Académique Documents
Professionnel Documents
Culture Documents
www.elsevier.com/locate/compstruc
Abstract
A beam element formulation and solution procedure for progressive collapse analysis of planar frame structures is
presented. Unlike previous research, the current study addresses the signicance of dynamic load redistribution fol-
lowing the failure of one or more elements. The developed beam-column element utilizes a multi-linear, lumped
plasticity model, and it also accounts for the interaction of axial force and bending moment. Strength and stiness
degradation are included through use of a damage-dependent constitutive relationship. A damage index is used to
determine the onset of member failure. Following the failure of an element, the analysis continues in an ecient manner
through use of a modied member stiness procedure. This approach does not require the introduction of any addi-
tional degrees-of-freedom or modication of the element connectivity denitions. Finally, a methodology for updating
the state of a structure at the time of member failure is presented. Analysis results indicate that dynamic redistribution
of loads is a signicant feature of the progressive collapse problem and should be accounted for in order to avoid
estimates of capacity that are not conservative.
2004 Elsevier Ltd. All rights reserved.
Keywords: Progressive collapse; Structural dynamics; Nonlinear analysis; Building collapse; Damage; Software
0045-7949/$ - see front matter 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compstruc.2003.12.001
640 G. Kaewkulchai, E.B. Williamson / Computers and Structures 82 (2004) 639651
from the structural model for the purposes of analysis. ments and axial force along the length of the element.
The remaining structure is then analyzed to determine if In comparison, a stiness-based element formulation
other member failures result. Because progressive col- would depend upon displacement interpolation func-
lapse is an extreme event with a low likelihood of tions that provide the necessary compatibility require-
occurrence, the design limit state is taken to be the ments. The eect of axial force on yield moment is also
prevention of widespread failure propagation. Accord- incorporated in the element formulation through a
ingly, unfactored loads are used, and strength reduction momentaxial force interaction relationship. To capture
factors are ignored. The procedure continues until there cyclic behavior, multi-linear forcedeformation rela-
are no further member failures or the structure remains tionships, as well as the modied Mrozs hardening rule,
capable of supporting its loads despite the loss of vari- are employed. For the current research, the original
ous structural components. As a result of using the beam-column element was modied to account for large
Alternate Load Path Method for progressive collapse displacements through introduction of a geometric sti-
analysis, information on static load redistribution for ness matrix. Thus, both nonlinear geometry (i.e., P D
the structure under consideration is obtained. One crit- eect) and nonlinear material behavior are considered.
icism of this method is that it fails to consider dynamic
eects that inevitably result following the failure of one 2.2. Damage model
or more load carrying members.
Pretlove et al. [8] discussed the importance of Many materials, including steel and concrete, expe-
dynamic load redistribution in their research on the rience strain softening under cyclic loading as a result of
progressive failure of a tension spoke wheel. These damage. The original beam-column element of Kim [12]
researchers demonstrated that a static analysis predict- incorporated these eects, in an empirical fashion, in the
ing a damaged structure to be safe from progressive formulation of the constitutive model. Thus, damage
failure may not be conservative if inertial eects are and inelasticity eects were coupled in the original
taken into consideration. The authors (Kaewkulchai and development. Various researchers, however, have pro-
Williamson [9]) also demonstrated the importance of posed that these eects be treated independently, and
considering inertial eects for frame structures through a observations of the response of various structural sys-
simple frame example. Although dynamic eects on the tems lend credibility to this approach. For example, it is
response of truss structures during progressive failure possible for a brittle system to sustain severe damage yet
have been presented in the research literature (Malla and show little plasticity. For this reason, the beam-column
Nalluri [10,11]), few researchers have considered dy- element of Kim was modied for the current research so
namic load redistribution in the progressive collapse that inelasticity and damage are considered separately.
analysis of frame structures. In this paper, we present Therefore, to track the evolution of damage, a model
the development of a beam-column element for use in representing its evolution is required.
