10NCEE
10NCEE
10NCEE
net/publication/290557911
CITATIONS READS
3 596
2 authors:
All content following this page was uploaded by Panon Latcharote on 17 October 2017.
ABSTRACT
In a target area, thousands of buildings are modeled to predict individual and overall damage from
earthquake and tsunami. All buildings are represented as a different set of structural element
models, such as beam, column, and wall, to perform nonlinear structural analysis. For each
building, a structural frame is generated from Geographic Information System (GIS) data, such as
building shapes, building types, number of floors, and construction year. Based on a building
design code, strength capacity of structural element models is estimated to predict structural
damage of each building. This paper proposes nonlinear structural analysis of RC buildings
suffering damage from earthquake and subsequent tsunami. In the case of earthquake, dynamic
structural response analysis is performed to predict damage of RC buildings, which are subject to
earthquake ground motions. Sequentially, static structural analysis is performed to predict more
damage to RC buildings, which are subject to hydrodynamic force from tsunami. For earthquake
and subsequent tsunami, nonlinear structural analysis of RC buildings is performed by means of
the same hysteresis models.
1
Graduate Student, Dept. of Infrastructure System Engineering, Kochi University of Technology, Japan, M. Eng.
2
Professor, Dept. of Infrastructure System Engineering, Kochi University of Technology, Japan, Dr. Eng.
Latcharote P, Kai Y. Nonlinear structural analysis of buildings suffering damage from earthquake and subsequent
tsunami. Proceedings of the 10th National Conference in Earthquake Engineering, Earthquake Engineering Research
Institute, Anchorage, AK, 2014.
Tenth U.S. National Conference on Earthquake Engineering
Frontiers of Earthquake Engineering
July 21-25, 2014
10NCEE Anchorage, Alaska
ABSTRACT
In a target area, thousands of buildings are modeled to predict individual and overall damage from
earthquake and tsunami. All buildings are represented as a different set of structural element models,
such as beam, column, and wall, to perform nonlinear structural analysis. For each building, a
structural frame is generated from Geographic Information System (GIS) data, such as building
shapes, building types, number of floors, and construction year. Based on a building design code,
strength capacity of structural element models is estimated to predict structural damage of each
building. This paper proposes nonlinear structural analysis of RC buildings, suffering damage from
earthquake and subsequent tsunami. In the case of earthquake, dynamic structural response analysis
is performed to predict damage of RC buildings, which are subject to earthquake ground motion.
Sequentially, static structural analysis is performed to predict more damage to RC buildings, which
are subject to hydrodynamic force from tsunami. For earthquake and subsequent tsunami, nonlinear
structural analysis of RC buildings is performed by means of the same hysteresis models.
Introduction
In order to reduce the social and economic impact of earthquake and tsunami, earthquake and
tsunami simulation can be used for predicting and illustrating structural damage of buildings. In
earthquake and tsunami simulation, thousands of buildings are modeled to predict structural
damage of each individual building and overall damage in a target area. Integrated Earthquake
Simulation (IES) is an earthquake simulation tool for predicting and illustrating structural damage
of all buildings simultaneously [1] as shown in Fig. 1. For tsunami simulation, all buildings can be
modeled by means of the same approach as IES to determine hydrodynamic force, which may vary
upon building arrangements and shapes. Object-Based Structural Analysis (OBASAN) is a
structural analysis program which can perform static structural analysis and dynamic structural
response analysis of buildings [6]. In earthquake and tsunami simulation, RC buildings are
represented as a structural frame, and OBASAN can be used to perform reliable nonlinear
structural analysis, which has a significant effect on reliable prediction of structural damage from
earthquake and tsunami. Due to earthquake ground motion and hydrodynamic force in earthquake
and tsunami simulation, structural damage of each building can be predicted from structural
analysis results and the estimated strength capacity of structural elements. The earthquake and
1
Graduate Student, Dept. of Infrastructure Systems Engineering, Kochi University of Technology, Japan, M. Eng.
2
Professor, Dept. of Infrastructure Systems Engineering, Kochi University of Technology, Japan, Dr. Eng.
