THESIS
2019
x, 145 pages : color illustrations ; 30 cm
Abstract
Landslides pose a large threat to both human lives and infrastructures. Hence, slope health
monitoring is an important task to protect the slopes from catastrophic failures. In this research, vibration data continuously collected from the sensors installed on the landslide-prone areas
were used to infer slope properties such as the site directivity and peak frequencies using the horizontal to vertical spectral ratio (HVSR) method, and then to identify any potential changes
in the slope health condition.
Over 2 years of data obtained from two study sites located in Lushan and Lishan of Taiwan were analyzed. The effects of earthquake amplitude and epicenter location on site directivity
and HVSR was investigated, and analysis results showed that derivation of directivity and peak freq...[
Read more ]
Landslides pose a large threat to both human lives and infrastructures. Hence, slope health
monitoring is an important task to protect the slopes from catastrophic failures. In this research, vibration data continuously collected from the sensors installed on the landslide-prone areas
were used to infer slope properties such as the site directivity and peak frequencies using the horizontal to vertical spectral ratio (HVSR) method, and then to identify any potential changes
in the slope health condition.
Over 2 years of data obtained from two study sites located in Lushan and Lishan of Taiwan were analyzed. The effects of earthquake amplitude and epicenter location on site directivity
and HVSR was investigated, and analysis results showed that derivation of directivity and peak frequencies using the HVSR technique were unaffected by the shaking magnitude and
epicenter location. Besides, in order to facilitate the classification of signals under various situations for subsequent risk management in the future, signals collected by the sensors were browsed and classified to establish patterns that can be used to identify future signals for different responses. 4 different sources of data, including ambient, earthquake, slope movement, and vehicle signals were classified and summarized. Finally, a comparison of site directivity
and HVSR before and after a landslide event was completed with both sensors to check the health condition of the slope. While no significant changes were found for both sites, the capability to derive slope condition parameters was demonstrated.
Post a Comment