0% found this document useful (0 votes)
52 views8 pages

IOEGC 2019 Summer 046

This document discusses dynamic increase factors used in progressive collapse analysis of structures. It analyzes reinforced concrete buildings of varying heights and bay sizes to determine how the dynamic increase factor varies with the actual level of inelasticity experienced. The effects of seismic design level on the dynamic increase factor are also examined.

Uploaded by

umesh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
52 views8 pages

IOEGC 2019 Summer 046

This document discusses dynamic increase factors used in progressive collapse analysis of structures. It analyzes reinforced concrete buildings of varying heights and bay sizes to determine how the dynamic increase factor varies with the actual level of inelasticity experienced. The effects of seismic design level on the dynamic increase factor are also examined.

Uploaded by

umesh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Proceedings of IOE Graduate Conference, 2019-Summer

Peer Reviewed
Year: 2019 Month: May Volume: 6
ISSN: 2350-8914 (Online), 2350-8906 (Print)

Dynamic Increase Factor for Progressive Collapse Analysis due


to Sudden Column Removal
Umesh Lamichhane a , Hari Darshan Shrestha b
a, b
Department of Civil Engineering, Pulchowk Campus, IOE, Tribhuvan University, Nepal
Corresponding Email: a [email protected], b [email protected]

Abstract
Assessment of progressive collapse potential of structures can be carried out through alternate path approach.
In the alternate path approach, load bearing structure is suddenly removed and the ability of other member to
withstand the added forces are examined. Linear static, nonlinear static and nonlinear dynamic analysis can
be employed for determination of the structural response during progressive collapse analysis. In the dynamic
analysis, the inertial forces are directly considered through the equations of motion. But in nonlinear static
method, a Dynamic Increase Factor (DIF) is introduced for modifications of dead and live loads due to the
unaccounted inertial effects. The Unified Facilities Criteria (UFC) and General Service Administration (GSA)
guidelines initially adopted Dynamic Increase Factor of two. Latter both guidelines use Dynamic Increase
Factor (DIF) based on plastic rotation capability that the damaged frame experiences. In this study, variation
of dynamic increase factor with actual level of inelasticity is found out. For this purpose, three to seven storey
symmetrical reinforced concrete building of varying bay sizes designed as per Indian Standard are analysed.
The effect of sesmic design level on Dynamic Increase Factor (DIF) is also found out.
Keywords
Progressive collapse, Dynamic Increase Factor, RC structure, Column removal, SAP2000

1. Introduction until the complete collapse of the structure.

After the collapse of the World Trade Center towers in The General Services Administration, “Progressive
2001, Progressive collapse has been an increasing Collapse Analysis and Design Guidelines”[1] and The
concern in the structural engineering community. Department of Defense Unified Facilities Criteria
Progressive collapse of existing building is initiated 4-023-03 “Design of Buildings to Resist Progressive
by the sudden failure of one or more of its major load Collapse”[2] are the available guidelines for the
bearing elements, typically columns or walls, progressive collapse analysis. Both the guidelines
followed by redistribution of the loads and failure of follow alternate load path method where the analysis
the next elements in the vicinity in a chain-like of structure is carried out to determine collapse
reaction until the failure of the whole building. The potential after the removal of load bearing elements.
cause of the phenomenon may be a result of one Different analytical procedures like linear static,
specific event or a combination of causes that lead to nonlinear static and nonlinear dynamic method are
local failure like vehicular impact, earthquakes, fire, employed for determination of the structural response
explosions as well as human error in design or during progressive collapse. Among these method,
construction of the structure. A typical example of nonlinear dynamic analysis procedure gives accurate
this would be the intentional removal of a column by and better results but is more complicated, tedious and
an explosion. The structural components of the floors time consuming[3]. Due to that reason simple, less
above this column would experience a sudden tedious static analysis is being carried out with certain
increase in stress as well as large deflections. This magnification factors that accounts for dynamic and
amplification of the load may continue to cause failure nonlinear effects so that the final responses are more
in other primary members of the structure until the similar and exact to dynamic analysis. Load Increase
building stabilizes with noticeable deformations or Factor (LIF) used in linear static method accounts
inertial and nonlinear effects whereas Dynamic

Pages: 343 – 349


Dynamic Increase Factor for Progressive Collapse Analysis due to Sudden Column Removal

