Professional Documents
Culture Documents
INTERACTION EFFECTS
By H. L. Wong 1 and J. E. Luco, 2 Member, ASCE
Downloaded from ascelibrary.org by WALTER SERIALS PROCESS on 07/13/13. Copyright ASCE. For personal use only; all rights reserved.
INTRODUCTION
For the last 20 years there has been increased interest in the possible use
of active control techniques to limit the seismic response of structures. Some
of the advances in this area have been reviewed recently by Yang and Soong
(1988). A common assumption in most studies on active control of the
seismic response of structures is that soil-structure interaction effects are
small and, in particular, that the rocking motion of the base is negligible.
The objective of this study is to remove these assumptions and to consider
the seismic response of tall structures subjected to active control when the
flexibility of the soil is included in the analysis.
In this paper, the structure is modeled as a uniform shear beam supported
on a rigid foundation embedded in the soil represented by a uniform visco-
elastic half-space (Fig. 1). The seismic excitation is represented in the form
of vertically incident SH-waves. The kinematic interaction effects associated
with the embedment of the foundation together with the inertial interaction
effects result in a base motion that includes translational and rocking re-
sponse components. The seismic response of the structure, including soil-
structure interaction effects, is modified by use of a control force acting at
the top of the structure. The active control strategy used here is based on
the work of Vaughan (1968) and von Flotow (1986), in which the energy
flow within the structure is modified by controlling the reflection and/or
transmission of waves at end points or at joints. In this study, the active
control force is selected to simulate an absorbing boundary such that all
upward propagating waves are absorbed at the top of the structure and no
downward propagating waves are reflected at that point. Applications of
this approach to the active control of the seismic response to tall structures
H
+ (-pBAB$l + w 2 P s / B ) ( - ) LGB (10)
sin
P*
u(x) = uBe~ /(un/Pfl) ,-i(mHlf,B) (12)
The solution in this case includes both upward and downward propagating
waves in the structure but, as in the previous case, no resonance is obtained
within the superstructure. The corresponding expressions for the base force
and base overturning moment are given by
FB = -mpBAB$BuB - pBAB£B [1 - e-'-(«"'f»»>] 0, (13)
and
MB = -pBAB?,2B [1 - c-'(»"'&»] uB + ^pBIB(H/L)LQB
sin
PB
-/(aitt/pa). (14)
PBAB&[ y L8 S
P«/ _j
Substitution of the general solution given by (6) into the boundary conditions
given by (4) and (15) permits us to determine the unknown coefficients A
and B with the result
2240
(16)
cos —
The corresponding expressions for the base shear force FB and the base
overturning moment MB are
n ^wH
1 — cos —
F B = bipBAB$Btm\ — )uB + pBAB% — (17)
V PA / /to//
cos ——
VP*
and
/to//
1 - cos —
MB = pBAB$ \p«
cos
(t)
tan —
// \P« W
+ < -pBABft W * I ^ ) f £e fl . . . (18)
L to//
P«/ -1
Eqs. (16), (17), and (18) indicate that u(x), FB, and MB can be very large
in the vicinity of the fixed-base characteristic frequencies for which cos(co///
P«) = o.
-to 2 M 0 U 0 = FB F0 (19)
T
where U0 = (u0,LB0) is the generalized displacement of the bottom of the
foundation; and F B 0 = (FB0,MB0/L)T and - F ? = - ( F 0 , M 0 / L ) r are the
generalized forces that the structure and the soil exert on the foundation,
respectively. The generalized mass matrix of the foundation M 0 is given by
1 0 MF 0 1 (hl2L)
(20)
QiIlL) 1 0 (IF/L2) 0 1
2241
F„ = -G,LKB(o>)UB (21)
T
where JJB = (uB,LQB) . The elements of the matrix KB have been normalized
by the half-width L of the foundation and by the shear modulus of the soil
10
i—f—PI —I—I—i—r ~I—i—i—i—i—r
PS/PB=IO P S /PB = IO
n—i—I—i—i—i—r "1 1 1 1 1 1
A/PB=I MPB=I
ICQ.
<
J i i i u
3
/(Hz)
2242
PS/PB=10 PS/PB=IO
- No Control
- Exact AB Control
- Approx. AB Control
Downloaded from ascelibrary.org by WALTER SERIALS PROCESS on 07/13/13. Copyright ASCE. For personal use only; all rights reserved.
