Seismic analysis comparative study by using Eurocode 8 and RPA99

N.LAOUAMI
CGS,National Center of Applied research in earthquake engineering, 1Rue Kadour Rahim, Hussein Dey BP252, Algiers, Algeria

A.CHEBIHI
USTHB, FGC, BP 32, El Alia Bab Ezzouar, Algiers, Algeria.

ABSTRACT: This paper deals with a buildings seismic analysis comparative study by using Eurocode 8 and the Algerian code RPA99. The parameter which is taken into account in this study is the design spectrum defined in each of the two codes. Basic differences exist between the two design spectra, such as the site coefficients, the spectral shape of the spectrum and also the integration by Eurocode 8 of the near and far field concept. The effects of all these differences are studied by considering the four types of soils, and two types of buildings, flexible and rigid. The obtained results on the base shear and the lateral displacement show important differences between the two codes. Keywords: Building, RPA99, Eurocode 8, Seismic analysis, Design spectrum

1. INTRODUCTION During these recent years, seismic codes, as well in Algeria as in Europe, have been revised and updated, especially after some destructive earthquakes. The first version of the Algerian seismic code which calls Algerian earthquake resistant regulations, was RPA 81, it was modified to RPA88, and then to RPA99. Unfortunately, the May 21st 2003 destructive Boumerdes earthquake (Mw = 6.8) (Laouami, 2006) occurred in Algeria, and resulted in more than 2.000 caused deaths and 11.000 injuries. More than 100.000 buildings were heavily damaged and some 13.300 others totally collapsed. All these results led to a partial revision of the RPA99 which became RPA99 version 2003. The two codes take into account site effects by introducing different categories of sites. The Eurocode8 defines five main types of soil and two special types with a soil factor S for each type, whereas RPA99/2003 considers four types S1, S2, S3 and S4 without soil factor. The site classification system is based on definitions of mean shear waves velocity, standard penetration test, unconfined compression test, and relative density. Borcherdt (1994) recommended the shear wave velocity VS30 as a tool of classifying sites for building codes. Boore et al. (1994) indicate that the ideal parameter would be the average shear-wave velocity to a depth of one-quarter wavelength for the period of interest, as was used by Joyner and Fumal (1984). By the quarter-wavelength rule, 30m is the appropriate depth for period of 0.16 s for stiff soil and period values tend to increase as the soil gets softer (Boore et al., 1994). It should be noted that the design spectra depending on ground types are provided only for cases where the 30m of soil immediately below the site dominates the frequency content of the design motions. Table 1 shows the different soil types with shear wave velocity defined in both codes and the values of site factor S for spectra type 1 and type 2 of EC8.

Table 1. Ground types defined in EC8 and RPA99/2003 Soil type Eurocode 8 Rock (A-S1) Firm (B-S2) Soft (C-S3) Very soft(DS4) Vs,30>800 m/s 360<Vs,30<800 m/s 180<Vs,30<360 m/s Vs,30< 180 m/s Type 1 S=1 S=1.2 S=1.15 S=1.35 Type 2 S=1 S=1.35 S=1.5 S=1.8