Increase Factor (DIF) used in nonlinear static Secondary members aren’t included. The connection
procedure accounts only for inertial effects. at the foundation are modeled as fixed connection.
Initially the General Services Administration,
“Progressive Collapse Analysis and Design
Guidelines” and The Department of Defense Unified
Facilities Criteria 4-023-03 “Design of Buildings to
Resist Progressive Collapse” both adopted the
Dynamic Increase Factor of 2 on analysis[4, 5]. Latter
this value is modified in both guidelines and adopted
the same formula which is based on material
properties of affected structural members only. It isn’t
based on the actual plastic deformation level that the a
damaged frame experiences. It is possible that a frame
can still remain elastic even after a certain column has
been removed which is particularly true for the frames
that were originally designed to withstand large lateral
loads such as those from earthquakes and winds. Such
frame have significantly extra capacity against gravity
induced progressive collapse. Also even after the
damaged frame enters in inelastic range, the actual
level of inelasticity is not necessarily so high that the
plastic rotation of the controlling beam reaches
maximum allowable plastic hinge rotation. So the
variation of DIF with gravity loading and structural
capacity in terms of Mu/My is found out in this paper
which can be useful for DIF formulation.

2. Analytical Models
Figure 1: Three dimensional model of building
Three-dimensional reinforced concrete buildings are
modeled and analyzed using finite element program
SAP2000. The table 1 shows the parameters of All beams are confined by shear reinforcement
building taken in this study. adequately so that beams are not shear controlled and
hinges are governed by flexural stress only. The lump
Table 1: parameters of building nonlinearity is included by assigning plastic hinges at
the ends and midspan of every beam elements. The
Storey of buildings 3,4,5,6,7 generalized force displacement curve assigned for
Height of each floor 3m each hinge is shown in figure 2. Point A is always
Size of bay 3m,4m,5m,6m origin. B represents yielding, Point C represents the
ultimate capacity. Point D represents a residual
strength and Point E represents total failure. The
All together 20 symmetrical reinforced concrete
nonlinear modeling parameters a, b, c and acceptance
building are taken. The building are seismically
criteria are taken from Table 4-1 of UFC 4-023-03[2]
designed using Indian Standard code[6]. The beam
depending on the structural configuration, shear
and column sizes are different along with their
demand, and reinforcement ratios. The hinge
reinforcement details.
definition is designed to allow strain hardening of 5%
During the analysis, beams and columns are modeled at the point expected to be the maximum allowed
with two noded frame elements as shown in figure 1. rotation which is different from 10% hardening used
The beam and column connection are moment in ASCE 41[7]. Geometric nonlinearity such as
resistant. The columns are stronger than beam so that P-delta effect is also included in analysis. Columns
the plastic hinges will form on the body of beam only. are assumed to have adequate strength to resist

344
Proceedings of IOE Graduate Conference, 2019-Summer

additional load redistribution after loss of the primary In nonlinear dynamic analysis, the simulation of
column. instantaneous removal of column is done by replacing
the column with equivalent reaction[9]. For the
determination of equivalent reaction, a linear static
analysis is performed first using the ASCE 07[8]
extreme load case 1.2DL+0.5LL and internal forces at
the top joint of the column to be removed are
calculated. Then, the column is removed from the
model and the calculated forces are applied at the
column joint in opposite direction as a reaction. After
the column has been substituted with reaction forces,
a new linear static analysis is again performed and the
resulting flexural moments diagrams and deflections
Figure 2: Force-Deformation Relation[7] are compared with results obtained from the initial
linear static analysis that included the column as
The modulus of elasticity and compressive strength of shown in figure 4 . If both linear static analysis with
concrete is taken as 22360 Mpa and 20 Mpa the column and with substitute reactions resulted in
respectively. The yield strength of reinforcement is identical moment diagrams and deflections, then that
taken as 500 Mpa. reaction is taken as equivalent reaction for column
removal.
The ASCE 7 extreme-event load case 1.2DL+0.5LL is
used for all analysis[8]. .The structural loading applied
are taken as listed below.
Live load= 3KN/m2
Roof load= 1.5KN/m2
Wall load= 10KN/m
Slab dead load= 3.125kN/m2
The dead load of beam and column will be taken
automatically by SAP2000. The area load such as slab
dead load, floor finish, floor live load and roof live
load are transferred into the corresponding beam as
per tributary area.

