Prediction of On-Site Peak Ground Motion Based on Machine Learning and Transfer Learning
-
摘要: 在现地地震预警中，为了提高中国仪器地震烈度计算中地震动峰值（PGA和PGV）预测的准确性，提出了基于机器学习和迁移学习的现地地震动峰值预测方法.基于日本K-NET台网记录的强震动数据，使用神经网络建立了预训练的现地地震动峰值预测模型；基于中国的强震动数据和预训练的现地地震动峰值预测模型，通过迁移学习建立了用于中国的现地地震动峰值预测模型.对于日本和中国测试数据集以及泸定6.8级地震，在P波到达后3 s，和传统的现地地震动峰值预测方法相比，本研究提出的方法对于PGA预测和PGV预测有更小的平均绝对误差和标准差.结果表明，本研究提出的方法可以在一定程度上提高现地地震预警地震动峰值预测的可靠性，对于现地地震预警系统具有重要意义.Abstract: To improve the accuracy of peak ground motion (peak ground acceleration (PGA) and peak ground velocity (PGV) prediction in Chinese instrument seismic intensity calculation for on-site earthquake early warning (EEW), a prediction method of on-site peak ground motion based on machine learning and transfer learning is proposed. A pretrained on-site peak ground motion prediction model was established using neural networks based on strong motion data recorded by the K-NET network in Japan. Based on strong motion data from China and the pretrained on-site peak ground motion prediction model, an on-site peak ground motion prediction model for China was established through transfer learning. For the Japanese and Chinese test dataset and Luding M6.8 eathquake, at 3 s after the arrival of P-wave, compared to traditional on-site peak ground motion prediction method, the method proposed in this study has smaller mean absolute error and standard deviation for PGA prediction and PGV prediction. The results indicate that the method proposed in this study can improve the reliability of predicting peak ground motion in on-site EEW to a certain extent, which is of great significance for the development of on-site EEW systems.
-
Key words:
- earthquakes /
- machine learning /
- transfer learning /
- P wave /
- peak ground motion /
- artificial intelligence
-
图 1 日本数据集震中(a)和K-NET台站(b)的分布
Fig. 1. Distribution of epicenters (a) and K-NET stations (b) in the Japanese dataset
图 2 日本数据集震中距(a)和震级(b)在记录数量上的分布以及PGA(c)和PGV(d)在震中距上的分布
Fig. 2. The distribution of epicentral distance (a) and magnitude (b) in the number of records, and the distribution of PGA (c) and PGV (d) in epicentral distance in the Japanese dataset
图 3 中国数据集震中(a)和台站(b)的分布
Fig. 3. Distribution of epicenters (a) and stations (b) in the Chinese dataset
图 4 中国数据集震中距(a)和震级(b)在记录数量上的分布以及PGA(c)和PGV(d)在震中距上的分布
Fig. 4. The distribution of epicentral distance (a) and magnitude (b) in the number of records, and the distribution of PGA (c) and PGV (d) in epicentral distance in the Chinese dataset
图 6 OSnet-PGA模型(a)和OSnet-PGV模型(b)的损失曲线
Fig. 6. Loss curves of OSnet-PGA model (a) and OSnet-PGV model (b)
图 7 不同特征参数之间的相关系数
Fig. 7. Correlation coefficients between different characteristic parameters
图 8 TLOSnet-PGA模型(a)和TLOSnet-PGV模型(b)的损失曲线
Fig. 8. Loss curves of TLOSnet-PGA model (a) and TLOSnet-PGV model (b)
