LABasin-SAF | PFRAME3D & FRAME3D | Doctoral | Masters | Presentations |
LABasin-SAF PROJECT
What would happen to engineered buildings in the Los Angeles and San Fernando basins if an 1857-like magnitude 7.9 earthquake were to occur on the San Andreas fault?
Two such earthquakes have been simulated and the nonlinear response of two 18-story steel moment-frame buildings has been modeled with the aid of high-performance computing.
Key information:
1. Two scenarios - one with rupture initiating at Parkfield in central California and propagating north-to-south a distance of 290 km and the other with the slip distribution and rupture direction flipped such that rupture starts north of Los Angeles and proceeds south-to-north terminating at Parkfield.
2. Both are magnitude 7.9 earthquakes. Kinematic source model of the Denali (Alaska) fault earthquake of 2002 has been mapped on to the San Andreas fault.
3. Domain of simulation includes a uniform grid of 636 analysis sites spaced at 3.5 km in each direction.
4. 3-D models of two 18-story steel moment-frame buildings (one existing building designed according to the 1982 Uniform Building Code, and the other being the same building redesigned according to the stricter 1997 UBC) have been analyzed subjected to 3-component ground motion from the two San Andreas fault earthquakes at each of the 636 sites.
5. Building response summed up in the form of maps of peak interstory drift ratio.
Watch movies of the seismic wave propagation and building response.
 |
 |
NORTH-TO-SOUTH RUPTURE SCENARIO
Fault rupture and seismic wave propagation:
Play Movie (18 MB)
Movies of the shaking of existing and new 18-story steel moment-frame
building models at various sites in Southern California. Of these, the
existing building models in Thousand Oaks and Northridge are seen collapsing,
while there is significant permanent tilt in most other cases.
Anaheim: Play Movie
(8 MB)
Baldwin Park: Play
Movie (8 MB)
Downtown Los Angeles:
Play Movie (8 MB)
Long Beach: Play Movie
(8 MB)
Northridge: Play Movie
(8 MB)
Santa Ana: Play Movie
(8 MB)
Thousand Oaks: Play Movie
(8 MB)
West Los Angeles: Play Movie
(8 MB)
SOUTH-TO-NORTH RUPTURE SCENARIO
Fault rupture and seismic wave propagation:
Play Movie (18 MB)
SOFTWARE USED
SPECFEM3D for the seismic wave propagation.
FRAME3D for the building analyses.
REFERENCES
"Performance of Two 18-Story Steel Moment-Frame Buildings in Southern California During Two Large Simulated San Andreas Earthquakes," Swaminathan Krishnan, Chen Ji, Dimitri Komatitsch, and Jeroen Tromp, vol. 22(4), November 2006, Earthquake Spectra.
"Case Studies of Damage to Tall Steel Moment-Frame Buildings in Southern California During Large San Andreas Earthquakes," Swaminathan Krishnan, Chen Ji, Dimitri Komatitsch, and Jeroen Tromp, vol. 96(4), August 2006, Bulletin of the Seismological Society of America.
"Performance of 18-Story Steel Moment-Frame Buildings During a Large San Andreas Earthquake - A Southern California-Wide End-to-End Simulation", Swaminathan Krishnan, Chen Ji, Dimitri Komatitsch, and Jeroen Tromp, Technical Report - CaltechEERL:EERL-2005-01, California Institute of Technology, Pasadena, California, 2005.
COLLABORATORS
Dr. Jeroen Tromp, Caltech
Dr. Chen Ji, University of California, Santa Barbara
Dr. Dimitri
Komatitsch, University of Pau, France
P-FRAME3D & FRAME3D
FRAME3D is a program for the three-dimensional nonlinear analysis of steel buildings. It aims to overcome the computational challenges posed by full 3D analysis of buildings subject to earthquake ground motion. The element library consists of a plastic hinge beam element, an elastofiber beam element, a panel zone element, a 4-noded diaphragm element to model floor slabs, and an elastic translational/rotational spring element to model foundations and supports. The program utilizes a Netwon-Raphson iteration strategy applied to an implicit Newmark time-integration scheme to solve the nonlinear equations of motion at each time-step. Geometric nonlinearity and shear deformation are included in the formulation.
P-FRAME3D is the parallel version of FRAME3D currently under development. Upon completion, P-FRAME3D can be executed on a parallel computer cluster to solve complex structural engineering problems such as the strong shaking of a super-highrise building or a long-span cable-supported bridge under a large earthquake occurring on a nearby fault.
Related Links:
FRAME3D Program - http://www.frame3d.caltech.edu.
FRAME3D Program User Guide - FRAME3D - A Program for Three-Dimensional Nonlinear Time-History Analysis of Steel Buildings: User Guide.
Doctoral
Strong ground motion from a nearby fault has frequency content in the same range as the natural frequencies of tall buildings. This may have serious repercussions and is the topic of my Ph.D. dissertation. Buildings are designed per building code standards. But, are the code provisions adequate? Strong motion from large earthquakes has been recorded only in recent times in the near-source region. Have the current codes used this information to update tall structure design guidelines? Considerable damage has been observed in tall buildings from the Northridge, Kobe, Turkey, and Taiwan earthquakes. How will tall buildings designed per the latest code regulations perform if they were to be shaken by any of these earthquakes? My thesis attempts to answer these questions.