Figure 3: location of column removal

For each building, the corner column, external column


near middle of long side and short side in first storey Figure 4: comparision of bending moment diagram
are removed one at a time as shown in figure 3. Hence with column and without column with substituate
all together there are 60 cases for 20 models. reactions

345
Dynamic Increase Factor for Progressive Collapse Analysis due to Sudden Column Removal

In this way the column is successfully replaced by trial DIF applied to extreme event load case only
equivalent superimposed reaction forces obtained from on the bay around loss location. The process
a static analysis of building using extreme event load is repeated untill the maximum plastic hinge
case applied to entire structure. The reaction obtained rotation or maximum vertical displacement at
is then removed over time as shown figure 5. the column location matched with NLD analysis
response.

Figure 5: Transition of structural model for column


removal

Figure 6: Procedure for determination of DIF


For dynamic analysis, following parameters are
taken[9].
Damping ratio = 1% 3. For the same model, conduct NLS analysis
Column removal time = T/20 without amplified extreme event load case to
Analysis Time Step = T/200 determine maximum ratio of moment demand
The period (T) used for the column removal duration (Mu) to yield moment capacity (My) of beams
is the period of the first mode to exhibit vertical around loss location[11]. Finally DIF value for
motion at the location of the removed column after corresponding Mu/My is obtained.
the column has been removed.
For Nonlinear Static Analysis, Non-linear staged
construction feature in SAP2000 is used to simulate
the column removal[10]. Three stages using 100 steps
per stage is used in analysis. In first stage, all the floor
are loaded with ASCE extreme loading.In second
stage, the additional load as per trial DIF are loaded
over the bays around the column removal only. In
third stage, column is finally removed .

3. Procedure for calculation of DIF


Figure 7: Procedure for determination of Mu
The procedure for the calculation of DIF consists of
following steps as shown in figure 6.

1. Obtain maximum plastic hinge rotation and


vertical displacement among all the beams of
the bay affected by column removal location by
conducting NLD analysis without amplified
extreme event load case

2. Conduct NLS analysis with the same model with Figure 8: Beam and column notation

346
Proceedings of IOE Graduate Conference, 2019-Summer

4. Results and discussion Table 4: calculation of Mu /My


Mu + Mu −
The tables 2,3 below shows DIF calculation for Mu+ Mu− My+ My− My My
external column removal near centre of shorter side in 32 78.46 126.06 81.86 142.40 0.96 0.89
three storey RC building of 4m bay spacing building 33 80.77 135.83 81.86 142.40 0.99 0.95
shown in figure 8. For each model with different 51 42.80 130.68 81.86 142.40 0.52 0.92
location of column removal, trial amplification factors 63 78.27 129.98 81.86 142.40 0.96 0.91
are applied to the load in static analysis to match 64 80.59 138.34 81.86 142.40 0.98 0.97
deformation level obtained from dynamic analysis. 82 49.68 130.84 81.86 142.40 0.61 0.92
1 56.37 82.38 57.16 97.35 0.99 0.85
2 56.71 92.99 57.16 97.35 0.99 0.96
20 46.18 95.09 57.16 97.35 0.81 0.98

From figure 9, it is observed that the maximum


displacement above the column removal location is
91.88 mm and corresponding maximum plastic
rotation of beams nearer or above the column removal
location is 0.212 radian which are obtained from
nonlinear dynamic analysis.To match response of
static analysis with the dynamic analayis, the load on
Figure 9: Displacement at column removal location static analysis are amplified with trial factor starting
from 1 and found out that the response are more
similar when the amplification factor is 1.119 with
Table 2: Response from dynamic analysis only 2% error. Hence DIF value of 1.119 at that
column removal location estimates the dynamic
storey Beam dis(mm) rot(rad) effects due to instantaneous removal of column at that
1 32 91.88 0.0201 location. Finally the DIF value 1.119 is plotted in
1 33 91.88 0.0212 figure 10 for max Mu/My ratio of 0.99 which is
1 51 91.88 0.0167 obtained in table 4 .
2 63 91.88 0.0187
2 64 91.88 0.0197 In the similar way, building with variable bay spacing,
2 82 91.88 0.0181 column removal location and seismic design level are
3 1 91.88 0.0125 analysed and corresponding DIF with Mu/My are
3 2 91.88 0.0141 plotted as shown in figure 10.
3 20 91.88 0.0199

Table 3: Response from static analysis

Beam dis(mm) rot(rad) dis(mm) rot(rad)