图 9 OSnet-PGA模型(a)、HybridNet-PGA模型(b)、RNN-PGA模型(c)、Pd-PGA方法(d)、OSnet-PGV模型(e)、HybridNet-PGV模型(f)、RNN-PGV模型(g)、Pd-PGV方法(h)在日本测试集上的PGA和PGV预测
Fig. 9. PGA and PGV predictions of OSnet-PGA model (a), HybridNet-PGA model (b), RNN-PGA model (c), Pd-PGA method (d), OSnet-PGV model (e), HybridNet-PGV model (f), RNN-PGV model (g) and Pd-PGV method (h) for Japanese test dataset
图 10 TLOSnet-PGA模型(a)、OSnet-PGA (日本)模型(b)、OSnet-PGA (中国)模型(c)、Pd-PGA方法(d)、TLOSnet-PGV模型(e)、OSnet-PGV (日本)模型(f)、OSnet-PGV (中国)模型(g)、Pd-PGV方法(h)在中国测试集上的PGA和PGV预测
Fig. 10. PGA and PGV predictions of TLOSnet-PGA model (a), OSnet-PGA (Japan) model (b), OSnet-PGA (China) model (c), Pd-PGA method (d), TLOSnet-PGV model (e), OSnet-PGV (Japan) model (f), OSnet-PGV (China) model (g), Pd-PGV method (h) for Chinese test dataset
图 11 泸定6.8级地震的震中和台站分布
Fig. 11. The epicenter and station distribution of the Luding M6.8 earthquake
图 12 TLOSnet-PGA模型(a)、OSnet-PGA (日本)模型(b)、OSnet-PGA (中国)模型(c)、Pd-PGA方法(d)、TLOSnet-PGV模型(e)、OSnet-PGV (日本)模型(f)、OSnet-PGV (中国)模型(g)和Pd-PGV方法(h)对于泸定6.8级地震的PGA和PGV预测误差分布
Fig. 12. Distribution of PGA and PGV prediction errors for the Luding M6.8 earthquake using TLOSnet-PGA model (a), OSnet-PGA (Japan) model (b), OSnet-PGA (China) model (c), and Pd-PGA method (d), TLOSnet-PGV model (e), OSnet-PGV (Japan) model (f), OSnet-PGV (China) model (g), and Pd-PGV method (h)
表 1 输入参数的计算公式
Table 1. Calculation formula for input parameters
参数 计算公式 单位 峰值位移Pd $ {P}_{\mathrm{d}}=\underset{0\le t\le 3}{\mathrm{m}\mathrm{a}\mathrm{x}}\left[{d}_{ud}\left(t\right)\right] $ cm 峰值速度Pv $ {P}_{v}=\underset{0\le t\le 3}{\mathrm{m}\mathrm{a}\mathrm{x}}\left[{v}_{ud}\left(t\right)\right] $ cm/s 峰值加速度Pa $ {P}_{a}=\underset{0\le t\le 3}{\mathrm{m}\mathrm{a}\mathrm{x}}\left[{a}_{ud}\left(t\right)\right] $ cm/s2 速度平方积分IV2 $ \mathrm{I}\mathrm{V}2={\int }_{0}^{3}{v}_{ud}^{2}\left(t\right)\mathrm{d}t $ cm2/s 累积绝对速度CAV $ \mathrm{C}\mathrm{A}\mathrm{V}={\int }_{0}^{3}\left|a\left(t\right)\right|\mathrm{d}t $ cm/s 阿里亚斯烈度IA $ {I}_{A}=\frac{\mathrm{\pi }}{2\times 9.8\times 100}{\int }_{0}^{3}{a}^{2}\left(t\right)\mathrm{d}t $ cm/s 竖向加速度之和SVA $ \mathrm{S}\mathrm{V}\mathrm{A}=\underset{0\le t\le 3}{\mathrm{s}\mathrm{u}\mathrm{m}}\left[\left|{a}_{ud}\left(t\right)\right|\right] $ cm/s2 竖向速度之和SVV $ \mathrm{S}\mathrm{V}\mathrm{V}=\underset{0\le t\le 3}{\mathrm{s}\mathrm{u}\mathrm{m}}\left[\left|{v}_{ud}\left(t\right)\right|\right] $ cm/s 竖向位移之和SVD $ \mathrm{S}\mathrm{V}\mathrm{D}=\underset{0\le t\le 3}{\mathrm{s}\mathrm{u}\mathrm{m}}\left[\left|{d}_{ud}\left(t\right)\right|\right] $ cm 构造参数TP $ {\tau }_{c}=\frac{2\mathrm{\pi }}{\sqrt[]{{\int }_{0}^{3}{v}_{ud}^{2}\left(t\right)\mathrm{d}t/{\int }_{0}^{3}{d}_{ud}^{2}\left(t\right)\mathrm{d}t}} $,
$ TP={\tau }_{c}\times {P}_{\mathrm{d}} $s·cm 