_i i i i i i i i i i_ -J A I I I I I I I I I
n—i—i—i—i—r ~I—i—I—I—r ~I i I I r
PS/PB = 2 PS/PB = 2
Gs = P*P?> where ps and f}s are the density and shear wave velocity of the
soil, respectively. Considering the relations
1 0
Fso - (22)
(h/L) 1
and
1 (h/L)
Ufl = U0 (23)
0 1
leads to
2243
F B 0 = -G s LK B 0 (o))U 0 (24)
in which
1 0 1 (hIL)
KHn — K, (25)
(ML) 1 0 1
The motion of the soil in contact with the foundation can be described
by the expression
2244
ps/pB = io p s /p B =io
Downloaded from ascelibrary.org by WALTER SERIALS PROCESS on 07/13/13. Copyright ASCE. For personal use only; all rights reserved.
• No Control
Exact AB Control
Approx. AB Control
i
I I I I I I I I I I I I I I I II I I
P S /PB = 2
i i i i i i i i
I I I I I I I I I I I I I I I I I I •••[ 1 M M II 1 1 1 1 1 1 1 1 1 1 1
P~ S /PB=I PS/PB=I
lea
<
-
z'" \_,
1 1M i l I I I ! E ! 1 1
/(Hz) /(Hz)
p s /p B = io PS/PB=IO
No Control
Exact AB Control
Approx. AB Control
Downloaded from ascelibrary.org by WALTER SERIALS PROCESS on 07/13/13. Copyright ASCE. For personal use only; all rights reserved.
i i i i i i i i i i i i i i m~
P S /PB = 2
i i i i i i n - m i i i i i i i ~r
PS/PB=I
/(Hz) /(Hz)
P , / P B = 10
- No Control
- Exacl AB Control
- Approx. AB Control
Downloaded from ascelibrary.org by WALTER SERIALS PROCESS on 07/13/13. Copyright ASCE. For personal use only; all rights reserved.
/(Hz) /(Hz)
force FB, the base overturning moment MB and the required control force
FT can be calculated easily.
PARAMETRIC STUDY
and foundation inertia ratio (IF/psL5) are also listed in Table 1. The soil
was modeled as a uniform viscoelastic half-space_characterized by complex
wave velocities a,. = a s (l + 2r'£„)1/2 and P.v = P5(l + 2/£p)1/2 for P- and
•S-waves, respectively. To take advantage of the numerical results presented
by Mita and Luco (1989) for the impedance functions and scattering^ coef-
ficients of square embedded foundations it was assumed that as = 2$s{v ~
1/3), &, = 0.0005, and £p =_CJJ)01. Three values of the soil shear wave
velocity $s corresponding to ps/pfi = 10, 2, and 1 were used to represent
stiff, intermediate, and soft soil conditions, respectively. For (3S = 150
m/s, PS/(3B = 10 corresponds to a rock with p, = 1,500 m/s and $J$B = 1
corresponds to a soft soil with ps = 150 m/s.
Numerical results for the 10- and 100-story-building models were obtained
for three soil conditions and for three cases corresponding to the absence
of control, control by the absorbing boundary defined by (8), and control
by the approximate absorbing boundary defined by (11). Results in the
frequency domain for the 10-story building are shown in Figs. 2, 3, and 4.
The corresponding results for a 100-story building are shown in Figs. 5,6,
and 7.
The amplitudes of the transfer functions [u-jJu] where uT = u(H) is the
total motion at the top of the structure and us is the free-field motion of
the ground surface are shown in Figs. 2 and 5 for 10- and 100-story buildings,
respectively. Also shown in Figs. 2 and 5 are the normalized amplitudes
\Frl( — mpBAB$Bu^\ of the control forces for the two control cases. A first
observation referring to the response at the top of the structure is that both
the exact and the simplified absorbing boundaries drastically reduce the
response and eliminate all resonant behavior. Both absorbing boundaries
lead to almost the same response at the top. The results for \ujJug\ also
indicate that the beneficial effects introduced by active absorbing boundaries
are not reduced in any way by soil-structure interaction effects. The nor-
malized values for the amplitude of the required control force FT also shown
in Figs. 2 and 5 indicate that: (1) Both types of absorbing boundaries lead
to almost the same control force; (2) as the soil becomes softer and the
2248
translational and rocking response at the base of the structure are presented
in Figs. 3 and 6 for 10- and 100-story buildings, respectively. These results
indicate that the use of absorbing boundaries drastically reduces the inertial
interaction effects on the base translation and rotation. In particular, the
rocking response is drastically reduced when absorbing boundaries are used.
