EARTHQUAKE ENGINEERING AND STRUCTURAL DYNAMICS Earthquake Engng Struct. Dyn.

2007; 36:24452453 Published online 2 August 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/eqe.738

SHORT COMMUNICATION

Inuence of the soilstructure interaction on the fundamental period of buildings

Louay Khalil , Marwan Sadek, , and Isam Shahrour
Laboratoire de M ecanique de Lille (UMR 8107), Universit e des Sciences et Technologies de Lille, Villeneuve dAscq, F-59655, France

SUMMARY This paper includes an investigation of the inuence of the soilstructure interaction (SSI) on the fundamental period of buildings. The behaviour of both the soil and the structure is assumed to be elastic. The soil-foundation system is modelled using translational and rotational discrete springs. Analysis is rst conducted for one-storey buildings. It shows that the inuence of the SSI on the fundamental frequency of building depends on the soilstructure relative rigidity K ss . Analysis is then extended for multi-storey buildings. It allows the generalization of the soilstructure relative rigidity K s to such complex structures. Charts are proposed for taking into account the inuence of the SSI in the calculation of the fundamental frequency of a wide range of buildings. Copyright q 2007 John Wiley & Sons, Ltd.
Received 10 October 2006; Revised 19 June 2007; Accepted 21 June 2007 KEY WORDS:

building; charts; fundamental period; relative rigidity; soilstructure interaction; seismic

1. INTRODUCTION Building codes generally use the fundamental period of buildings to assess their response to seismic loadings. This parameter is generally calculated using empirical formulas provided by seismic codes. These formulas generally ignore the soil exibility, which could drastically affect the fundamental period of buildings and consequently their overall seismic response. On the basis of measurements collected from buildings during earthquakes, Goel and Chopra [1] concluded that in the case of shear-wall structures, empirical formulas such as those proposed by the Uniform
Correspondence

to: Marwan Sadek, Laboratoire de M ecanique de Lille (UMR 8107), Universit e des Sciences et Technologies de Lille, Villeneuve dAscq, F-59655, France. E-mail: Marwan.Sadek@polytech-lille.fr Associate Professor. Graduate Student. Professor.

Copyright q

2007 John Wiley & Sons, Ltd.

2446

L. KHALIL, M. SADEK AND I. SHAHROUR

Building Code [2] are inadequate. Through a series of micro-vibrations tests, Ghrib and Mamedov [3] conrmed the inadequacy of the fundamental period formula given by the National Building Codes of Canada [4] or the Uniform Building Code [2], since they do not include the foundation exibility. The role of the soilstructure interaction (SSI) is usually considered benecial to the structural system under seismic loading since it lengthens the lateral fundamental period and leads to higher damping of the system. This conclusion could be misleading. Indeed, recent case studies and postseismic observations suggest that the SSI can be detrimental and neglecting its inuence could lead to unsafe design for both the superstructure and the foundation especially for structures founded on soft soil [5, 6]. Mylonakis and Gazetas [6] reported three cases of earthquakes (Bucharest 1977, Mexico City 1985 and Kobe 1995) where SSI caused an increase in the seismic-induced response of structures despite a possible increase in damping. They reported that Mexico earthquake was particularly destructive to l012 storey buildings founded on soft clay, whose period increased from about 1.0 s (assumption of a xed base structure) to nearly 2.0 s due to the SSI. Several authors attempted to determine the inuence of the SSI on the seismic response of buildings. First studies were conducted by Veletsos and Meek [7], Veletsos and Nair [8] and Bielak [9]. The exible-base period T is evaluated from Velestos and Meek [7] for a structure with a surface foundation as follows: T k kh 2 = 1+ + T ku k (1)

where T denotes the xed-base system period; ku and k stand for the rotational and translational springs, h and k designate the height and exural rigidity of the structure, respectively. Similar formula is recommended by the Building Seismic Safety Council [10]. On the basis of data recorded on 57 building covering a wide range of structural and geotechnical conditions, Stewart et al. [11] concluded that approaches proposed by Velestos and Bielak to predict T / T can reliably predict the effect of inertial interaction but are limited to single degree of freedom (SDOF) oscillators. The objective of the present paper concerns the elaboration of a simple procedure for taking into account the inuence of the SSI in the determination of the fundamental frequency of buildings. Analyses conducted for both one-storey and multi-storey buildings for various geotechnical conditions allow the construction of charts that give the fundamental frequency of a wide range of buildings in terms of the relative soilstructure stiffness.