下载: 导出CSV
表 2 不同输入特征参数数量下OSnet模型的预测结果
Table 2. Prediction results of OSnet models for different numbers of input characteristic parameters
输入 PGA PGV MAE Std MAE Std Pa&Pv&Pd 0.213 0.266 0.227 0.285 Pa&Pv&Pd&IV2 0.206 0.258 0.218 0.274 Pa&Pv&Pd&IV2&CAV 0.201 0.251 0.213 0.268 Pa&Pv&Pd&IV2&CAV&IA 0.199 0.248 0.209 0.264 Pa&Pv&Pd&IV2&CAV&IA&SVA 0.198 0.248 0.209 0.264 Pa&Pv&Pd&IV2&CAV&IA&SVA&SVD 0.197 0.247 0.209 0.264 Pa&Pv&Pd&IV2&CAV&IA&SVA&SVD&TP 0.197 0.247 0.208 0.263 Pa&Pv&Pd&IV2&CAV&IA&SVA&SVD&TP&SVV 0.197 0.247 0.208 0.262
下载: 导出CSV
表 3 OSnet模型和基线模型在日本测试集上的PGA和PGV预测平均绝对误差和标准差
Table 3. Mean absolute error and standard deviation of PGA and PGV prediction using OSnet models and baseline models for Japanese test dataset
方法 MAE Std OSnet-PGA 0.197 0.247 HybridNet-PGA 0.200 0.250 RNN-PGA 0.199 0.251 Pd-PGA 0.256 0.321 OSnet-PGV 0.208 0.262 HybridNet-PGV 0.215 0.266 RNN-PGV 0.214 0.270 Pd-PGV 0.241 0.303
下载: 导出CSV
表 4 不同PGA范围下，OSnet-PGA模型和基线模型对于日本测试集的绝对误差（AE）在不同范围下的占比
Table 4. The proportions of different absolute errors (AE) of OSnet-PGA model and baseline models in different PGA ranges for Japanese test dataset
方法 PGA≤45.7 cm/s2 PGA > 45.7 cm/s2 AE < 0.4 AE≥0.4 AE < 0.4 AE≥0.4 OSnet-PGA 90.30% 9.70% 71.84% 28.16% HybridNet-PGA 90.05% 9.95% 66.09% 33.91% RNN-PGA 89.77% 10.23% 65.51% 34.49% Pd-PGA 81.29% 18.71% 27.01% 72.99%
下载: 导出CSV
表 5 不同PGV范围下，OSnet-PGV模型和基线模型对于日本测试集的绝对误差（AE）在不同范围下的占比
Table 5. The proportions of different absolute errors (AE) of OSnet-PGV model and baseline models in different PGV ranges for Japanese test dataset
方法 PGV≤3.81 cm/s PGV > 3.81 cm/s AE < 0.4 AE≥0.4 AE < 0.4 AE≥0.4 OSnet-PGV 87.56% 12.44% 65.33% 34.67% HybridNet-PGV 86.65% 13.35% 60.66% 39.34% RNN-PGV 87.13% 12.87% 44.00% 56.00% Pd-PGV 82.15% 17.85% 38.66% 61.34%
下载: 导出CSV
表 6 不同方法对于中国测试集的PGA和PGV预测的平均绝对误差和标准差
Table 6. The mean absolute error and standard deviation of PGA and PGV prediction for Chinese test dataset using different methods
方法 MAE Std TLOSnet-PGA 0.208 0.268 OSnet-PGA (日本) 0.217 0.285 OSnet-PGA (中国) 0.243 0.310 Pd-PGA 0.286 0.368 TLOSnet-PGV 0.233 0.297 OSnet-PGV (日本) 0.250 0.324 OSnet-PGV (中国) 0.251 0.326 Pd-PGV 0.271 0.349
下载: 导出CSV
表 7 不同PGA范围下，不同方法在中国测试集上的绝对误差（AE）在不同范围下的占比
Table 7. The proportions of different absolute errors (AE) of different methods in different PGA ranges for Chinese test dataset
方法 PGA≤45.7 cm/s2 PGA > 45.7 cm/s2 AE < 0.4 AE≥0.4 AE < 0.4 AE≥0.4 TLOSnet-PGA 88.65% 11.35% 74.22% 25.78% OSnet-PGA (日本) 86.79% 13.21% 71.87% 28.13% OSnet-PGA (中国) 81.80% 18.20% 75.78% 24.22% Pd-PGA 77.77% 22.23% 25.00% 75.00%
下载: 导出CSV
表 8 不同PGV范围下，不同方法在中国测试集上的绝对误差（AE）在不同范围下的占比
Table 8. The proportions of different absolute errors (AE) of different methods in different PGV ranges for Chinese test dataset
