The Complementary Characteristics of GPS and Accelerometers in Monitoring Structural Deformation
The effects of natural phenomena on buildings and other structures have profound engineering and safety implications. For several years, efforts have been under way to use high-precision GPS techniques and, separately or together with, accelerometer technology to monitor these effects in real-time. In this article, researchers from Australia and Japan describe a new technique for converting and combining measurements from an integrated system, as well as the field results from an installation subjected to earthquake and typhoon.
Severe loading conditions such as strong winds and earthquakes acting on modern tall buildings and structures can cause significant loads and vibrations. Recent trends toward slender, flexible, and light-weight buildings have left a large number of buildings susceptible to wind-induced motion. Furthermore, human perception of building motion has become a critical consideration in modern building design.
More complex building shapes and structural systems further accentuate eccentricities between the mass center, the elastic center, and the instantaneous point of application of aerodynamic loads, and consequently will generate significant torsional effects.
Verifying dynamic structural analysis requires the development of direct dynamic measurement tools and techniques in order to determine the natural frequencies, damping characteristics, and mode shapes. Among these tools accelerometers have played the most important part in analyzing structural response due to severe loading conditions. However, they provide only a relative acceleration measurement. The displacement from acceleration measurement cannot be obtained directly by double integration.
In contrast to accelerometers, GPS can directly measure position coordinates, thereby providing an opportunity to monitor, in real-time and full scale, the dynamic characteristics of a structure. GPS used in the real-time kinematic mode (GPSRTK) offers direct displacement measurements for dynamic monitoring. Earlier studies by the authors and other researchers, referenced in the Additional Resources section at the end of this article, have shown the efficiency and feasibility of structural deformation monitoring by combining accelerometer and GPS-RTK.
However, GPS-RTK has its own limitations. For example, the measurement accuracy can be affected by multipath and depends strongly on satellite geometry. Moreover, the typical GPS-RTK 20Hz sampling rate will limit its capability in detecting certain high mode signals of some structures. The new 100Hz GPS-RTK systems need to be further tested in order to ensure the independence of the measurements.
In order to exploit the advantages of both GPS-RTK and accelerometers, two data processing strategies have typically been used, namely to convert GPS measured displacement to acceleration through double differentiation and compare it with the accelerometer measurements (what we refer to as forward transformation), or to convert the accelerometer measurements into displacement through double integration and compare it with GPS measured displacement (the reverse transformation).
The latter approach is much more challenging because we have to determine two integration constants in order to recover all the components of displacement (static, quasi-static and dynamic). If the structure to be monitored is subject to a quasi-static force, as in the case of a typhoon, this further complicates the analysis.
Although earlier research has proposed a lab-based threshold setting for accelerometers to deal with the quasi-static issue, we believe that avoiding this procedure and developing new ways to recover the false and missing measurements from GPS by acceleration transformation would provide a preferred approach.
This article discusses recent efforts to design such a system based on a new integration approach that employs the correlation signals directly detected from a GPS-RTK system and an accelerometer to transform one form of measurement to the other. The methodology consists of a Fast Fourier Transform (FFT) for correlated signal identification, a filtering technique, delay compensation, and velocity linear trend estimation from both GPS and accelerometer measurements. We also present results derived from its installation on structures in Japan that subsequently experienced the effects of an earthquake and typhoon.
(For the rest of this story, please download the complete article using the PDF link above.)
The GPS receiver used in the Japan installation was an MC1000, Leica Geosystems, Heerbrugg, Switzerland. Other equipment included a servo type accelerometer, JA-24MA03, Japan Aviation Electronics Industry, Limited, Tokyo, Japan; a three-cup AF860 anemometer and vane, Makino Applied Instruments, Japan; and an FLA-5-11-5LT strain gauge, Tokyo Sokki Kenkyujo Co., Ltd., Tokyo, Japan.
Jean (Xiaojing) Li holds a B.Eng. in optical engineering from the Wuhan Technical University of Surveying and Mapping (WTUSM), P.R. China. From 1997 to 1998 she was a research assistant at the Meteorological Research Institute of Japan, researching real-time seismology. She is currently a Ph.D. student at the University of New South Wales (UNSW), Sydney, Australia. Her current research interests are the integration of GPS-RTK, accelerometer, optical fiber sensors, and digital signal processing.
Chris Rizos is a professor and head of school at the School of Surveying & Spatial Information Systems, UNSW. He obtained a bachelor of surveying and a Ph.D. both from the UNSW. Rizos has been researching the technology and high precision applications of GPS since 1985 and is currently leader of the Satellite Navigation and Positioning group at UNSW. He is a fellow of the Australian Institute of Navigation, a fellow of the International Association of Geodesy (IAG), and is currently president of the IAG’s Commission 4 “Positioning and Applications.”
Linlin Ge is a senior lecturer at the School of Surveying & Spatial Information Systems, UNSW. He graduated with a B.Eng. from the Wuhan Technical University of Surveying and Mapping WTUSM, a M.Sc. from the Institute of Seismology, and a Ph.D. from the UNSW. His research interests include continuous GPS, radar interferometry, and structural deformation monitoring. He is the cochair of the IAG Sub-Commission 4.4 “Applications of Airborne and Space-borne Imaging Systems.”
Eliathamby Ambikairajah is an associate professor and deputy head of the School of Electrical Engineering and Telecommunications, UNSW. He has been teaching and researching digital signal processing and applications since 1982 in the Athlone Institute of Technology, Ireland, and subsequently at UNSW. He has been active in a number of research areas, including speech and audio coding and speech enhancement, neural networks and pattern recognition, and cochlear modeling.
Yukio Tamura is a professor of the Tokyo Polytechnic University, and the director of the 21st Century Center of Excellence (COE) Program named “Wind Effects of Buildings and Urban Environment (WEBUE)” authorized by the Ministry of Education, Culture, Sports, Science and Technology of Japan. He currently holds several responsible positions in the wind engineering society, including president of the Japan Association for Wind Engineering, chairman of the Committee on Wind Loading of the Architectural Institute of Japan, and convener of WG-E (Dynamic Responses) of the Codification Working Group of the International Association for Wind Engineering.
Akihito Yoshida is a research associate of the Tokyo Polytechnic University. He obtained a bachelor of architectural engineering and a master of architectural engineering, both from the Tokyo Polytechnic University. Yoshida has been researching wind engineering and full-scale monitoring using GPS-RTK. He is also a researcher of the WEBUE under the 21st Century Center of Excellence (COE) Program.
Copyright © 2017 Gibbons Media & Research LLC, all rights reserved.