Latcharote P, Kai Y. Nonlinear structural analysis of buildings suffering damage from earthquake and subsequent
tsunami. Proceedings of the 10th National Conference in Earthquake Engineering, Earthquake Engineering Research
Institute, Anchorage, AK, 2014.
tsunami simulation output can lead to prevention measures and raise awareness of disaster
prevention among people in the target area. This paper aims to focus on nonlinear structural
analysis of RC buildings, which can be a designated evacuation shelter for earthquake and tsunami.
Drift angle
0.0 0.02 0.04 0.06 0.08 0.1
In earthquake and tsunami simulation, thousands of buildings are modeled to predict structural
damage of each individual building and overall damage in a target area. All buildings are
represented as a structural frame which is a set of structural element models, such as beam, column,
and wall, to perform nonlinear structural analysis. Part of IES, Common Modeling Data (CMD)
is a building modeling approach to generate a structural frame of each building for earthquake
simulation [1] as shown in Fig. 2. For tsunami simulation, nonlinear structural analysis of buildings
can be performed by means of the same structural frame as earthquake simulation.
Quantum GIS (QGIS) is an open software which can collect Geographic Information
System (GIS) data, such as building dimensions, building types, roof types, building categories,
floor area, number of floors, and construction year as shown in Fig 3. With GIS data, each building
is modeled as a structural frame from common span length and floor height in a building design
code. For each floor of a building, weight and mass are estimated from building types (e.g., RC,
steel, and wooden building), roof types (e.g., tile, cement, and flat roof), and building categories
(e.g., house, apartment, hotel, hospital, factory, and school). For the structural frame of each
building, foundation modeling is a fixed type or a pinned type, and damping ratio is estimated
from building type and construction year. In order to obtain reliable prediction of structural
damage, building modeling is very important because it has a significant effect on dynamic
structural response analysis and static structural analysis of buildings due to earthquake and
tsunami.
Coordinate
Address
Roof types
Construction year
Building types
Floor area
Building categories
Number of floors
Bending moment
My
Mc
For each structural element of a RC building, the bending moment-curvature, shear stress-
strain relationship, and hysteresis models are constructed to perform nonlinear structural analysis.
For the bending moment-curvature and shear stress-strain relationship, strength capacity of each
structural element is estimated based on a building design code with span length, weight, and
material properties. The reduced strength capacity is cnsidered from construction year of a RC
building. Based on the building design code, building categories have a significant effect on the
importance of earthquake resistance design, so strength capacity of each structural element is
varied by means of private and public buildings, such as houses, apartments, hotels, hospitals,
factories, and schools. From the strength capacity of each structural element, section properties are
calculated to construct the bending moment-curvature and shear stress-strain relationship. In order
to perform nonlinear structural analysis of a RC building, different hysteresis models of bending,
shear, and axial load are developed for each structural element, such as beam, column, and wall.
For the hysteresis models, some parameters are defined to construct a hysteresis loop in order to
represent the restoring force characteristics of each structural element. With the bending moment-
curvature relationship, the shear stress-strain relationship, and hysteresis models, nonlinear
structural analysis of each RC building is performed to predict structural damage from earthquake
and subsequent tsunami.
In earthquake scenarios, an earthquake occurs beneath a city area. Then earthquake ground motion
shakes all buildings and causes some structural damage to all buildings in the city area.
Subsequently, a tsunami reaches the city area and hydrodynamic force causes more structural
damage to the buildings. With building modeling, each RC building is represented as a set of
structural element models, such as beam, column, and wall, which can be analyzed by inputting
sequential excitation loads of earthquake and tsunami as shown in Fig. 5. In the case of
earthquake, dynamic structural response analysis is performed to predict structural damage of RC
buildings, which are subject to earthquake ground motions. Sequentially, static structural analysis
is performed to predict more damage to RC buildings, which are subject to hydrodynamic force
from a tsunami. In the future, structural analysis of buildings should consider liquefaction,
buoyancy force, and debris due to a tsunami.