Tall buildings by their nature are computationally intensive to analyze. They consist of thousands of degrees of freedom and when subjected to strong ground motion from a nearby source, exhibit inelastic response. Modeling this inelastic response requires an iterative approach that is computationally expensive. Furthermore, a large class of buildings, classified as irregular, exhibits complex behavior that can be studied only when the structures are modeled in their entirety. To this end, a three-dimensional analysis program, FRAME3D, has been developed incorporating two special beam-column elements -- the plastic hinge element and the elastofiber element that can model beams and columns in buildings accurately and efficiently, a beam-column joint element that can model inelastic joint deformation, and 4-noded elastic plane-stress elements to model floor slabs acting as diaphragms forcing the lateral force resisting frames in a building to act as one unit. The program is capable of performing time-history analyses of buildings in their entirety.
Six 19-story irregular steel moment frame buildings (with buildings 2A and 3A being variants of buildings 2 and 3, respectively) have been designed per the latest code (Uniform Building Code, 1997). Two of these buildings have reentrant corners and the other two have torsional irregularity. Their strength and ductility are assessed by performing pushover analyses on them. To assess their performance under strong shaking, FRAME3D models of these buildings are subjected to near-source strong motion records from the Iran earthquake (Mw = 7.3, Tabas Station) of 1978, the Northridge earthquake (Mw = 6.7, Sylmar Station) of 1994 and the Kobe earthquake (Mw = 6.9, Takatori Station) of 1995. None of the buildings collapsed under these strong events in the computer analyses. However, when compared against the acceptable limits for various performance levels in FEMA 356 document, the damage in terms of plastic deformation at the ends of beams and columns and at joints would render the buildings inadequate in terms of life safety in quite a few cases and would even indicate possible collapse in a couple of cases. Thus, in these terms, the code falls short of achieving its life safety objective, and the near-source factors introduced in the code in 1997 in recognition of the special features of near-source ground motion seem to be inadequate. The ductility demand, in terms of plastic rotation at the ends of beams and columns and in joints, on these buildings during this class of earthquakes is up to 6% of a radian, which is far greater than a typical limiting plastic rotation of 3% associated with fracture and consequent failure of large wide-flanged steel sections during experiments. Thus, if strength degradation due to fractures, local buckling, etc., were to be included in the analysis, then the results would likely to be worse, as far as the ability of these buildings to withstand these earthquakes without collapse is concerned.
Due to damage localization, the peak drifts observed in the structure far exceeded the inelastic drift limit in the code of 0.02 (in some cases up to 3 times). This points to serious non-structural damage to facades, interior dry wall, etc. Furthermore, large roof permanent offsets after the events indicate significant post-earthquake repair requiring considerable disruption and building closure. Column yielding was minimal thus validating the strong-column weak-beam criterion in the code. Redundancy factors used to assess the redundancy in the system need to take into account the case of torsionally sensitive structures where frames in both principal directions are simultaneously activated. Stress concentration was not observed at the reentrant corners in L-shaped buildings.
Finally, the data catalogued in my thesis could be useful for future code development and tall structure design guidelines.
Related Links:
FRAME3D Program - http://www.frame3d.caltech.edu.
FRAME3D Program User Guide - FRAME3D - A Program for Three-Dimensional Nonlinear Time-History Analysis of Steel Buildings: User Guide.
Building Animations - http://www.frame3d.caltech.edu/anim.html.
Key Analyses Results - http://www.frame3d.caltech.edu/bldgdb.html.
Masters
Structural system identification is the process of deducing the properties of a structural system from its measured response to ambient vibration by fitting a mathematical model. The objectives of my Masters research were to: (1) investigate the use of existing system identification methods, and (2) develop new system identification methods, in order to evaluate both loading and structural modal parameters of ambient excited structures for which the load process is difficult to measure. An example is the case of offshore platforms subjected to sea waves. The study considers the inverse problem from the viewpoint of continuous-time linear dynamic systems. An existing structural identification technique defined for the deterministic case (when the loading is known) and based on sequences of modal minimizations (called modal sweeps) is formulated for the general case of multiple-input multiple-output (MIMO) systems. A global minimization technique based upon the Levenberg-Marquardt algorithm for nonlinear least-squares problems is developed for the same case. Both identification techniques are applied to a multi-degree-of-freedom (MDOF) shear building model and the results are shown to be consistent. Using the framework of random vibration theory, these techniques are then converted to the stochastic case, namely when the loading process is known only statistically. These identification methods, both individually and combined, were tested based on simulated cases of increasing complexity. The results obtained are promising and indicate that under certain conditions, both load and structural parameters can be estimated from the measured response and the statistical properties of the loading process. However, the load parameters are not as well estimated as the structural parameters, since the load process is further removed from the response process than the structural filter. Although a generic shear building model has been used throughout this study to simulate the dynamic response of real structures, the results obtained are believed to apply to linear multi-degree-of-freedom systems in general.
Thesis: System Identification of Dynamic Structural Systems Using Continuous-Time Domain Methods.