方法 PGV≤3.81 cm/s PGV > 3.81 cm/s AE < 0.4 AE≥0.4 AE < 0.4 AE≥0.4 TLOSnet-PGV 83.89% 16.11% 50.98% 49.02% OSnet-PGV (日本) 80.57% 19.43% 60.78% 39.22% OSnet-PGV (中国) 80.11% 19.89% 64.70% 35.30% Pd-PGV 77.51% 22.49% 33.33% 66.67%
下载: 导出CSV
表 9 不同方法对于泸定6.8级地震的PGA和PGV预测的平均绝对误差和标准差
Table 9. Mean absolute error and standard deviation of PGA and PGV prediction for the Luding M6.8 earthquake using different methods
方法 MAE Std TLOSnet-PGA 0.246 0.296 OSnet-PGA (日本) 0.283 0.324 OSnet-PGA (中国) 0.313 0.385 Pd-PGA 0.481 0.438 TLOSnet-PGV 0.240 0.299 OSnet-PGV (日本) 0.273 0.326 OSnet-PGV (中国) 0.368 0.385 Pd-PGV 0.280 0.315
下载: 导出CSV
表 10 不同PGA范围下，不同方法对于泸定6.8级地震的绝对误差（AE）在不同范围下的占比
Table 10. The proportion of absolute error (AE) of different methods for the Luding M6.8 earthquake in different PGA ranges
方法 PGA≤45.7 cm/s2 PGA > 45.7 cm/s2 AE < 0.4 AE≥0.4 AE < 0.4 AE≥0.4 TLOSnet-PGA 82.55% 17.45% 79.74% 20.26% OSnet-PGA (日本) 71.62% 28.38% 68.35% 31.65% OSnet-PGA (中国) 66.74% 33.26% 70.88% 29.45% Pd-PGA 42.09% 57.91% 70.88% 29.12%
下载: 导出CSV
表 11 不同PGV范围下，不同方法对于泸定6.8级地震的绝对误差（AE）在不同范围下的占比
Table 11. The proportion of absolute error (AE) of different methods for the Luding M6.8 earthquake in different PGV ranges
方法 PGV≤3.81 cm/s PGV > 3.81 cm/s AE < 0.4 AE≥0.4 AE < 0.4 AE≥0.4 TLOSnet-PGV 84.10% 15.90% 82.00% 18% OSnet-PGV (日本) 77.99% 22.01% 68.00% 32.00% OSnet-PGV (中国) 61.44% 38.56% 52.00% 48.00% Pd-PGV 74.51% 25.49% 72.00% 28.00%
下载: 导出CSV
-
Allen, R. M., Melgar, D., 2019. Earthquake Early Warning: Advances, Scientific Challenges, and Societal Needs. Annual Review of Earth and Planetary Sciences, 47: 361-388. https://doi.org/10.1146/annurev-earth-053018-060457 Brondi, P., Picozzi, M., Emolo, A., et al., 2015. Predicting the Macroseismic Intensity from Early Radiated P Wave Energy for On-Site Earthquake Early Warning in Italy. Journal of Geophysical Research: Solid Earth, 120(10): 7174-7189. https://doi.org/10.1002/2015jb012367 Chen, M., Wang, H., 2022. Explainable Machine Learning Model for Prediction of Ground Motion Parameters with Uncertainty Quantification. Chinese Journal of Geophysics, 65(9): 3386-3404 (in Chinese with English abstract). Chung, J., Gulcehre, C., Cho, K., et al., 2014. Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. ArXivv: 1412.3555. https://arxiv.org/abs/1412.3555v1 Fu, Z. Y., Li, D. Q., Wang, S., et al., 2023. Landslide Susceptibility Assessment Based on Multitemporal Landslide Inventories and TrAdaBoost Transfer Learning. Earth Science, 48(5): 1935-1947 (in Chinese with English abstract). Hsu, T. Y., Huang, C. W., 2021. Onsite Early Prediction of PGA Using CNN with Multi-Scale and Multi-Domain P-Waves as Input. Frontiers in Earth Science, 9: 626908. https://doi.org/10.3389/feart.2021.626908 Hsu, T. Y., Pratomo, A., 2022. Early Peak Ground Acceleration Prediction