PDF
Full
Outline
Ground motion parameters quantify the intensity of ground motion and their impact on building structures, making the selection of appropriate parameters crucial for pre-earthquake seismic design and post-earthquake damage assessment. Ground motion parameters and structural seismic responses are often statistically related through traditional correlation analysis and regression methods. However, data are mostly sourced from numerical simulations, which makes it difficult to capture the true nonlinear mapping relationship between the two. Therefore, this paper collected and organized nearly 1.28 million actual damage records from the Great East Japan Earthquake on March 11, 2011, and the complex mapping relationship between ground motion parameters and building damage levels was established based on four machine learning classification models, namely, XGBoost (eXtreme gradient boosting), RF(random forest), LightGBM (light gradient boosting machine), and CatBoost (categorical boosting). The SMOTE oversampling and Bayesian hyperparameter optimization algorithms were introduced to optimize the model, and the optimal combination of seven ground motion parameters was selected using two methods for evaluating feature importance. The results indicate that the XGBoost algorithm performs the best, with an overall accuracy of 71.39% on the test set. The optimal combination of ground motion parameters includs PGA, Td, VSI, PGD, PGV/PGA, PGV, and Sa. The amplitude, spectrum, and duration parameters of the ground motion show a strong correlation with post-earthquake building damage, while the cumulative energy parameters exhibit a weaker correlation. Finally, an earthquake loss prediction model based on the XGBoost algorithm was established using actual damage data from three earthquakes in New Zealand, validating the completeness, reliability, and regional generalization capability of the selected parameter combination. The research results can provide a theoretical basis and engineering reference for the seismic design of buildings and earthquake risk assessment.
PDF
70
33
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62199257
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