progressive collapse analysis of 2-D frames. For this Several damage models have been proposed in the
research, inertial eects play a key role, and, thus, a literature that depend upon a damage index, D, having a
solution methodology for tracking the dynamic response value ranging from 0 (no damage) to 1 (total damage).
of a structure modeled with the developed beam-column Examples include those of Park and Ang [13] and Rao
element is presented. The procedure is computationally et al. [14] for concrete members, and models by Kra-
ecient and has the capability of determining the re- winkler and Zohrei [15], Ballio and Castiglioni [16], and
sponse of a frame in which members can fail at any time Azevedo and Calado [17] for steel members. For the
throughout the response history. current research, we propose a formulation that depends
linearly upon the maximum deformation and the accu-
mulated plastic energy. To wit, damage, D, is given by
2. Beam element formulation the formula
The beam-column element originally developed by where a, b are constant (material) parameters, U d is a
Kim [12] for the analysis of steel moment frames sub- function that depends upon the maximum deformation,
jected to earthquake excitation forms the foundation of and W d is a function that depends upon the accumu-
the element that was developed for the current study. lated plastic energy.
The element employs a lumped plasticity model in which By varying the values of a and b, one can represent
inelasticity is assumed to occur only at the element ends dierent rates of damage accumulation so that many of
or hinges. In addition, the element utilizes a exibility- the models presented in the literature can be represented
based formulation which relies on force interpolation (Williamson and Hjelmstad [18]). Because the damage-
functions that satisfy the equilibrium of bending mo- dependent response of frame members during a collapse
G. Kaewkulchai, E.B. Williamson / Computers and Structures 82 (2004) 639651 641
2 Da Fe 3
event is not known, the current model provides the 11
0 0
1Da
exibility to investigate the relative importance of the 6 Di Fe22 7
Fd 6
4 0 1Di
0 7
5 5
terms in the damage model. Thus, until data are avail- Dj Fe33
able to suggest a model of damage that may be better 0 0 1Dj
suited to collapse analysis than the current model, the
choice was made to use a simple model that could be where Di is the damage index at hinge i, Dj is the damage
varied so as to investigate how the evolution of damage index at hinge j, and Da is an axial damage index. The
aects the computed results. Therefore, unlike previous axial damage index Da is chosen as the larger value of Di
researchers, we allow a and b to vary as a function of the and Dj because we assume that, at the onset of failure,
properties of the structural system. both the axial and rotational stiness at the failed end go
The use of a damage index to indicate failure is to zero, thereby implying separation of the failed
preferable to an arbitrary rule that may be based on member end from the remainder of the intact structure.
some percentage reduction of strength or stiness. When Di or Dj is nonzero, the member stiness is
Extending the basic model to account for both axial and modied. As discussed below, the damage indices are
rotational components, a modied damage index at each updated at each time step throughout the analysis pro-
hinge can be expressed as cedure so that the development of damage aects the
subsequent response as a function of time. Traditionally,
hmi dma hmi dma damage indices are viewed as simply a quantication of
Di ai
hy dy hy dy a design limit state. If the damage index exceeds a cer-
iP a P i a P P tain value, then one concludes that a member or struc-
Epi Epa Epi Epa
bi 2 ture is no longer able to function within the design
Eoi Eoa Eoi Eoa
specications. Under these circumstances, damage plays
where hm , hy are the maximum and the yield rotations a passive roledamage is recorded, but it does not aect
respectively, dma , dya are the maximum and the yield the evolution of the dynamic response. In this context
axial displacements, respectively, and Eo is the initial there is no need for more than an indirect connection
elastic energy prior to yield. The rst two terms within between physical damage and the damage index. In the
each set of the parentheses in Eq. (2) represent a basic present study, however, damage modies the structural
extension of the traditional Park and Ang [13] model in properties and hence contributes to the evolution of the
which damage is assumed to vary linearly as a function response. While this concept is not new (see, for exam-
of maximum deformation and hysteretic energy dissi- ple, work by Baber and Wen [21]), it is one that has not
pated. The last term within each set of parentheses ac- been used directly to evaluate structural performance.