RPA 99/2003 Vs800 m/s 400 Vs<800m/s 200Vs<400 m/s 100Vs<200 m/s

As seen from table 1, shear wave velocities for EC8 are taken for a depth of 30 m, whereas for RPA, the depth is 10 to 20 first meters, the site factor is taken into account only by EC8 and it is more important for spectra type 2. 2. ELASTIC AND INELASTIC DESIGN SPECTRA The design spectrum is an important parameter in the seismic code. The earthquake induced ground shaking is generally represented in the form of acceleration design spectra or displacement design spectra. Earthquake parameters such as soil condition, epicentral distance, magnitude, duration, and source characteristics influence the shape and amplitudes of response spectra. While the effects of some parameters may be studied independently, the influences of several factors are interrelated and cannot be discussed individually. Ambraseys et al. (1996) and Bommer and Acevedo (2004) presented and discussed the effects of earthquake magnitude, source-to-site distance, site classification, and style-of faulting on the strong-motion accelerograms and consequently on the response spectra. As known, the damping ratio and structural vibration period are other parameters affecting the response spectra. In all current seismic codes, the earthquake actions are represented in the form of a spectrum of absolute acceleration. EC8 defines two types of spectra: Type 1 for the far field and Type 2 for the near field. If the earthquakes that contribute most to the seismic hazard defined for the site for the purpose of probabilistic hazard assessment have a surface-wave magnitude, Ms not greater than 5.5, it is recommended that the Type 2 spectrum is adopted, if not, Type 1 is recommended. RPA99/2003 defines only one type of spectra depending on seismic zone and some other factors according to the building. The ordinates of elastic design spectra Se for EC8, and inelastic design spectra Sd, Sa respectively for EC8 and RPA99/2003 are given by their expressions in table 2. In this table, shows lower bound factor for the horizontal design spectrum, recommended value for is 0.2. The periods of elastic design spectra of EC8 and RPA99/2003 depend on soil type and are detailed in table 3. Seismic hazard is expressed in EC8 by a parameter namely reference peak ground acceleration at the rock surface for a reference return period. The reference return period recommended for the non-collapse performance level is the 475 year, corresponding to 10% probability of exceedance in 50 years. In EC8 the design ground acceleration ( ) is equal to times the importance factor I. The Algerian seismic code subdivided the territory into five zones of increasing seismicity as: Zone 0: neglected seismicity, Zone I: low seismicity, Zones IIa and IIb: moderate seismicity, Zone III: high seismicity. It defines a coefficient of zone acceleration A according to the seismic zone and the using group of the building. In table 2, S is the soil factor defined in EC8 depending on ground types and is the damping correction factor with a reference value of = 1 for 5% viscous damping. Elastic design spectra are drawn as shown in fig.1 using the expressions shown in table 2 for the soil types defined in the respective codes.

Table 2. Ordinates of elastic and inelastic spectra for EC8 and RPA99/2003 Eurocode 8 = RPA99/2003 ]

Table 3. Periods of design spectra defined in EC8 and RPA99/2003 Eurocode 8 Soil type A Type 1: Type 2: Soil type B Type 1: Type 2: Soil type C Type 1: Type 2: Soil type D Type 1: Type 2: Soil type E Type 1: Type 2: Soil type S1 , , Soil type S2 , Soil type S3 , , Soil type S4 , , , , , , , , RPA99/2003 ,

Fig.1 shows the design spectra proposed respectively by the Eurocode 8 and the RPA99. As mentioned before, there are some differences as well on the spectral form as on the spectral amplitude. The RPA99/2003 neglect the soil factor and considers the same spectral amplitude peak values for the four soil types, while the EC8 gives different spectral amplitudes taking account the soil factor and the near and far fields. The inelastic design spectra fig.2 is obtained by dividing the elastic response spectra by the behavior factor (q) in EC8 and (R) in RPA99/2003. The values of the maximum allowable behavior factor are taken considering the type of structural system, regularity in elevation and prevailing failure mode in the system with walls in EC8, whereas, the behavior factor in RPA depends only on the structural system. RPA99/2003 defined a quality factor of the structure (Q) which depends on the redundancy and the geometry of the constituent elements, the regularity in plan and in elevation, and the quality of the control of construction.

5 spectral acceleration 4 3 2 1 0 0 0.5 1

soil type S1-A EC8 type 1 EC8 type 2 RPA

soil type S2-B 5 spectral acceleration 4 3 2 1 0 0 0.5 1 1.5 periods (s) soil type S4-D EC8 type 1 EC8 type 2 RPA 5 spectral acceleration 4 3 2 1 0 0 0.5 1 1.5 periods (s) 2 2.5 3 EC8 type 1 EC8 type 2 RPA 2 2.5 3 EC8 type 1 EC8 type 2 RPA

1.5 periods (s) soil type S3-C

2.5

5 spectral acceleration 4 3 2 1 0 0 0.5 1 1.5 periods (s) 2

2.5

Figure 1. Elastic design spectra defined in EC8 and RPA99/2003.