2. ONE-STOREY BUILDING 2.1. Numerical model Analysis of the inuence of the SSI on the seismic response of buildings constitutes a complex task, since it depends on multiple factors that affect the fundamental frequency of buildings. In order to analyse this problem, the case study that will be rstly investigated concerns one-storey buildings modelled as a reinforced concrete frame founded on a homogeneous elastic soil layer (Youngs modulus = 50 MPa; density = 20 kN/m3 ; Poissons ratio = 0.3; shear wave velocity = 98 m/s). This frame is composed of a slab supported by two reinforced concrete square columns (Youngs modulus = 32 000 MPa; density = 24.5 kN/m3 ) and based on two isolated square footings (2.0 m 2.0 m). The height of each column is 4.0 m and its cross
Copyright q 2007 John Wiley & Sons, Ltd. Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

INFLUENCE OF THE SOILSTRUCTURE INTERACTION

2447

Figure 1. Numerical model for one-storey buildings.

section is 0.53 0.53 m2 (EI = 420 000 kN m2 ). Dimensions of the slab oor are 5.0 5.0 m2 and its thickness is equal to 0.3 m. Elastic beam elements are used to model the superstructure, while the soil-foundation system is represented by the means of translational (horizontal K h and vertical K v ) and rotational ( K ) discrete springs as illustrated in Figure 1. The rigidities of these springs are evaluated from the following formulas [12]: Kv = Gs (1 )
z

A,

K h = 2(1 +

s )G s x

A,

K =

1+ 4

Gs

x (a

+ b2 ) A

(2)

a and b denote the dimensions of the footing; G s and s designate the soil shear modulus and Poissons ratio, respectively. The factors x and z account for the geometry of the foundation (b/a ). For the present case study, springs rigidities are equal to K v = 115 000 kN/m; K h = 100 000 kN/m and K = 100 000 kN m/rd. Dynamic analyses were carried out using the nite element software Effel. It gives a fundamental frequency f = 7.23 Hz(T = 0.138 s) that differs about 18% from that obtained for the xed model ( f FB = 8.82 Hz) due to the SSI effect. 2.2. Soilstructure relative rigidity K ss The fundamental frequency of a xed-base structure f FB under lateral loading depends on the parameters E c , Ic and H which stand for Youngs modulus, the moment of inertia of the vertical resisting element and the storey height. The inuence of these parameters on the buildings fundamental frequency also depends on the SSI. For example, considering the reference case, the ratio f / f FB decreases from 0.819 to 0.576 (decrease of 30%) when the inertia moment of the vertical elements increases from Ic = 0.0135 to 0.135 m4 . This difference is less signicant for a stiffer soil with a shear wave velocity Vs = 200 m/s where f / f FB decreases from 0.949 to 0.823 (decrease of 13%) for the same variation in Ic . On the other hand and as indicated before, the inuence of the SSI is taken into account through the rotational and translational springs (Equations (1)(3)), which are function of the soil shear wave velocity Vs (= G s / ) and the foundation area A ( designates the soil density). As a consequence, the fundamental frequency f could be expressed as: f = F (Vs , H , E c , Ic , A)
Copyright q 2007 John Wiley & Sons, Ltd.

