Rapid on-site peak ground acceleration estimation based on support vector regression and P-wave features in Taiwan

Comparing the Performance of the NEEWS Earthquake Early Warning System Against the CWB System During the 6 February 2018 Mw 6.2 Hualien Earthquake

“…Hsu, tyhsu@mail.ntust The NEEWS estimates the peak ground acceleration (PGA) of an imminent earthquake from the same station by relying on a recently developed support vector machine (SVM) technique. During the Meinong earthquake in 2016, which had a moment magnitude of 6.5, the performance of the NEEWS was remarkable, the details of which have been summarized and published (Hsu et al, 2013(Hsu et al, , 2016. These features include the predominant period, peak acceleration amplitude, peak velocity amplitude, peak displacement amplitude, cumulative absolute velocity, and the integral of the squared velocity.…”

“…The on-site EEW approach has recently been the subject of numerous studies worldwide (Allen et al, 2009;Böse et al, 2012;Carranza et al, 2013;Caruso et al, 2017;Emolo et al, 2016;Hsu et al, 2013;Kanamori, 2005;Nakamura, 1998;Odaka et al, 2003;Peng et al, 2015;Wu et al, 2013;Zollo et al, 2010). The on-site EEW approach has recently been the subject of numerous studies worldwide (Allen et al, 2009;Böse et al, 2012;Carranza et al, 2013;Caruso et al, 2017;Emolo et al, 2016;Hsu et al, 2013;Kanamori, 2005;Nakamura, 1998;Odaka et al, 2003;Peng et al, 2015;Wu et al, 2013;Zollo et al, 2010).…”

Two Novel Approaches to Reduce False Alarm Due to Non‐Earthquake Events for On‐Site Earthquake Early Warning System

“…These numbers are listed in Table as being 595 and 1,371 for earthquakes and unknown events, respectively. In addition, for the purpose of reducing the time consumed for training the SVC model, only 3,942 vibration records selected from the 71 representative earthquakes (Hsu et al., ) with local magnitudes, between 3.0 and 7.3 in database I are selected for training the SVC model, and the training and total numbers of vibration records in database I are shown in Table . After the establishment of the SVC model, the proposed approaches are then tested by utilizing all the data in database I.…”

“…For regions that are close to the epicenter, where seismic intensity is usually much higher than regions outside, the lead‐time before a destructive wave arrives provided by the regional warning can be null. However, an on‐site warning can provide more lead‐time at the regions that are close to an epicenter because only the seismic information of the target site is required, and the intensity of the forthcoming earthquake is predicted by the extraction of some P‐wave features from the first few seconds of vertical ground acceleration of a single station (Wu and Kanamori, ; Tsai et al., ; Nakamura et al., ; Böse et al., ; Hsu et al., ). Therefore, the increase in accuracy of on‐site warnings is one of the key advantages for improving the effect of EEW techniques in Taiwan.…”

Estimation of peak floor acceleration based on support vector regression and p-wave features

“…In order to enable earlier earthquake alerts to reduce possible casualties when an earthquake occurs, the National Center for Research on Earthquake Engineering (NCREE) developed an on-site EEWS, which can predict the PGA of the whole time history of the earthquake before the major vibration arrives [1]. The entire operation process is shown in Figure. 1.…”