Determination of High-pass Filter Frequency with Deep Learning for Ground Motion
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Accurate seismic vulnerability assessment requires high quality and large amounts of ground motion data. Ground motion data generated from time series contains not only the seismic waves but also the background noise. Therefore, it is crucial to determine the high-pass cut-off frequency to reduce the background noise. Traditional methods for determining the high-pass filter frequency are based on human inspection, such as comparing the noise and the signal Fourier Amplitude Spectrum (FAS), f2 trend line fitting, and inspection of the displacement curve after filtering. However, these methods are subject to human error and unsuitable for automating the process. This study used a deep learning approach to determine the high-pass filter frequency. We used the Mel-spectrogram for feature extraction and mixup technique to overcome the lack of data. We selected convolutional neural network (CNN) models such as ResNet, DenseNet, and EfficientNet for transfer learning. Additionally, we chose ViT and DeiT for transformer-based models. The results showed that ResNet had the highest performance with R2 (the coefficient of determination) at 0.977 and the lowest mean absolute error (MAE) and RMSE (root mean square error) at 0.006 and 0.074, respectively. When applied to a seismic event and compared to the traditional methods, the determination of the high-pass filter frequency through the deep learning method showed a difference of 0.1 Hz, which demonstrates that it can be used as a replacement for traditional methods. We anticipate that this study will pave the way for automating ground motion processing, which could be applied to the system to handle large amounts of data efficiently.
Abstract
Accurate seismic vulnerability assessment requires high quality and large amounts of ground motion data. Ground motion data generated from time series contains not only the seismic waves but also the background noise. Therefore, it is crucial to determine the high-pass cut-off frequency to reduce the background noise. Traditional methods for determining the high-pass filter frequency are based on human inspection, such as comparing the noise and the signal Fourier Amplitude Spectrum (FAS), f2 trend line fitting, and inspection of the displacement curve after filtering. However, these methods are subject to human error and unsuitable for automating the process. This study used a deep learning approach to determine the high-pass filter frequency. We used the Mel-spectrogram for feature extraction and mixup technique to overcome the lack of data. We selected convolutional neural network (CNN) models such as ResNet, DenseNet, and EfficientNet for transfer learning. Additionally, we chose ViT and DeiT for transformer-based models. The results showed that ResNet had the highest performance with R2 (the coefficient of determination) at 0.977 and the lowest mean absolute error (MAE) and RMSE (root mean square error) at 0.006 and 0.074, respectively. When applied to a seismic event and compared to the traditional methods, the determination of the high-pass filter frequency through the deep learning method showed a difference of 0.1 Hz, which demonstrates that it can be used as a replacement for traditional methods. We anticipate that this study will pave the way for automating ground motion processing, which could be applied to the system to handle large amounts of data efficiently.
참고문헌 (24)
- Ancheta TD, "NGA-West2 database", Earthquake Spectra, 30(3), pp.989~1005 (2014)
- Chiou B, "NGA Project strongmotion database", Earthquake Spectra, 24(1), pp.23~44 (2008)
- Goulet CA, "PEER NGA-East database", Earthquake Spectra, 37(1_suppl), pp.1331~1353 (2021)
- Akkar S, "Reference database for seismic ground-motion in Europe(RESORCE)", Bull Earthquake Eng, 12(1), pp.311~339 (2014)
- Choi S-W, "A Study on Development of an Earthquake Ground-motion Database Based on the Korean National Seismic Network", J Earthq Eng Soc Korea, 24(6), pp.277~283 (2020)
- Choi I, "A Fundamental Study on the Database of Response History for Historical Earthquake Records on the Korean Peninsula", KSCE Journal of Civil and Environmental Engineering Research, 39(6), pp.821~831 (2019)
- Boore DM, "Processing of strong-motion accelerograms : needs, options and consequences", Soil Dynamics and Earthquake Engineering, 25(2), pp.93~115 (2005)
- Ross ZE, "Generalized Seismic Phase Detection with Deep Learning", Bulletin of the Seismological Society of America, 108(5A), pp.2894~2901 (2018)
- Zhu W, "PhaseNet : A Deep-Neural-Network-Based Seismic Arrival Time Picking Method", Geophyiscal Journal Inter-national, 216(1), pp.261~273 (2019)
- Mousavi SM, "Earthquake transformer—an attentive deep-learning model for simultaneous earthquake detection and phase picking", Nat Commun, 11(1), pp.3952~ (2020)
- Wang B, "Deep-learning-based seismic data interpolation : A preliminary result", Geophysics, 84(1), pp.1JF~Z5 (2019)
- Derakhshani A, "Predicting the principal strong ground motion parameters : A deep learning approach", Applied Soft Computing, 80, pp.192~201 (2019)
- Seydoux L, "Clustering earthquake signals and background noises in continuous seismic data with unsupervised deep learning", Nat Commun, 11, pp.3972~ (2020)
- Yang Y, "Rapid Seismic Waveform Modeling and Inversion With Neural Operators", IEEE Trans Geosci Remote Sensing, 61, pp.1~12 (2023)
- Liu B, "The Cut-Off Frequency of High-Pass Filtering of Strong-Motion Records Based on Transfer Learning", Applied Sciences, 13(3), pp.1500~ (2023)
- He K, "Deep Residual Learning for Image Recognition", 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.770~778 (2016), IEEE
- "Densely Connected Convolutional Networks" (2017)[URL]
- Tan M, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks", PMLR, pp.6105~6114 (2019)
- "An image is worth 16x16 words: Transformers for image recognition at scale" (2020)[URL]
- Touvron H, "Training data-efficient image transformers & distillation through attention", PMLR, pp.10347~10357 (2021)
- Seo J, "Deep Learning-Based, Real-Time, False-Pick Filter for an Onsite Earthquake Early Warning(EEW)System", J Earthq Eng Soc Korea, 25(2), pp.71~81 (2021)
- Shakeel M, "EMC: Earthquake Magnitudes Classification on Seismic Signals via Convolutional Recurrent Networks", 2021 IEEE, 2021, 388~393
- "mixup: Beyond Empirical Risk Minimization" (2018)[URL]
- "Decoupled Weight Decay Regularization" (2019)[URL]