(3)

Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

2448

L. KHALIL, M. SADEK AND I. SHAHROUR

This equation shows that the fundamental frequency of the soil-foundationstructure system depends on ve parameters that characterize the soil (Vs ), the foundation ( A) and the structure ( H , Ic , E c ). Note that the inuence of span length has been investigated prior to the present study and was found insignicant. In the sequel, analyses are conducted for different values of inuential parameters in order to identity a dimensionless factor representing the soilstructure relative rigidity: Vs : 98, 107, 140, 200, 250 and 300 m/s (stiff soil); H : 2, 4, 7, 10 and 13 m; E c : 20 000, 32 000 and 42 000 MPa; Ic : 0.000139, 0.00135, 0.0131 and 0.1357 m4 . Results of these analyses are summarized in Table I. They were approximated using the following multilinear regression: log( f / f FB ) = A1 log(Vs ) + A2 log( H ) + A3 log( E c ) + A4 log( Ic ) f FB designates the fundamental frequency of the xed-base structure (without SSI effect).
Table I. Inuence of K ss on the fundamental frequency of one-storey buildings.
f / f FB 1 1 1 1 1 1 0.973 0.976 0.985 0.995 0.998 0.999 0.819 0.833 0.891 0.949 0.967 0.976 0.576 0.589 0.686 0.823 0.876 0.908 0.548 0.584 0.682 0.804 0.862 0.896 0.819 0.833 0.891 0.949 0.967 0.976
Copyright q

(4)

Vs (m/s) 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300

H ( m) 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 2 2 2 2 2 2 4 4 4 4 4 4

E c (MPa) 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000

I c ( m4 ) 0.000139 0.000139 0.000139 0.000139 0.000139 0.000139 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.135752 0.135752 0.135752 0.135752 0.135752 0.135752 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315

2007 John Wiley & Sons, Ltd.

Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

INFLUENCE OF THE SOILSTRUCTURE INTERACTION

2449

Table I. Continued.
f / f FB 0.963 0.983 0.963 0.983 0.989 0.993 0.966 0.972 0.983 0.992 0.995 0.996 0.979 0.985 0.987 0.994 0.997 1 0.981 0.985 0.991 0.996 0.997 0.998 0.961 0.967 0.981 0.991 0.994 0.996 0.881 0.899 0.936 0.968 0.979 0.985 0.787 0.814 0.877 0.936 0.958 0.969 0.666 0.699 0.785 0.878 0.917 0.939 0.523 0.558 0.657
Copyright q

Vs (m/s) 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140

H ( m) 7 7 7 7 7 7 10 10 10 10 10 10 13 13 13 13 13 13 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

E c (MPa) 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 20 000 20 000 20 000 20 000 20 000 20 000 42 000 42 000 42 000 42 000 42 000 42 000 20 000 20 000 20 000 20 000 20 000 20 000 42 000 42 000 42 000 42 000 42 000 42 000 20 000 20 000 20 000 20 000 20 000 20 000 42 000 42 000 42 000

I c ( m4 ) 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.00135 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.01315 0.135752 0.135752 0.135752 0.135752 0.135752 0.135752 0.135752 0.135752 0.135752

2007 John Wiley & Sons, Ltd.

Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

2450

L. KHALIL, M. SADEK AND I. SHAHROUR

Table I. Continued.
f / f FB 0.784 0.845 0.883 0.316 0.338 0.413 0.557 0.661 0.735 0.842 0.863 0.911 0.954 0.971 0.978 0.911 0.924 0.953 0.977 0.986 0.987 0.953 0.955 0.974 0.988 0.993 0.995 Vs (m/s) 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 98 107 140 200 250 300 H ( m) 4 4 4 2 2 2 2 2 2 7 7 7 7 7 7 10 10 10 10 10 10 13 13 13 13 13 13 E c (MPa) 42 000 42 000 42 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 32 000 I c ( m4 ) 0.135752 0.135752 0.135752 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315 0.1315