Figure 5. Sequential excitation loads of earthquake ground motion and hydrodynamic force
As mentioned above, each RC building is represented as a set of structural element models which
can be analyzed by inputting sequential excitation loads of earthquake and tsunami. In earthquake
and tsunami simulation, earthquake ground motion causes structural damage and then
hydrodynamic force causes more structural damage to RC buildings in a target area. For nonlinear
structural analysis of RC buildings, dynamic structural response analysis and static structural
analysis can be performed due to earthquake and tsunami respectively.
Fy
Fc
B A C
Damage Prediction
Due to nonlinear structural analysis, structural damage can be predicted from structural analysis
results to consider individual and overall damage of RC buildings in earthquake and tsunami
simulation. In order to consider structural damage of a RC building, structural analysis results are
classified by node output, element output, modal output, and hysteresis output as shown in Table
1. Node output is structural analysis results of each node, such as displacement, velocity,
acceleration, and reaction force; element output is structural analysis results of each structural
element, such as stress, strain, deformation, and internal force (e.g., bending moment, shear force,
and axial force); modal output is dynamic characteristics of a RC building, such as eigenvalue,
eigenvector, period, and frequency; hysteresis output is nonlinear analysis results of each structural
element, such as ductility factor and stiffness degrading factor.
Table 1. Structural analysis results of a RC building
For each story of a RC building, story displacement is calculated from node output, and
drift angle is calculated from relative story displacement and floor height. Based on a building
design code, maximum drift angle can classify the level of structural damage approximately. For
each structural element of a RC building, bending moment, deformation, stress, and strain are
calculated from element output, and structural damage can be classified from the range of the
bending moment-curvature and shear stress-strain relationship, especially for RC walls. For all RC
buildings in a target area, the classification level of structural damage from earthquake and tsunami
is very useful for designated evacuation shelters.
Nonlinear structural analysis of a 7-story RC building, which are subject to earthquake ground
motion and subsequent hydrostatic force, is conducted in this section. As shown in Fig. 7, the 7-
story RC building is represented as a wall-frame model, including beam, column, and shear wall,
from GIS data and building modeling. For each structural element, strength capacity and section
properties are estimated based on a Japanese building design code. This typical building is the first
attempt to perform nonlinear structural analysis of RC buildings suffering damage from earthquake
and subsequent tsunami.
Number of floors = 7
Floor height = 3.5 m
Number of spans (major) = 3
Number of spans (minor) = 2
Span length = 6 m
X Y Z
800
th
600 5
400
th
200 4
Acceleration (gal)
qz= ρg(αh - z)
0
0 1 2 3 4 5 6 7 8 9 10 rd α = 2.0
-200
3
-400 nd
2
-600
-800 st
1
-1000
Time (sec)
From results of nonlinear structural analysis, the stress-strain relationships of columns are
shown in Fig 9. Three columns at the first floor are selected to observe structural behavior due to
earthquake and subsequent tsunami. Inada model [9] is used to represent hysteretic behavior of
these columns in nonlinear structural analysis. As can be seen in Fig. 9, the red line represents
nonlinear structural analysis of this building due to earthquake and the blue line represents
nonlinear structural analysis of this building due to tsunami. Due to earthquake, nonlinear
structural analysis is performed during earthquake ground motion and also free vibration.
Sequentially, nonlinear structural analysis is performed due to tsunami. Therefore, the stress-strain
relationships of columns are continuous as shown in Fig. 9 and are quite similar to the idealization
as shown in Fig. 2. It tends to get better results from more realistic behavior due to hydrodynamic
force. Then, structural damage can be predicted from this sequential analysis. The ratio between
stress and strain in these results are the normalized stiffness of these columns which is not over
than 1.0. From these results, structural damage of columns can be identified to elastic, functional,
or safety limit.
For earthquake and tsunami simulation, this application of nonlinear structural analysis can
be conducted to investigate structural damage of all RC buildings in a target area.