for On-Site Earthquake Early Warning Using LSTM Neural Network. Frontiers in Earth Science, 10: 911947. https://doi.org/10.3389/feart.2022.911947 Hu, J. J., Ding, Y. T., Zhang, H., et al., 2023. A Real-Time Seismic Intensity Prediction Model Based on Long Short-Term Memory Neural Network. Earth Science, 48(5): 1853-1864 (in Chinese with English abstract). Kamigaichi, O., Saito, M., Doi, K., et al., 2009. Earthquake Early Warning in Japan: Warning the General Public and Future Prospects. Seismological Research Letters, 80(5): 717-726. https://doi.org/10.1785/gssrl.80.5.717 Kanamori, H., 2005. Real-Time Seismology and Earthquake Damage Mitigation. Annual Review of Earth and Planetary Sciences, 33: 195-214. https://doi.org/10.1146/annurev.earth.33.092203.122626 Kanamori, H., 2015. Earthquake Hazard Mitigation and Real-Time Warnings of Tsunamis and Earthquakes. Pure and Applied Geophysics, 172(9): 2335-2341. https://doi.org/10.1007/s00024-014-0964-y Kong, Q. K., Wang, R. J., Walter, W. R., et al., 2022. Combining Deep Learning with Physics Based Features in Explosion-Earthquake Discrimination. Geophysical Research Letters, 49(13): e2022GL098645. https://doi.org/10.1029/2022gl098645 LeCun, Y., Bengio, Y., Hinton, G., 2015. Deep Learning. Nature, 521(7553): 436-444. https://doi.org/10.1038/nature14539 Li, Z. F., 2022. A Generic Model of Global Earthquake Rupture Characteristics Revealed by Machine Learning. Geophysical Research Letters, 49(8): e2021GL096464. https://doi.org/ 10.1029/2021GL096464 Liu, C., Li, X. J., Jing, B. B., et al., 2019. The Distance Segmentation Characters of PGV-Pd Relationship Parameters for Earthquake Early Warning. Chinese Journal of Geophysics, 62(4): 1413-1426 (in Chinese with English abstract). Lü, S., Fang, L. H., Ren, H. Y., et al., 2024. Overview of the Earthquake Monitoring Status in the United States. China Earthquake Engineering Journal, 46(2): 431-448 (in Chinese with English abstract). Ma, Q., 2008. Study and Application on Earthquake Early Warning (Dissertation). Institute of Engineering Mechanics, China Earthquake Administration, Harbin (in Chinese with English abstract). Ma, Q., Jin, X., Li, S. Y., et al., 2013. Automatic P-Arrival Detection for Earthquake Early Warning. Chinese Journal of Geophysics, 56(7): 2313-2321 (in Chinese with English abstract). Mousavi, S. M., Beroza, G. C., 2020. A Machine-Learning Approach for Earthquake Magnitude Estimation. Geophysical Research Letters, 47(1): e2019GL085976. doi: 10.1029/2019GL085976 Pan, S. J., Yang, Q., 2010. A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10): 1345-1359. https://doi.org/10.1109/TKDE.2009.191 Peng, C. Y., Ma, Q., Jiang, P., et al., 2020. Performance of a Hybrid Demonstration Earthquake Early Warning System in the Sichuan-Yunnan