counts for coupling between axial and exural behavior Because we incorporate the damage index in the con-
that is consistent with the constitutive model describing stitutive denition of the model, the stiness of the beam
the behavior of the plastic hinges. While other forms of member degrades as damage evolves. Note that for Di or
the damage index can be developed, the model adopted Dj 1, hinge i or j will have innite exibility (i.e., zero
for the present study oers a simple, yet eective way to stiness), and that hinge can be considered as discon-
account for the degradation of structural properties. nected or no longer attached to the structure. Thus,
Moreover, the parameters a and b can be calibrated with failure is assumed to take place at the member end in
experimental data of impulsively loaded beams to rep- which the damage index has achieved a value of one.
resent variable rates of damage accumulation. In addition to stiness degradation, the eect of
To incorporate the eects of stiness and strength strength degradation of the member is captured using
degradation at member hinges using the damage index, a the modied yield function as shown in Eq. (6).
simplied method suggested by Cipollina et al. [19] and
f
My 6
Inglessis et al. [20] is implemented. With their approach, 1 D
Fig. 1. (a) W21 57 cantilever beam with a suddenly applied point load P . (b) Tip displacement response history with increasing values
of damage (a, b).
Moment (k-in)
sponse parameters, ranging from zero damage (i.e., Damage Rate
a b 0), to severe damage (a b 0:09 for this
example). As might be expected, the response of the 4000
beam for this load case depends strongly upon the value
of a but weakly upon b. For cyclic loading, however, b
plays an important role (Williamson [22]). Because the 2000
0.00 0.01 0.02
load scenario prior to the initiation of a progressive Rotation, (rad)
collapse event can vary widely, the choice was made to
allow the damage model to depend upon the accumu- Fig. 2. Momentrotation (Mh) relationships for dierent rates
lated plastic energy. While progressive collapse can be of damage (a, b).
initiated by blast or impact, some researchers have
proposed that, in the context of retrotting older
structures in areas of high seismicity, it may also occur
as the result of an earthquake (Moehle et al. [23]). In 5.0
Tip deflection (in)
EI EI
k1 k1 and k2 k2 9
0 0 L L
-2 0 2 -2 0 2
the governing dierential equation can be solved by
-80 -80 successive integration to obtain the following equations:
(c) Deflection, (in) (d) Deflection, (in)
x4 x3 x2
EIwx q0 C1 C2 C3 x 10a
Fig. 4. Response of the damaged cantilever beam with dierent 24 6 2
rates of damage (a, b). (a) No damage (a b 0), (b) slow
damage (a b 0:01), (c) moderate damage (a b 0:03), q0 L q0 L 12 6k1 2k2 k1 k2
(d) severe damage (a b 0:1). C1 10b
4 4 12 4k1 4k2 k1 k2
3. Solution procedure
Fig. 6. Element degrees of freedom for complete and essential
In this section, a detailed description of the meth- sets.
odology used to develop the computer program for
progressive collapse analysis of planar frame structures
deformations (r1 r3 ) are needed. Other forcedeforma-
is given.
tion variables can be used; the critical requirement is
that the system under consideration be free of rigid body
3.1. Solution method motion. For the beam element shown in Fig. 6, the
relative element forces associated with the essential set
The solution methodology incorporates an implicit are denoted by Q1 , Q2 , and Q3 .
direct integration scheme, the Newmark-beta method From linear geometry theory, the relationship be-
(Newmark [25]), for solving the governing equations of tween member displacements and member deformations
equilibrium described by Eq. (13). can be determined from the following expression
MU 00 CU 0 RInt RExt 13 rAu 14
In Eq. (13), U, U 0 and U 00 are the displacement, velocity where A the kinematic transformation matrix
and acceleration vectors, respectively, M and C are the 2 3
1 0 0 1 0 0
system mass and damping matrices, respectively, and
4 0 1=L 1 0 1=L 0 5
RInt and RExt are the internal and external force vectors.