As in the case of the response at the top, both types of absorbing boundaries
(exact and approximate) lead to almost the_same response at the base.
The normalized amplitudes \FBI — mpBAB$Bug\ and \MBI-iu>pBAB$BHug\
of the base shear force and base overturning moment are shown in Figs. 4
and 7 and 10- and 100-story buildings, respectively. The results in Figs. 4
and 7 indicate that the use of absorbing boundaries strongly reduces the
base shear force and the base overturning moment in the vicinity of the
characteristic frequencies of the system without control. At other frequen-
cies these quantities may be increased by the use of control through ab-
sorbing boundaries.
The results in Figs. 2-7 indicate that the introduction of absorbing bound-
aries eliminates the resonant response of the superstructure and, conse-
quently, reduces the force and moment that the superstructure exerts on
the foundation in the vicinity of the characteristic frequencies of the (un-
controlled) system. This reduction, in turn, leads to a decrease of the inertial
interaction effects and, in particular, of the rocking response.
The effects of soil-structure interaction on the amplitude of normalized
control force FT/( — iu>pBABfiBug) are summarized in Fig. 8. These results
indicate that the amplitude of the required control force decreases as the
soil becomes softer. It appears that this reduction is mainly associated with
kinematic interaction effects and, consequently, is stronger for the larger
structure founded on a deeper foundation.
CONCLUSIONS
The effects that the interaction between the structure and the soil may
have on the possibility of using active control techniques to modify the
seismic response of structures have been studied. It has been found that the
rocking of the foundation resulting from the kinematic and inertial inter-
action effects changes the form of the control rule required to obtain an
active absorbing boundary at the top of shear wall models of structures.
Active control by means of an exact absorbing boundary, which includes
the rocking effects, and by an approximate absorbing boundary, which
ignores the rocking effects, result in large reductions in the structural re-
sponse even when soil-structure interaction effects are included. In fact, the
amplitudes of the required control force and of the structural response
decrease as the soil becomes softer. The use of control by means of absorbing
boundaries also reduces the inertial interaction effects and, in particular,
drastically reduces the rocking response of the structure.
ACKNOWLEDGMENT
The work described here was supported by grants from the Ohsaki Re-
search Institute, Shimizu Corp. to the University of Southern California,
and the University of California, San Diego.
2249
B./|i B = 2
P,/Pll=l
_i I L_J I I L L_L_J
I I I I I I I TTTT
Approx. AB Control Approx. AB Comrol
H)-slorcy KH)-Morey
/'(Hz)
APPENDIX. REFERENCES
Apsel, R. J., and Luco, J. E. (1987). "Impedance functions for foundations embed-
ded in a layered medium: An integral equation approach." Earthquake Engrg.
and Struct, Dynamics, 15(2), 213-231.
Luco, J. E., and Wong, H. L. (1987), "Seismic response of foundations embedded
in a layered half-space." Earthquake Engrg. and Struct. Dynamics, 15(2), 233-
247.
Luco, J. E., Wong, H. L., and Mita, A. (1991). "Active control of the seismic
response of structures by combined use of base isolation and absorbing bounda-
ries." Earthquake Engrg. and Struct. Dynamics, (submitted for publication).
Mita, A., and Luco, J. E. (1989). "Impedance functions and input motions for
embedded square foundations." /. Geotech. Engrg., ASCE, 115(4), 491-503.
Mita, A., and Luco, J. E. (1990a). "Active vibration control of a shear beam with
variable cross section." Proc. of the 1990 Dynamics and Design Conference, Japan
Society of Mechanical Engineers, Jul., 276-279.
Mita, A., and Luco, J. E. (19906). "New active control strategy for tall buildings."
Proc. Eighth Japan Earthquake Engrg. Symp., Dec. 1869-1874.
Vaughan, D. R. (1968). "Application of distributed parameter concept to dynamic
analysis and control of bending vibrations." /. Basic Engrg., Jun., 157-166.
von Flotow, A. H. (1986). "Travelling wave control of large spacecraft structures."
J. of Guidance, Control and Dynamics, 9(4), 462-468.
Yang, J. N., and Soong, T. T. (1988). "Recent advances in active control of civil
engineering structures." Probabilistic Engrg. Mech., 3, 179-188.
2250