Values of the coefcients of the multilinear regression are: A1 = 0.233, A2 = 0.33, A3 = 0.12 and A4 = 0.07. They are obtained with a determination coefcient R 2 = 0.7. Normalizing the coefcients of the regression by A3 gives A1 / A3 2, A2 / A3 3 and A4 / A3 0.75. As a consequence, the proposed dimensionless parameter K ss representing the soilstructure relative rigidity could be approximated as follows: K ss = Vs2 H 3 E c ( Ic )3/4 (5)

Expressions of discrete springs (Equation (2)) reveals that the inuence of the square of soil shear wave velocity (Vs2 ) is equivalent to the square root of foundation area ( A); hence, the expression of the soilstructure relative rigidity K ss is turned into Vs2 H 3 E c ( Ic )3/4 A A0

K ss =

(6)

where A0 denotes a reference area (1 m2 ). Figure 2(a) shows the variation of the ratio f / f FB with the relative rigidity K ss . It can be observed that this variation can be approximated using a simple chart. For log( K ss )>1.5, the ratio f / f FB
Copyright q 2007 John Wiley & Sons, Ltd. Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

INFLUENCE OF THE SOILSTRUCTURE INTERACTION

2451

1.2 1 0.8 f/fFB 0.6 0.4 0.2 0 0 1 Log (Kss)

-4

-3

-2

-1

(a)
1.2
Ns

1.2 1 Ns f/fFB Nbt Nbl 0.8 0.6 0.4 0.2 0 0 Log (Kss)

1 0.8

f/fFB

0.6 0.4 0.2 0 0 1 Log (Kss)

-4

-3

-2

-1

-4

-3

-2

-1

(b)

(c)

Figure 2. Inuence of K ss on the fundamental frequency of buildings: (a) one-storey buildings; (b) multi-storey buildings with one span in the lateral and transverse directions; and (c) multi-storey buildings with multiple spans.

attains an asymptote f / f FB = 1. In this case, the fundamental frequency of the soil-foundation structure system is close to that of a xed-base structure. For log( K ss )<1.5, an important variation is observed for f / f FB with log( K ss ). In this case, neglecting the SSI could lead to a signicant misestimation of the fundamental frequency the soil-foundationstructure system and consequently to a poor prediction of the overall dynamic response of the structure. For stiff structures on soft soils (log( K ss )< 1.5), the frequency of the xed-base model f FB exceeds by three times the frequency of the soil-foundationstructure system. 3. MULTI-STOREY BUILDINGS Investigations were conducted for multi-storey buildings with different conditions for both the soil and the structure. Results obtained for buildings with different values of the total number of stories Ns (1, 3, 5, 7 and 10) and one span in both the longitudinal and transversal directions shows that the soilstructure relative rigidity could be expressed as follows: Vs2 H 3 A A0

K ss =
Copyright q 2007 John Wiley & Sons, Ltd.

Ns E c ( Ic )3/4

(7)

Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

2452

L. KHALIL, M. SADEK AND I. SHAHROUR

1.2

0.8

f/fFB

0.6

0.4

0.2

-4

-3

-2

-1

0 Log (Kss)

Figure 3. Chart for the consideration of the SSI in the determination of the fundamental frequency of buildings.

This expression indicates a decrease of K ss with the increase in the number of stories, and consequently to an increase in the inuence of the SSI on the fundamental frequency of the building. Figure 2(a) depicts the variation of f / f FB with the relative rigidity K ss . It shows that this variation could be approximated using a simple chart that is similar to that obtained for one-storey buildings. Calculations were also conducted for multi-storey buildings with different values of the number of spans in both the transversal and longitudinal directions ( Nbt and Nbl ) and different foundation areas ( A1 , A2 and A3 ) proportional to the transmitted loads. Analyses show that the relative rigidity for these buildings could be expressed as follows: Nbt Nbl Vs2 H 3 A A0