0 1=L 0 0 1=L 1
For every time step, the well-known NewtonRaphson
method is employed for solving the nonlinear system of Subsequently, the internal forces associated with the
equations. Details of the methodology can be found in complete set R can be obtained from the relative element
many references including Bathe [26], Belytschko et al. forces by
[27], etc.
R AT Q 15
3.2. Element state determination The following determination of the relative element
forces Q is with respect to an iteration step i in the
As described earlier, the beam-column element used
NewtonRaphson method during an incremental time
in the current study is a exibility-based element in
step of the Newmark-beta method.
which, during the determination of its internal forces,
equilibrium rather than compatibility is satised. The
1. Calculate the element deformation increments
determination of the element forces employs the current
Dri A Dui .
element deformations in conjunction with the element 2. Calculate the relative element force increments
stiness matrix obtained from the last converged itera-
DQi K i1t Dri .
tion step. More information on exibility-based ele-
3. Check if the current element forces Qi Qi1 DQi
ments can be found in the literature (e.g., Spacone et al. violate yield functions
[28], Neuenhofer and Filippou [29]). i(i) If there is no yielding, continue to Step 9.
Generally, a planar beam element consists of 6 de-
(ii) If yielding occurs, calculate a factor FAC (Fig.
grees of freedom (often referred to as the complete
7) for each yielded hinge as follows:
set) including two translations and one rotational dis- q
placement at each end (u1 u6 ), as shown in Fig. 6. In a My M2 Fy F 2
exibility-based formulation, however, rigid body mo- FAC q
tions need to be removed. Hence, the element response dM2 dF 2
can be described in terms of only three independent
degrees of freedom (referred to as the essential set). where dM and dF are the moment and axial force
Thus, only the axial deformation and two rotational increments at the yielded hinge.
G. Kaewkulchai, E.B. Williamson / Computers and Structures 82 (2004) 639651 645
Fig. 7. Yield surface and element force increments. Fig. 8. An element force increment with elasticperfectly plastic
material behavior. (a) Yielding during an incremental step, (b)
yielding at the end of an incremental step.
where subscripts c and r refer to contracted and Based on the discussion above, together with Eqs. (21),
released respectively. The contracted set consists of (23) and (26), the procedure for dynamic progressive
incremental force and displacement vectors corre- collapse analysis with the modied member stiness
sponding to the element degrees of freedom 1 through 3 approach only involves modication of the stiness
at the intact end. Likewise, the released set contains matrix and xed end forces of a failed member. Thus,
those for the element degrees of freedom 4 through 6 at Eq. (21) can be employed with the modied member
the released end. Because the released element force stiness matrix K cc and the modied xed-end force
vector DRr is zero, the released displacement vector Dur vector DRFc . These matrices correspond to the con-
can be written in terms of Duc as tracted degrees of freedom at the intact end, and they
can be determined form Eqs. (27) and (28).
Dur K rr
1 K rc Duc DRFr
20
K cc K eff cc K eff cr K 1
eff rr K eff rc
27
As a result, the incremental equilibrium equations for
the contracted set can be expressed by DRFc DReff Fc K eff cr K 1
eff rr DReff Fr
28
DRc K cc Duc DRFc 21 With K cc and DRFc , analysis after member failure can
where K cc K cc K cr K 1 continue with little modication to the main analysis
rr K rc
is the modied member
stiness matrix, and DRFc DRFc K cr K 1 routine because no new degrees of freedom are added to
rr DRFr
is the
modied xed-end force vector. the system. At the end of a converged time step, the
It is interesting to observe that the relationships de- released displacement vector Dur at the failed end of the
rived in Eqs. (20) and (21) are based on static equili- member can be obtained from the contracted displace-
brium of the element, and therefore do not apply for ment vector Duc at the intact end using Eq. (25). Thus,
dynamic analyses. However, if the Newmark-beta using the approach just outlined, the assembly process
method is employed, similar equations, which are valid for the stiness matrix and the applied force vector of
for dynamic analyses, can be developed because the the main structure does not change. In addition, the
relationship among displacement, velocity, and acceler- equation solver still determines the same number of
ation is assumed known. Thus, the governing equations unknown degrees of freedom. Accordingly, the ap-
of dynamic equilibrium can be cast in terms of unknown proach is computationally ecient.
