다른 간행물
- 십진분류KDC 531DDC 624
권호리스트/논문검색
제30권 제5호(통권 제173호) (2026년 9월) 4건
1.
2026.09
구독 인증기관 무료, 개인회원 유료
김수빈, 이기학, 김형근, 신지욱
Piloti-type buildings are vulnerable to earthquakes because the soft story formed on the first floor concentrates structural damage in the lower story during seismic events. This highlights the need for a methodology that can rapidly and accurately predict the backbone curve, a key indicator of seismic performance. Accordingly, this study developed a code-based combined model for predicting the backbone curve of piloti-type RC buildings using regression-based machine learning. The model used nine input variables, one of which was the failure mode, derived from a previously developed prediction model. Optimal models for predicting displacement and strength at the three key points of the backbone curve—yield, ultimate, and residual—were selected based on regression performance metrics and combined in code to develop the final prediction model. To verify the proposed methodology, a comparative analysis with experimental results of piloti-type buildings was conducted based on key indicators of lateral resistance capacity: effective stiffness, strength ratio, and ductility. The results confirmed that the developed machine learning model reliably predicts the backbone curve, demonstrating its potential as a rapid and efficient alternative to conventional numerical analysis methods.
4,000원
2.
2026.09
구독 인증기관 무료, 개인회원 유료
김영우, 이건찬, 이창석, 전종수
Shape memory alloy (SMA) rebars have been investigated as a seismic retrofit for reinforced concrete (RC) columns because of their superelastic behavior and self-centering capability. Since SMA rebars are more expensive than conventional steel rebars, they should be selectively used in the plastic hinge region, where seismic damage concentrates, which requires a reliable estimate of the plastic hinge length. In this study, a model for predicting the plastic hinge length of circular SMA-RC columns was developed using extreme gradient boosting. A numerical database of 52,200 cases was constructed by combining column and section dimensions from circular RC column databases with probabilistically sampled material properties, of which 37,235 valid cases were used for model development. Moment-curvature analysis of sampled column sections was then performed to calculate the plastic hinge length from the maximum section moment and the moment at the forward transformation start strain of the compression-side SMA rebar. The proposed model achieved a coefficient of determination (R2) of 0.998 and a root mean square error (RMSE) of 0.010, whereas the existing equations for SMA-RC columns yielded negative R2 values. Shapley additive explanations analysis was then conducted to interpret the model, showing that the aspect ratio, transverse reinforcement ratio, and SMA transformation characteristics had the strongest influence on the predicted plastic hinge length. By combining predictive accuracy with interpretability, the proposed model can serve as an effective data-driven tool for estimating the plastic hinge length of SMA-RC columns.
4,000원
3.
2026.09
구독 인증기관 무료, 개인회원 유료
김태현, 초우드리 엠디 아드난 하크, 노화성
This study comparatively analyzes conventional machine learning models and a stacking ensemble model for predicting the peak displacement response of concrete columns under seismic loading. Using geometric parameters, material properties, and ground motion characteristics as inputs, a database of 1,080 cases was constructed via finite-element time-history analysis. Hyperparameter optimization was applied to six conventional machine learning models based on tree-based, kernel-based, and neural network approaches. A stacking ensemble is constructed with Extra Trees and XGBoost as base models and RidgeCV as the meta-model, and its optimal configuration is derived by examining meta-model selection, passthrough application, and the number of cross-validation folds. Predicted peak displacement responses were converted into drift ratios normalized by column height to evaluate predictive performance across models. The tree-based models show stable performance with a MAPE of 3-4%, while the stacking ensemble achieves the lowest error rate at 2.68% MAPE. Interval-based analysis shows that the stacking ensemble yields lower error rates than conventional machine learning models across a range of geometric parameters, including column height, cross-sectional dimensions, and aspect ratio, confirming stable predictive performance under variations in these parameters.
4,000원
4.
2026.09
구독 인증기관 무료, 개인회원 유료
박예원, 김지현, 전종수
Seismic resilience assessment of railway networks requires fragility functions for all bridges in the network. However, developing detailed probabilistic models and conducting nonlinear time-history analyses for each bridge configuration requires substantial computational effort. To overcome this limitation, this study proposes adjustment factors for system-level seismic fragility curves for prestressed concrete box-girder bridges, the most common class of high-speed railway bridges in Korea, with different numbers of spans. A database was constructed from design documents of 69 bridges in the first phase of the Honam High-Speed Railway, and major uncertain model parameters were identified. Three-dimensional probabilistic track-bridge models were generated using Latin hypercube sampling, and nonlinear time-history analyses were performed for 2-, 3-, 4-, 5-, 10-, and 20-span bridge models. Component fragility curves were derived using probabilistic seismic demand and limit-state models, and system fragility curves were obtained under a series-system assumption. Bearing responses dominated the slight-limit-state system fragility, followed by column responses. As the number of spans increased from 2 to 20, the median value of the fragility curve decreased by approximately 67%. Finally, adjustment factors for the median value and dispersion of the fragility curve were proposed to estimate the seismic vulnerability of bridges with different numbers of spans without additional computational effort.
4,000원