0.4 spectral acceleration 0.3 0.2 0.1 0 0

soil type S1-A EC8 type 1 EC8 type 2 RPA

soil type S2-B 0.4 spectral acceleration 0.3 0.2 0.1 0 0 EC8 type 1 EC8 type 2 RPA

0.5

1.5 periods (s) soil type S3-C

2.5

0.5

1.5 periods (s) soil type S4-D

2.5

0.4 spectral acceleration 0.3 0.2 0.1 0 0

spectral acceleration

EC8 type1 EC8 type2 RPA

0.4 0.3 0.2 0.1 0 0

EC8 type1 EC8 type2 RPA

0.5

1.5 periods (s)

2.5

0.5

1.5 periods (s)

2.5

Figure 2. Inelastic design spectra for ground types defined in EC8 and RPA99/2003

5. STRUCTURAL DATA AND FINITE ELEMENT MODELING OF BUILDING The structure studied here is a 10 story mixed moment resistant frame-shear walls. The building height is 38.08 m, the two first stories height is 4.08 m and the other ones 3.74 m. The plan dimensions of the first and second floor are: 33 m by 38.4 m, and 22.2 m by 28.10 m for the other ones.

To evaluate the seismic response of the building, elastic analyses were performed by the response spectrum method using the computer program SAP2000. The seismic analyses of the building are carried out separately in the longitudinal and the transverse directions. However, seismic responses only for x direction are comparatively presented with graphs and tables in this paper for the sake of brevity. Sample finite element model is shown in fig.3. Degrees of freedom at the base nodes are fixed, for other nodes are left free. Therefore, there is no finite element model for subsoil to consider soil-structure interaction. Columns and beams are modeled with frame elements, structural walls are modeled with shell elements. Slabs have been considered as a rigid diaphragm in each story level. In the analysis, Youngs modulus and unit weight of concrete are taken to be 32000MPa and 25 KN/m3, respectively. The damping ratio is assumed as 5% in all modes. It is assumed that the building is sited in high seismicity zone, so the reference peak ground acceleration is taken to be 0.25 g that is recommended in high seismicity zone (zone 3) in RPA, and the same value is taken for for EC8 to make the comparison. 6. MODAL AND SEISMIC ANALYSIS OF THE BUILDING The mode numbers taken into account for this building is 12. The first eight modes with periods and participating mass ratios of the building are presented in Table 4. In the first and second mode the building vibrates respectively in the x and y direction. The third mode takes place as torsional mode.

Figure 3. View of three dimensional finite element model of the building

Table 4. First eight modes and modal participating ratios of the building Individual mode (%) mode 1 2 3 4 Period 0,628774 0,612355 0,470333 0,147303 Ux 60,757 0,002 0,13 24,469 Uy 0,002 61,07 0 0 Uz 0 0,021 0 0 Ux 60,757 60,759 60,889 85,359 Cumulative sum (%) Uy 0,002 61,072 61,072 61,073 Uz 0 0,021 0,021 0,021

5 6 7 8

0,146465 0,122867 0,072354 0,071793

0 0,024 9,029 0

24,006 0 0 9,076

0,048 0 0 0,014

85,36 85,384 94,412 94,412

85,078 85,078 85,078 94,155

0,069 0,069 0,069 0,084

As shown from table 4, only eight modes give a cumulative sum of the participating ratios greater than 90% for x and y directions. The base shears of the building and the lateral displacements of floors were acquired from seismic analysis using the design spectra corresponding to 5% critical damping and considering fixed base condition. Seismic analyses of building were carried out for four ground types defined in RPA and their equivalent in EC8. Fig.5 presents the base shears of the building. As seen from fig.5, base shears become more important for soft soils because of the low fundamental frequency of the building. The results show also that EC8 type 1 gives the maximum base shears for all soil types, because the ordinate of inelastic spectra of fundamental period of the building which is 0.628 s is more important for EC8 type 1, whereas EC8 type 2 gives values of base shear greater than RPA only for very soft soil (soil type D-S4), which can be explained by the importance of soil factor defined by EC8 for this soil type which is equal to 1.8. This result shows the importance of taking far field earthquake (Type1) when we deal with flexible structures.