DIF = DIF = DIF = DIF =
1.12 1.12 1.119 1.119
32 103 0023 89.12 0.0193
33 103 0.0239 89.12 0.0197
51 103 0.0197 89.12 0.0166
63 103 0.0216 89.12 0.0179
64 103 0.0214 89.12 0.0185
82 103 0.0219 89.12 0183
1 103 0.0129 89.12 0.0123
2 103 0.0162 89.12 0.0135
20 103 0.0234 89.12 0.0195 Figure 10: Variation of DIF with max(Mu/My)

347
Dynamic Increase Factor for Progressive Collapse Analysis due to Sudden Column Removal

From figure 10, it is seen that DIF generally decreases 5. Conclusions


with increase in level of nonlinearity. Two distinct
trendlines are clearly noticed. The trendline’s slope is • It is observed that the nonlinear static analysis
gentle when the value of max(Mu/My) is less than 1 with suitable DIF can estimate the structural
and the slope is steep after max(Mu/My) is more than dynamic responses of beams within the affected
1. The value of max(Mu/My) less than 1 indicates that bays with accuracy.
the damaged structure is in elastic stage before
• Numerical results from analysis shows that the
applying the DIF. However the response of dynamic
dynamic amplification factor decreases as the
analysis will situate in inelastic stage. To drive the
nonlinearity level increases where max(Mu/My)
structure from elastic to certain point in inelastic stage
represents nonlinearity level. The slope of
from lower value of max(Mu/My), more gravity load
trendline of DIF with max(Mu/My) is gentle
is required in static analysis leading to the higher
when max(Mu/My) is less than 1 and the slope
value of DIF. Hence DIF decreases when
is steep after max(Mu/My) is greater than 1.
max(Mu/My) increases. Similarly max(Mu/My) more
than 1 represents the damaged structure is in post • The DIF value in all cases are less than 2 which
yield stage before the application of DIF. So small indicates that the use of amplification factor 2 in
load is enough for the structure to reach in final earlier version of guidelines was conservative.
inelastic stage response of dynamic analysis. Due to
that reasons the value of DIF is smaller for latter case • The latest version of guideline (UFC2009 and
than the earlier. The steep slope is due to the presence GSA 2013) showed the dependency of dynamic
of high level of nonlinearity and ductility of structure Increase Factor with ultimate level of
in post-yield stage which is consistent with the belief inelasticity only but from results, it is seen that
that structures with large deformation capacity will it is dependent on actual level of plastic
withstand less dynamic effect. To study the effect of deformation level.
seismic design level on dynamic Increase factor,
structural model 20 is redesigned considering the • It is also observed that when the seismic design
importance factor of 1.2. The dynamic increase factor level of structure increases, the DIF value
is calculated for model and compared with initial increases but nonlinearity level decreases. Also
model 20 having importance factor 1. The higher is the seismic design level, the building
comparision of dynamic increase factor on these two is less vulnerable to progressive collapse.
case is shown in figure 11 below.

References
[1] General Service Administration. Alternate path
analysis & design guidelines for progressive collapse
resistance (GSA 2013). washington (DC).
[2] Department of Defence. Design of buildings to resist
progressive collapse (UFC 4-023-03). washington
(DC).
[3] S. M. Marjanishvili. Progressive analysis procedure
for progressive collapse. 18(2):79–85.
[4] General Service Administration. Progressive collapse
Figure 11: Variation of DIF with seismic design level analysis and design guidelines for new federal office
buildings and major modernization projects (GSA
2003). washington (DC).
It is seen that in all column removal case, the value of [5] Department of Defence. Design of buildings to resist
progressive collapse (UFC 4-023-03). washington
dynamic increase factor increases with seismic design (DC).
level. Also the value of dynamic increase factor is
[6] Indian Standard Code. IS 1893 (part 1), criteria for
more in case of corner column removal case than earthquake resistant design of structures.
others. [7] American Society of Civil Engineers. Seismic
rehabilitation of existing buildings (ASCE 41-13).
new york (NY).

348
Proceedings of IOE Graduate Conference, 2019-Summer

[8] American Society of Civil Engineers. Minimum [10] Milan Bandyopadhyay and Atul Krishna Banik.
design loads for buildings and other structures (ASCE Progressive collapse of rigid and semi-rigid jointed
7-10). new york (NY). steel frames according to GSA 2013 and GSA 2003
[9] K. Marchand, A. McKay, and D. J. Stevens. guidelines. 6(3):211–223.
Development and application of linear and non-linear [11] American Concrete Institute. Building code
static approaches in UFC 4-023-03. In Proceedings requirements for structural concrete (ACI 318-08)
of the 2009 Structures Congress, pages 1–10. and commentary.

349

You might also like