Use this URL to cite or link to this record in EThOS:
Title: GNSS-seismometry integration for rapid far-field displacement estimation
Author: Li, C.-Y.
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Accurate and rapid estimation of permanent surface displacement is required immediately after a slip event for earthquake monitoring or tsunami early warning. It is difficult to achieve the necessary accuracy at high- and low-frequencies using GNSS or seismometry alone, particularly where the sensor network has a low spatial resolution, such as in the third world or developing countries. Kalman filter algorithms with displacement and velocity states have been developed to combine GNSS and accelerometer observations to overcome the limitations of each other and to obtain the optimal displacement solutions. However, sawtooth-like phenomena caused by the bias or tilting of the sensor decrease the accuracy of the displacement estimates. In this study, a three-dimensional Kalman filter algorithm with an additional baseline error state has been developed. An experiment with both a GNSS receiver and a strong motion seismometer mounted on a movable steel plate on the platform and subjected to known displacements was carried out on the roof of the National Physical Laboratory. The results clearly show that the additional baseline error state enables the Kalman filter to estimate the instrument's sensor bias and tilt effects and correct the state estimates. Additional validation was conducted using data from the University of California, San Diego large outdoor shake table experiment. Finally, the proposed Kalman filter algorithm has been validated with data sets from the 2010 Mw 7.2 El Mayor-Cucapah Earthquake. The results indicate that the additional baseline error state cannot only eliminate the linear and quadratic drifts but also reduce the sawtooth-like effects from the displacement solutions. Conventional seismometric baseline-corrected results fail to resolve the permanent displacements after an earthquake; the two-state Kalman filter can only provide stable and optimal solutions if the strong motion seismometer had not been moved or tilted by shaking. The proposed Kalman filter algorithm can achieve rapid and stable displacements by estimating and correcting for the baseline error at each epoch. The integration filters out noise-like distortions and thus improves the real-time detection and measurement capability. The proposed approach includes not only a new form of state but also a new approach in seismometry for tuning the Kalman filter. Including baseline error state in the Kalman filter improves accuracy and information content (both permanent displacements and earthquake waveforms) in the output of integrated GNSS and accelerometer systems, both of which are vital in earthquake monitoring.
Supervisor: Ziebart, M. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available