A column at the corner of the first floor
5.00E-04
4.00E-04
3.00E-04
2.00E-04
Stress
1.00E-04
0.00E+00
-8.00E-04 -6.00E-04 -4.00E-04 -2.00E-04 0.00E+00 2.00E-04 4.00E-04 6.00E-04 8.00E-04 1.00E-03
-1.00E-04
-2.00E-04
-3.00E-04
-4.00E-04
Strain
3.00E-04
2.00E-04
1.00E-04
Stress
0.00E+00
-8.00E-04 -6.00E-04 -4.00E-04 -2.00E-04 0.00E+00 2.00E-04 4.00E-04 6.00E-04 8.00E-04
-1.00E-04
-2.00E-04
-3.00E-04
-4.00E-04
-5.00E-04
Strain
4.00E-04
3.00E-04
2.00E-04
Stress
1.00E-04
0.00E+00
-4.00E-04 -2.00E-04 0.00E+00 2.00E-04 4.00E-04 6.00E-04 8.00E-04
-1.00E-04
-2.00E-04
-3.00E-04
Strain
Figure 9. The stress-strain relationship of columns due to earthquake and subsequent tsunami
Conclusions
In earthquake and tsunami simulation, thousands of RC buildings are modeled as a different set of
structural element models, such as beam, column, and wall, to perform nonlinear structural
analysis. Many parameters of each RC building are estimated based on GIS data and a building
design code. Due to earthquake and subsequent tsunami, nonlinear structural analysis of a RC
building can be performed sequentially by means of the same bending moment-curvature
relationship, shear stress-strain relationship, and hysteresis models. With results of nonlinear
structural analysis, structural damage can be predicted to consider individual damage and overall
damage in a target area. The classification level of structural damage, such as elastic, functional,
and safety limit, is very useful for designated evacuation shelters from earthquake and tsunami. In
this paper, nonlinear structural analysis of an RC building is performed due to earthquake and
subsequent tsunami. However, In the future, nonlinear structural analysis of all RC buildings in a
target area due to earthquake and subsequent tsunami will be performed to predict overall damage
which is important for disaster planning and management.
Acknowledgments
The authors greatly appreciate Kochi University of Technology (KUT) and Japan Student Service
Organization (JASSO) for supporting scholarship. The authors also appreciate Prof. Hunter of
KUT for supporting about English writing. Furthermore, the authors thank every students in Kai’s
laboratory for help.
References
1. Hori M, Ichimura T. Current state of integrated earthquake simulation for earthquake hazard and disaster. Journal
of Seismology 2008; 12: 307-321.
2. Hori M. Introduction to Computational Earthquake Engineering. Imperial College Press: London 2nd Edition,
2011.
3. Ichimura T, Hori M, Terada K, Yamakawa T. On integrated earthquake simulator prototype: combination of
numerical simulation and geographical information system. Structural Engineering and Earthquake Engineering
JSCE 2005; 22(2): 233s-243s.
4. Ichimura T, Hori M. Development of prototype of integrated earthquake disaster simulator using digital city and
strong ground motion simulator with high-resolution. 13th World Conference on Earthquake Engineering 2004;
1418.
5. Latcharote P, Kai Y. High performance computing of dynamic structural response analysis for the integrated
earthquake simulation. Proceedings of East Asia-Pacific Conference on Structural Engineering and Construction
2013.
6. Latcharote P, Kai Y. Object oriented modeling of dynamic structural response analysis. Proceedings of Thailand
Japan International Academic Conference 2012; 71-72.
7. Takeda T, Sozen M A, Nielsen N N. Reinforced concrete response to simulated earthquakes. Journal of the
Structural Division ASCE 1970; 96(12): 2557-2573.
8. Kai Y. A Study of RC Space Shear Wall Structure for Nonlinear Analysis (in Japanese). Doctoral Dissertation
University of Tokyo 2001.
9. Inada, Y. Restoring force characteristics of reactor buildings based on load tests and numerical analysis (in
Japanese), Part 1-3, Transactions of AIJ 1987; 371: 61-71, 1987; 378: 16-26, 1987; 382: 19-29.