(8) Ns E c ( Ic )3/4 where A is the mean value of foundations area A = ( A1 + A2 + + An )/ n . The increase in the number of spans ( Nbl or Nbt ) leads to higher values of the relative rigidity K ss and consequently to a decrease in the inuence of the SSI. Figure 2(c) depicts the variation of the fundamental frequency ratio ( f / f FB ) with K ss for buildings with multiple stories and spans. It conrms trends obtained for one-storey buildings. Figure 3 summarizes results obtained for one-storey and multi-storey buildings. It can be observed that the fundamental frequency ratio ( f / f FB ) can be determined in terms of the relative rigidity K ss (Equation (8)) using the chart described earlier for one-storey buildings. 4. CONCLUSION This paper included the analysis of the inuence of the soilstructure interaction (SSI) on the fundamental period of buildings. Analyses conducted for various soil and structure conditions
Copyright q 2007 John Wiley & Sons, Ltd. Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe

K ss =

INFLUENCE OF THE SOILSTRUCTURE INTERACTION

2453

showed that this inuence depends mainly on the soilstructure relative rigidity, which could be expressed in terms of the soil shear wave velocity (Vs ), the foundation area ( A), the exural rigidity of the building columns ( Ic , E c ), the storey height ( H ), the number of stories ( Ns ) and spans ( Nbt , Nbl ) as follows: Nbt Nbl Vs2 H 3 Ns E c ( Ic )3/4 A A0

K ss =

A chart is proposed for an ease consideration of the inuence of the SSI in the determination of the fundamental frequency of buildings. For exible buildings (log( K ss )>1.5), the SSI could be neglected. For lower values of K ss , ignoring the SSI could lead to a signicant misestimation of the fundamental frequency of buildings.
REFERENCES 1. Goel RK, Chopra AK. Period formulas for concrete shear wall buildings. Journal of Structural Engineering (ASCE) 1998; 124:426433. DOI: 10.1061/(ASCE)0733-9445(1998)124:4(426). 2. Uniform Building Code. International Conference of Building Ofcials, Whittier, CA, U.S.A., 1997. 3. Ghrib F, Mamedov H. Period formulas of shear wall buildings with exible bases. Earthquake Engineering and Structural Dynamics 2004; 33:295314. DOI: 10.1002/eqe.344. 4. NBCC (95)National Research Council Canada. Supplement to the National Building Code of Canada, Ottawa, Canada, 1995. 5. Gazetas G, Mylonakis G. Seismic SoilStructure Interaction: New Evidence and Emerging Issues. Emerging Issues Paper, Geotechnical Special Publication No. 75, vol. 3. ASCE: New York, 1998; 11191174. 6. Mylonakis G, Gazetas G. Seismic soilstructure interaction: benecial or detrimental? Journal of Earthquake Engineering 2000; 4(3):277301. DOI: 10.1142/S1363246900000175. 7. Veletsos AS, Meek JW. Dynamic behavior of building-foundation systems. Earthquake Engineering and Structural Dynamics 1974; 3(2):121138. 8. Veletsos AS, Nair VV. Seismic interaction of structures on hysteretic foundations. Journal of Structural Engineering (ASCE) 1975; 101(1):109129. 9. Bielak J. Dynamic behavior of structures with embedded foundations. Earthquake Engineering and Structural Dynamics 1975; 3(3):259274. 10. BSSCBuilding Seismic Safety Council. The 2003 NEHRP Recommended Provisions for New Buildings and Other Structures Part 2: Commentary (FEMA 450). Federal Emergency Management Agency, Washington, DC, 2003. 11. Stewart JP, Seed RB, end Fenves GL. Seismic soilstructure interaction in buildings II: empirical ndings. Journal of Geotechnical and Geoenvironmental Engineering (ASCE) 1999; 125(1):3848. DOI: 10.1061/(ASCE)10900241(1999)125:1(26). 12. Newmark NM, Rosenblueth E. Fundamentals of Earthquake Engineering. Prentice-Hall: Englewood Cliffs, NJ, 1971.

Copyright q

2007 John Wiley & Sons, Ltd.

Earthquake Engng Struct. Dyn. 2007; 36:24452453 DOI: 10.1002/eqe