Border Region. Seismological Research Letters, 91(2A): 835-846. https://doi.org/10.1785/0220190101 Serdar Kuyuk, H., Allen, R. M., Brown, H., et al., 2014. Designing a Network-Based Earthquake Early Warning Algorithm for California: ElarmS-2. Bulletin of the Seismological Society of America, 104(1): 162-173. https://doi.org/10.1785/0120130146 Song, J., Zhu, J., Li, S., 2023. MEANet: Magnitude Estimation via Physics-Based Features Time Series, an Attention Mechanism, and Neural Networks. Geophysics, 88(1): V33-V43. https://doi.org/10.1190/geo2022-0196.1 Song, J. D., Jiao, C. C., Li, S. Y., et al., 2018. Prediction Method of First-Level Earthquake Warning for High Speed Railway Based on Two-Parameter Threshold of Seismic P-Wave. China Railway Science, 39(1): 138-144 (in Chinese with English abstract). Song, J. D., Yu, C., Li, S. Y., 2021. Continuous Prediction of Onsite PGV for Earthquake Early Warning Based on Least Squares Support Vector Machine. Chinese Journal of Geophysics, 64(2): 555-568 (in Chinese with English abstract). State Administration for Market Regulation, Standardization Administration of China, 2020. GB/T‐17742‐2020, The Chinese Seismic Intensity Scale. China Quality and Standards Publishing & Media Co., Ltd., Beijing (in Chinese). Wald, D. J., Quitoriano, V., Heaton, T. H., et al., 1999. Relationships between Peak Ground Acceleration, Peak Ground Velocity, and Modified Mercalli Intensity in California. Earthquake Spectra, 15(3): 557-564. https://doi.org/10.1193/1.1586058 Wang, C. Y., Huang, T. C., Wu, Y. M., 2022. Using LSTM Neural Networks for Onsite Earthquake Early Warning. Seismological Research Letters, 93(2A): 814-826. https://doi.org/10.1785/0220210197 Wang, D., Sun, K., 2022. How the Big Data Seismology and AI Refine Rapid Determination of Source Parameters of Large Earth-Quakes?. Earth Science, 47(10): 3915-3917 (in Chinese with English abstract). Wang, J., Yang, Y., Mao, J. H., et al., 2016. CNN-RNN: A Unified Framework for Multi-Label Image Classification. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 2285-2294. https://doi.org/10.1109/CVPR.2016.251 Wu, L. Y., Tong, J. B., Wang, Z. F., et al., 2023. Classification of Damaged Grade on Rural Houses after Flood Disaster Based on Deep Convolutional Neural Network and Transfer Learning. Earth Science, 48(5): 1742-1754 (in Chinese with English abstract). Wu, Y. M., Kanamori, H., 2005. Rapid Assessment of Damage Potential of Earthquakes in Taiwan from the Beginning of P Waves. Bulletin of the Seismological Society of America, 95(3): 1181-1185. https://doi.org/10.1785/0120040193 Yu, C., Song, J. D., Li, S. Y., 2021. Prediction of Ground Motion for On-Site Earthquake Early Warning Based on SVM. Journal of Vibration and Shock, 40(3): 63-72, 80 (in Chinese with