Figure 5. Base shear of the building considering four types of soil

Fig.6 shows lateral displacements of stories of EC8 and RPA99. As seen from the plotted curves, the displacement increases when the soil gets softer and the maximum value is given by the last story. EC8 type 1 gives value of displacements greater than those of RPA and for all ground types, for very soft soil S4-D, the displacements given by EC8 type 2 and RPA are close.

0.04 displacement (m) 0.03 0.02 0.01 0 0

Soil type S1-A RPA EC8 type 1 EC8 type 2 displacement (m)

0.04 0.03 0.02 0.01 0 0

Soil type S2-B RPA EC8 type 1 EC8 type 2

8 5 story Soil type S3-C

4 5 story Soil type S4-D

0.04 0.03 displacement (m) 0.02 0.01 0 0

0.04 0.03 displacement (m) 0.02 0.01 0 0

RPA EC8 type 1 EC8 type 2

RPA EC8 type 1 EC8 type 2

4 5 story

4 5 story

Figure 6. Displacements of stories considering four ground types defined in EC8 and RPA99/2003

7. CON CLUSION The inelastic design spectra showed that acceleration values of RPA99/2003 have the same peak values for all ground types, whereas, EC8 type 1 and type 2 specifies different peak values depending on ground types, this difference is explained by the fact that EC8 takes into account two important parameters which is not the case of RPA99/2003: site effect by introducing site factor S, and near and far field. EC8 specifies the values of the maximum allowable behavior factor depending on type of structural system, regularity in elevation and prevailing failure mode in the system with walls, but the behavior factor defined in RPA99/2003 only depends on structural type of the structure. In this study, base shears and displacements increase when soil gets softer, so the maximum value is given for soil type S4-D (very soft soil). The displacement also increased with stories, which means that the maximum value of displacement is given in the last story. EC8 type 1 gives the maximum displacement and base shear values for all ground types, but for rock soil, site factor is equal to 1, so the differences with RPA may be explained by the importance of behavior factor of RPA99/2003 which is equal to 5 comparatively to that of EC8 which is equal to 3,9. The results show that EC8 type 2 gives base shears and displacements lower than RPA except for soil type S4-D, this is because of the importance of soil factor for this soil type.

REFERENCES Algerian earthquake resistant regulations RPA99/ version 2003. (2004), national center of applied research in earthquake engineering. Ambraseys, N. Simpson, K. Bommer, J. (1996). Prediction of horizontal response spectra in Europe. Earthquake Engineering and Structural Dynamics, 25:371-400. Bommer, JJ. Acevedo, AB. (2004). The use of real earthquake accelerograms as input to dynamic analysis. Journal of Earthquake Engineering. 8: 43-91. Boore, DM. Joyner, WB. Fumal, TE. (1994). Estimation of response spectra and peak accelerations from western North American earthquakes. An interim report, USGS Open-File Report: 94127.

Borcherdt, RD. (1994). Estimates of site-dependent response spectra for design (methodology and justification), Earthquake Spectra. 10: 617-653. Dogangn, A. Livaoglu, R. (2006). A comparative study of the design spectra defined by Eurocode 8, UBC, IBC and Turkish Earthquake Code on R/C sample buildings. Journal of Seismology. 10: 335-351. EC 8 (2004) Eurocode 8: design of structures for earthquake resistance Part 1: general rules, seismic actions and rules for buildings, European Norm. European Committee for Standardization, European Committee for Standardisation Central Secretariat, rue de Stassart 36, B-1050 Brussels. Joyner, WB. Fumal, TE. (1984) Use of measured shear wave velocity for predicting geologic site effects on strong ground motions. Proc 8thWorld Conference of Earthquake Engineering San Francisco. 2:777-783. Laouami, N. Slimani, A. Bouhadad, Y. Chatelain, J. L. and Nour, A. 2006. Evidence for fault-related directionality and localized site effects from strong motion recordings of the 2003 Boumerdes (Algeria) earthquake: consequences on damage distribution and the Algerian seismic code. