Effect of Interstorey Drift Limits On High Ductility in Seismic Design of Steel Moment Resisting Frames
Effect of Interstorey Drift Limits On High Ductility in Seismic Design of Steel Moment Resisting Frames
Effect of Interstorey Drift Limits On High Ductility in Seismic Design of Steel Moment Resisting Frames
net/publication/277990091
CITATION READS
1 542
1 author:
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Muhammad Tayyab Naqash on 10 June 2015.
Abstract
The current research activity deals with the seismic design of perimeter steel moment resisting
frames of 9, 7and 5 storeys with several span lengths(9.15m, 7.63m, 6.54m and 5.08m) using
Eurocode 8. In total 24 cases are designed and analysed using Ductility Class High(DCH) having
behaviour factor equals 6.5. In order to shed light on the drift limitations of Eurocode 8, the designed
frames are then checked by means of iteration to investigate the optimal behaviour factor. The
evaluated behaviour factor is then compared with the code provided behaviour factor and with the
evaluated ductility factor of frames, obtained through the use of static nonlinear analysis. Hence the
influence of drift criteria on the capacity design rules of Eurocode 8 is investigated. The frame
performances are measured in terms of over strength and redundancy factors, strength demand to
capacity and drift demand to capacity ratios allowing to the point highlighted conclusions.
Key Words: Moment Resisting Frames, High Ductility Class, Drift Limits, Eurocode
133
Pak. J. Engg. & Appl. Sci. Vol.16, Jan., 2015
8
4
45.75
7
7.63 7.63 7.63 7.63 7.63 7.63
4
1 2 3 4 5 6 7 6
4
5
36
45.77 4
6.55 6.54 6.54 6.54 6.54 6.54 6.54
4
3
1 2 3 4 5 6 7 8
4
2
4
45.75 1
4
134
Effect of Interstorey Drift Limits on High Ductility in Seismic Design of Steel Moment Resisting Frames
isperformedin order to check the performance of the designed using EN 10025-2 S275 grade structural
frames.The fundamental period of vibration from the steel with the following properties: Adopted steel
codified formulation is found 1.3secfor 9-storeys, grade: EN 10025-2 S275 having unit density (ρ) =
1.03sec for 7-storeys and 0.8 sec for 5-storeys, which 76.9 kN/m3, poissons ratio (ν) = 0.3, modulus of
is lower than the period obtained from the modal elasticity (E in MPa) = 2.10E+05, yield stress (fy in
response spectrum analysis (see Table2). MPa) = 275, ultimate stress (fu in MPa) = 275,
expected yield stress (fye in MPa) = 302.5 and
1.0 Sd [g] Sae [g] expected ultimate stress (fu in MPa) = 473.
0.6
q_5 q_6.5
3.1 General
Static pushover analysis has been carried out
0.4 using FEMA-356[12] recommendations for
evaluating the lateral load resisting performance of
0.2
the frames. For this reason triangular distribution
0.0
T [s] (unit load at roof level) of static incremental loads
0 1 2 3 4
has been applied and the displacement at the roof
level has been controlled. For the ultimate rotation
Fig. 2 Eurocode 8 Design Spectra for Various q capacity of an element, acceptance criteria is defined,
Factors this is represented as IO (Immediate Occupancy), LS
(Life Safety) and CP (Collapse Prevention).FEMA
This is due to the fact that simplified formulae 356 acceptance criteria for non-linear procedure are
given by seismic codes tend to underestimate the adopted here. Mechanical non-linearity of the
fundamental period of vibration, as they are based on members has been assumed to be concentrated in
empirical evaluation, therefore globally accounting plastic hinges at the ends (lumped plasticity) of the
for the stiffening effects of non-structural elements elements. Furthermore, as steel moment resisting
too, e.g. partition walls and in-fills etc. The frames own relatively long period therefore the
connections of the examined frames are assumed “equal displacement rule” is employed to evaluate the
fully rigid, therefore the detail discussions and their so-called common parameters like “over-strength
influences are assumed beyond the scope of the factor”, “ductility factor”, “elastic over-strength” and
current research. All the framing members are “redundancy factor”.
135
Pak. J. Engg. & Appl. Sci. Vol.16, Jan., 2015
Fig.3 Pushover Curves and Normalised Pushover Curves for 9 Storeys Frames: (a, d) Redundancy Factor, (b, e)
Pushover Curve and (c, f) Global over-strength
Fig.4 Pushover Curves and Normalized Pushover Curves for 7 Storeys Frames: (a, d) Redundancy factor, (b, e)
Pushover Curve and (c, f) Global over-strength
136
Effect of Interstorey Drift Limits on High Ductility in Seismic Design of Steel Moment Resisting Frames
Finally, in these graphs the total base shear (Vb) The effect of the drift limits L1 (0.01h) andL2
is normalised with respect to Vd(the design base (0.0075h) can also be observed, for instance, the
shear), therefore giving rise to global over strength redundancy factors are in the same range
factors ( EP ) as shown by Figure 3(c, f), Figure 4(c, approximately as expected, the base shear increases
f) and Figure 5(c, f). when drift limit L2 is employed for a corresponding
frame (see Figure 3b for L1 and Figure 3e for L2)
In the normalised graphs the top displacement
and the global over-strength also increases[13, 14].
(Dt) is normalised with 1 (the displacement
corresponding to the first plastic hinge) therefore 3.3 Stiffness and Over-stiffness of the
showing the corresponding ductility of the frames. In Analysed Frames
all these graphs the top row shows graphs for the In this section, stiffness and overstiffness of the
frames when drift limit L1 (0.01h) is employed in the designed frames are reported. It is normal that as the
design with high ductility, whereas the bottom row earthquake forces pushes the structure, the
shows graphs when drift limit L2 (0.0075h) is
redistribution of the seismic forces take place due to
employed in the design. It is evident that as the
the formation of plastic hinges. This redistribution
number of storey increases:
causes the reduction of stiffness of the structures
The global over-strength decreases, for thereby the ductility of the structure increases. The
example see Figure3c (9 storeys) and Figure reduction in stiffness due the increase in fundamental
4c (7 storeys) in which global over-strength period that accompanies ductile behaviour tends to
is high for 7 storey frame, increase the amount of displacement the structure
The base shear increases, for example see will experience as it is pushed by earthquake forces.
Figure3b (9 storeys) and Figure 4b (7 The over-stiffness (k) is given by eq (1):
storeys) where slightly high base shear can
be observed in the case of 9 storeys.
k
Vb
The redundancy factors remain Velastic
approximately in the same range, for Limit
example see Figure3a (9 storeys) and Figure (1)
4a (7 storeys).
Fig. 5 Pushover Curves and Normalized Pushover Curves for 5 Storeys Frames: (a, d)
Redundancy Factor, (b, e) Pushover Curve and (c, f) Global over-strength
137
Pak. J. Engg. & Appl. Sci. Vol.16, Jan., 2015
Where Vb is the base shear obtained from related to the damageability limit state) and ∆Limit is
pushover analyses, ∆ is the corresponding the Inter-storey drift limit.
displacement in the push-over, Velastic is the base The stiffness and the over-stiffness of the
shear obtained from modal analyses using the elastic designed frames are shown in Table 3 and are
spectrum reduced by a factor equals 2.0 (that allow illustrated by Figure 6, Figure 7 and Figure 8 for 9, 7
for the lower return period of the seismic event and 5 storey frames, respectively.
Fig.6. Over-stiffness Factors for 9 Storeys frames: (a) Limit = 0.01h and (b) Limit = 0.0075h
Fig. 7. Over-stiffness Factors for 7 Storeys Frames: (a) Limit = 0.01h and (b) Limit = 0.0075h
138
Effect of Interstorey Drift Limits on High Ductility in Seismic Design of Steel Moment Resisting Frames
Fig. 8 Over-stiffness Factors for 5 Storeys Frames: (a) Limit = 0.01h and (b) Limit = 0.0075h
The overstiffness for a given frame when its 3.4 Ductility, Redundancy and
around 1.0 shows that the design of the frame is Overstrength Factor
dictated by damageability (drift criteria), instead
In Figure 9, the calculated overstrength from the
when it is greater than 1.0 means that strength
codified formulations ( calc) is shown for all the 24
controls the design. It is clear in the case of 9 story
designed frames. From these graphs, an increasing
frame when sized for drift limit L2 (0.0075h), the
trend can be observed as the No. of storeys of frame
drift governed the design of the frames (see Figure
6b) and hence the assumed ductility (q equals 6.5) is decreases from 9 to 5. Furthermore, overstiffness of
not utilised completely. In order, to see the influences the frames ( k) are also reported which were
of drift limits in the forthcoming sections, the mentioned in the previous section. In addition, elastic
ductility factors are evaluated in this section. Further overstrength ( E ), global overstrength ( EP )
the optimum behaviour factor is evaluated for each demonstrate an increasing trend whereas redundancy
frame by iterations. factor ( P ) for each frame is always constant.
Limit = 0.0075h
6 6 6 Bays
Limit = 0.0075h
5 Bays
Limit = 0.01h
5 5 Limit = 0.01h
4 4
3 3
2 2
1 1
Ns Ns
0 0
9 7 5 - 9 7 5 9 7 5 - 9 7 5
calc E E, (a) calc E E, (b)
6 7 Bays 6 9 Bays
Limit = 0.01h Limit = 0.0075h
5 5 Limit = 0.01h Limit = 0.0075h
4 4
3 3
2 2
1 1
Ns Ns
0 0
9 7 5 - 9 7 5 9 7 5 - 9 7 5
calc E E, (c) calc E E, (d)
Fig. 9 Over-strength Factors for the Analysed Frames: (a) 5 Bays (b) 6 Bays (c) 7 Bays and (d) 9 Bays
139
Pak. J. Engg. & Appl. Sci. Vol.16, Jan., 2015
From all these parameters the effect of drift These are strictly related to the span of the
limit can be easily observed as all such parameters frames as well as to the drift limitations. From the
except overstiffness of frames increases when the design and analysis of 9 storeys frames, Eq (2) can be
drift limit changes from L1 (0.01h) to L2 (0.0075h). used to illustrate the case.
3.5 Optimum Versus Code Prescribed u
Behaviour Factors qcode qoptimum and qoptimum (2)
y
In Figure 10, Figure11 and Figure 12 the
behaviour factors are plotted but it is observed that Therefore, leads to declare that these frames will
the obtained behaviour factor from pushover analysis be suitable if designed with medium ductility (q =
are high from the code specified factor for short span 4.0) rather than 6.5.Similarly for 7 and 5 storey
frames in the cases of both drift limits (L1 and L2). It frames the relation as shown by Eq (3) holds,
has to be mentioned here that the ultimate base share representing that the ductility of the frames increases
is defined as the maximum obtained from the as the number of storeys decreases.
pushover analysis. The optimum q factors are
u
obtained by iterative procedure from response qcode qcalculated and qoptimum (3)
spectrum analysis. y
12 Limit = 0.01h 12
q Limit = 0.0075h
q
10 10
8 8
6 6
4 4
2 2
LS / Nb LS / Nb
0 0
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
u/y q_code q_optimum (a) u/y q_code q_optimum (b)
10 q
Fig. 10 Ductility andbehaviour
= 0.01h Factors for Storeys Frames: (a) Limit = 0.01h and (b) Limit = 0.0075h
9= 0.0075h
Limit Limit
8
6
Limit = 0.01h Limit = 0.0075h
12 q 12 q
4
10 10
8 2 8
Nb
6 0 6
5 6 7 9 -- 5 6 7 9
4 q_obtained q_code q_optimum 4
2 2
L S / Nb L S / Nb
0 0
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
u/y q_code q_optimum (a) u/y q_code q_optimum (b)
10 q
Fig. 11 Ductility and behaviour Factors for 7 Storeys Frames: (a) Limit = 0.01h and (b) Limit = 0.0075h
8
4
Limit = 0.01h Limit = 0.0075h
140
2
Nb
0
5 6 7 9 -- 5 6 7 9
q_obtained q_code q_optimum
Effect of Interstorey Drift Limits on High Ductility in Seismic Design of Steel Moment Resisting Frames
8 8
6 6
4 4
2 2
LS / Nb LS / Nb
0 0
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 (b)
u/y q_code q_optimum (a) u/y q_code q_optimum
Fig. 12 Ductility and behaviour Factors for 5 Storeys Frames: (a) Limit = 0.01h and (b) Limit = 0.0075h
10
In the above
6 designed frames the calculated The behaviour factor specified by the code for
behaviour factor from pushover analysis are strictly high ductility is not completely utilised due to the
4
related to the period of = 0.01h the
Limitthe structures, as Limit
= 0.0075h
period of high Interstorey drift limits given by the code, this
the frame increases the ductility and thus the needs either to design such frames with Medium
2
corresponding behaviour factor decreases. For Ductility or relaxing the capacity design rule.
N
instance for the0
analysed cases, the fundamental b
At the end it is found that if these frames are It is therefore required to propose the design of
designed with medium ductility (q equals 4.0) it frames in a more sophisticated way as it gives high
might result in optimum use of behaviour factor and performance and completely avoiding or relaxing the
therefore shall lead to more economical solution. capacity design rules.
Furthermore, in these cases if a frame is designed The code specified ductility class (high) is not
with high ductility (q equals 6.5) the capacity design compatible with the code proposed drift limits;
rules could be relaxed by redefining the over-strength instead it is strictly important to limit the ductility
factor or at least could limit the elastic over-strength. when the design is govern by drift.
141
Pak. J. Engg. & Appl. Sci. Vol.16, Jan., 2015
142