English abstract). Zhang, H., Jin, X., Wei, Y. X., et al., 2016. An Earthquake Early Warning System in Fujian, China. Bulletin of the Seismological Society of America, 106(2): 755-765. https://doi.org/10.1785/0120150143 Zhou, L. Q., Zhao, C. P., Zhang, M., et al., 2021. Machine-Learning-Based Earthquake Locations Reveal the Seismogenesis of the 2020 Mw 5.0 Qiaojia, Yunnan Earthquake. Geophysical Journal International, 228(3): 1637-1647. https://doi.org/10.1093/gji/ggab420 Zhu, J., Li, Z. F., Fang, L. H., 2023a. USTC-Pickers: A Unified Set of Seismic Phase Pickers Transfer Learned for China. Earthquake Science, 36(2): 95-112. https://doi.org/10.1016/j.eqs.2023.03.001 Zhu, J. B., Li, S. Y., Wei, Y. X., et al., 2023b. On-Site Instrumental Seismic Intensity Prediction for China via Recurrent Neural Network and Transfer Learning. Journal of Asian Earth Sciences, 248: 105610. https://doi.org/10.1016/j.jseaes.2023.105610 Zhu, J. B., Song, J. D., Li, S. Y., 2024. Influence of Low-Cost Sensors on Earthquake Early Warning Magnitude Estimation Using Convolutional Neural Network Model. Journal of Harbin Institute of Technology, 56(6): 81-90 (in Chinese with English abstract). Zhu, J. B., Li, S. Y., Song, J. D., 2022. Hybrid Deep-Learning Network for Rapid On-Site Peak Ground Velocity Prediction. IEEE Transactions on Geoscience and Remote Sensing, 60: 5925712. https://doi.org/10.1109/TGRS.2022.3230829 Zollo, A., Amoroso, O., Lancieri, M., et al., 2010. A Threshold-Based Earthquake Early Warning Using Dense Accelerometer Networks. Geophysical Journal International, 183(2): 963-974. https://doi.org/10.1111/j.1365-246X.2010.04765.x 陈蒙, 王华, 2022. 地震动强度参数估计的可解释性与不确定度机器学习模型. 地球物理学报, 65(9): 3386-3404. 付智勇, 李典庆, 王顺, 等, 2023. 基于多时空滑坡编录和TrAdaBoost迁移学习的滑坡易发性评价. 地球科学, 48(5): 1935-1947. doi: 10.3799/dqkx.2023.013 胡进军, 丁祎天, 张辉, 等, 2023. 基于长短期记忆神经网络的实时地震烈度预测模型. 地球科学, 48(5): 1853-1864. doi: 10.3799/dqkx.2022.338 刘辰, 李小军, 景冰冰, 等, 2019. 地震预警PGV-Pd关系参数的距离分段特征. 地球物理学报, 62(4): 1413-1426. 吕帅, 房立华, 任华育, 等, 2024. 美国地震监测现状综述. 地震工程学报, 46(2): 431-448. 马强, 2008. 地震预警技术研究及应用(博士学位论文). 哈尔滨: 中国地震局工程力学研究所. 马强, 金星, 李山有, 等, 2013. 用于地震预警的P波震相到时自动拾取. 地球物理学报, 56(7): 2313-2321. 国家市场监督管理总局, 国家标准化管理委员会, 2020. GB/T‐17742‐2020, 中国地震烈度表. 北京: 中国标准出版社. 宋晋东, 教聪聪, 李山有, 等, 2018. 基于地震P波双参数阈值的高速铁路Ⅰ级地震警报预测方法. 中国铁道科学, 39(1): 138-144. 宋晋东, 余聪, 李山有, 2021. 地震预警现地PGV连续预测的最小二乘支持向量机模型. 地球物理学报, 64(2): 555-568. 王墩, 孙琨, 2022. 地震大数据和AI如何改进全球大震参数快速测定?. 地球科学, 47(10): 3915-3917. doi: 10.3799/dqkx.2022.863 吴禄源, 仝敬博, 王自法, 等, 2023. 基于深度卷积神经网络和迁移学习的农村房屋洪涝灾害后受损等级分类. 地球科学, 48(5): 1742-1754. doi: 10.3799/dqkx.2022.502 余聪, 宋晋东, 李山有, 2021. 基于支持向量机的现地地震预警地震动峰值预测. 振动与冲击, 40(3): 63-72, 80. 朱景宝, 宋晋东, 李山有, 2024. 地震烈度仪对卷积神经网络模型地震预警震级估计的影响. 哈尔滨工业大学学报, 56(6): 81-90. -
点击查看大图
计量
- 文章访问数: 30
- HTML全文浏览量: 14
- PDF下载量: 5
- 被引次数: 0