displacements. Accordingly, inertial eects in the re- To verify the modied member stiness approach,
sponse are accounted for, and the procedure outlined results obtained from a dynamic analysis using this ap-
above can be used with only slight modication. proach are compared with those obtained from a con-
The incremental equations of motion for a dynamic ventional dynamic analysis. For this purpose, the system
system, when combined with the Newmark-beta method, shown in Fig. 10(a) is considered. The system is a xed
can be written as xed beam consisting of two elements and two degrees
of freedom as shown. A point load P of 20 kips acts at
DPeff K eff DU 22 Node 2. For static equilibrium, the system has shear and
moment diagrams as shown in Fig. 10(b). To illustrate
in which K eff and DPeff are expressed by
the modied member stiness method, the right end of
K eff A1 M A4 C K 23 member 2 is assumed to fail abruptly so that the support
is no longer available to resist loads. For a conventional
DPeff DP MA2 v A3 a
CA5 v A6 a
24 dynamic analysis, new degrees of freedom at Node 3
would need to be introduced. Hence, after failure at
where A1 A6 are the Newmark time integration con- Node 3, the structure can be analyzed by using a system
stants, v is the velocity vector, and a is the acceleration having four degrees of freedom with suddenly applied
vector. Because Eq. (22) is expressed in a form compa- forces S and M at Node 3 as shown in Fig. 11. For this
rable to Eq. (19), the released displacement vector Dur example, S 10 kips and M 300 k-in. For a dynamic
from Eq. (20) can be rewritten as
Dur K eff rr
1 K eff rc Duc DReff Fr
25
3.4. Updating
Generally, in nonlinear dynamic analyses using the Fig. 13. Equilibrium of a node, (a) before fracture, (b) after
incremental equations of motion, member stiness fracture.
-3.0 0
0.0 0.5 1.0
Dof 3 Mem2
-6.0 -500
(a) Time (s) (b) Time(s)
Fig. 12. Comparisons of computed results, (a) displacements for DOF 1 and 3, (b) bending moments of member 1 and 2 at Node 2.
G. Kaewkulchai, E.B. Williamson / Computers and Structures 82 (2004) 639651 649
Fig. 14. Two-bay, two-story frame, (a) uniform load of 0.4 kips/in, (b) 1st oor column failure due to an abnormal load.
650 G. Kaewkulchai, E.B. Williamson / Computers and Structures 82 (2004) 639651
0 0
(a) (b)
Displacement (in)
Displacement (in)
0 1.5 0 1.5
-3 -8
-6 -16
Time (s) Time (s)
Static analysis Dynamic analysis
Fig. 15. Vertical displacements at Node F , (a) elastic analysis, (b) inelastic analysis.
plex problem because the process is highly nonlinear, [11] Malla RB, Nalluri B. Dynamic nonlinear member failure
and it involves simultaneously the issues of member propagation in truss structures. Struct Eng Mech 2000;
instability, damage evolution, ruptures of member 9(2):11126.
joints, and impact forces of failed members. Most of [12] Kim KD. Development of analytical models for earth-
quake analysis of steel moment frames. PhD Dissertation,
these characteristics have not been conclusively identi-
The University of Texas at Austin, 1995.
ed in the literature, and little data exist to provide [13] Park YJ, Ang AHS. Mechanistic seismic damage model for
validation for computational models. Nevertheless, all reinforced concrete. J Struct Eng 1985;111(4):72239.
of these issues are considered to be of great importance [14] Rao PS, Sarma BS, Lakshmanan N, Stangenberg F.
to the current research and have been included in the Damage model for reinforced concrete elements under
developed computer program. Some of these features cyclic loading. ACI Mater J 1998;95(6):68290.
have been addressed in an approximate way, and some [15] Krawinkler H, Zohrei M. Cumulative damage in steel
have been addressed in a more detailed fashion. Future structures subjected to earthquake ground motions. Com-
research will focus on conducting parametric studies to put Struct 1983;16(14):53141.
identify key factors that contribute to the progressive [16] Ballio G, Castiglioni CA. Seismic behavior of steel
sections. J Construct Steel Res 1994;29:2154.
collapse of planar frame structures.
