OVERVIEW

Research Objectives:

The ultimate research objective is to assess structural performance during earthquakes using embedded networked sensing technology. We aim to address this fundamental need with a research program that comprises of: (1) gathering data to allow improvements in our ability to model structural responses to earthquakes; (2) studying system interactions, such as soil-structure interaction or non-structural/structural interaction, which are difficult to study in the laboratory; and (3) developing algorithms for damage detection and ultimately rapid post-earthquake assessment.

APPROACHES AND PROGRESS

1. Laboratory studies for system identification:
Laboratory studies using model-scale structures to assess the effectiveness of several existing system identification techniques. The model-scale structure consisted of a two bay, four story aluminum frame with variable stiffness properties. The structure was excited with both harmonic and realistic earthquake ground motions using a shaking table. Two different time domain system identification techniques, the ARX (Auto-Regressive model with Exogenous input) approach and N4SID (Numerical Algorithm for Subspace State Space System Identification), have been successfully employed to identify the structural modal properties such as frequencies, damping ratios, mode shapes. Physical parameters, especially the stiffness matrix of the structure, are further identified. Structural system identification results based on the experimental data have been compared with those by computed using Finite Element Analysis (SAP2000).

2. Factor building:
Collaborating with Earth and Space Science group (Paul, Allen, Igor), several earthquake and ambient vibration data from the Factor building have been analyzed. N4SID, which is probably the most advanced method for structural system identification when only output measurements are available, is used to accurately identify the modal properties such as frequencies, damping ratios and mode shapes of the building. The structural drawings of the building are now available which can be used for the modeling the building (structural model for dynamic responses).

3. Four Seasons building test:
A series of forced vibration tests on a four story reinforced concrete office building, herein termed the Four Seasons project, using the nees@UCLA equipment have been scheduled. Instrumentation that will deployed for this project includes the broadband force-balanced Episensor accelerometers, LVDTs and strain gauges to monitor displacements and subsurface accelerometers. The lease agreement for the testing of the building was recently executed, and we have obtained access to the building. The data archived through the tests could form the basis of detailed analytical studies for many years. Following the forced vibration testing of the Four Season building, the ARX and N4SID methods will be employed to identify the modal and physical properties of the building and the results are compared with those from finite element analyses.

FUTURE GOALS AND OBJECTIVES

1. Enhance laboratory studies by performing model improvement (periods, damping, mode shapes, structural stiffness). The collected dataset will serve as a benchmark dataset which will enable calibration of new system identification techniques.
2. Develop damage detection algorithms and sensor criteria requirements using the Four Seasons building. The Four Seasons project provides a unique opportunity to perform baseline field tests on a full-scale building, which can be supplemented with a series of tests with modified structural properties. The resulting dataset will enable a comparison of various parameters pre- and post-damage, against which future damage detection algorithms can be calibrated.
3. Implement and evaluate innovative network time protocols for time synchronization in wireless networked sensors in a building environment.
4. Implement and evaluate the performance of multi-hop and wireless data transfer using next-generation wireless MEMS sensors. These sensors will be co-located with traditional seismic sensors, enabling a direct evaluation of their performance.
   

PEOPLE

FACULTY

John W. Wallace
Ertugrul Taciroglu

STAFF

Daniel H. Whang
Ying Lei

GRADUATE STUDENTS

Derek Skolnik
William Elmer