[17] Azevedo J, Calado L. Hysteretic behavior of steel mem-
bers: Analytical models and experimental tests. J Construct
Steel Res 1994;29:7194.
[18] Williamson EB, Hjelmstad KD. Nonlinear dynamics of a
References harmonically excited inelastic inverted pendulum. J Eng
Mech 2001;127(1):527.
[1] Breen JE. Research workshop on progressive collapse of [19] Cipollina A, Lopez-Inojosa A, Florez-Lopez J. A simpli-
building structures: Summary report. HUD-PDR-182, ed damage mechanics approach to nonlinear analysis of
Department of Housing and Urban Development, 1976. frames. Comput Struct 1995;54(6):111326.
[2] Leyendecker EV, Ellingwood BR. Design methods for [20] Inglessis P et al. Model of damage for steel frame members.
reducing the risk of progressive collapse in buildings: NBS Eng Struct 1999;21(10):95464.
Building Science Series 98, 1977. [21] Baber TT, Wen YK. Random vibration hysteretic, degrad-
[3] Hakuno M, Meguro K. Simulation of concrete-frame ing systems. J Eng Mech 1981;107(6):106987.
collapse due to dynamic loading. J Eng Mech 1993;119(9): [22] Williamson EB. Evaluation of damage and P D eects for
170923. systems under earthquake excitation. J Struct Eng-ASCE
[4] Isobe D, Toi Y. Analysis of structurally discontinuous 2003;129(3):103646.
reinforced concrete building frames using the ASI tech- [23] Moehle JP, Elwood KJ, Sezen H. Gravity load collapse of
nique. Comput Struct 2000;76(4):47181. building frames during earthquakes. Special Publication
[5] International Building Code. USA: International Code (SP-197) Uzumeri Symposium, American Concrete Insti-
Council, 2000. tute, Farmington Hills, MI, 2002.
[6] DoD Interim Antiterrorism/Force Protection Construction [24] Felton LP, Nelson RB. Matrix structural analysis. New
Standards Progressive Collapse Design Guidance: Depart- York: Wiley; 1997.
ment of Defense, 2001. [25] Newmark NM. A method of computation for structural
[7] Progressive Collapse Analysis and Design Guidelines for dynamics. J Eng Mech 1959;85:6794.
New Federal Oce Buildings and Major Modernization [26] Bathe KJ. Finite element procedures. New Jersey: Prentice-
Projects: Central Oce of the GSA, 2000. Hall; 1995.
[8] Pretlove AJ, Ramsden M, Atkins AG. Dynamic eects in [27] Belytschko T, Liu WK, Moran B. Finite elements for
progressive failure of structures. Int J Impact Eng nonlinear continua and structures. New York: Wiley;
1991;11(4):53946. 2000.
[9] Kaewkulchai G, Williamson EB. Dynamic progressive [28] Spacone E, Ciampi V, Filippou FC. Mixed formulation of
collapse of frame structures. The 15th Engineering nonlinear beam nite element. Comput Struct 1996;58(1):
Mechanics Division Conference, ASCE, New York, 2002. 7183.
[10] Malla RB, Nalluri B. Dynamic eects of member failure on [29] Neuenhofer A, Filippou FC. Evaluation of nonlinear
response of truss-type space structures. J Spacecraft Rock- frame nite-element models. J Struct Eng 1997;123(7):
ets 1995;32(3):54551. 95866.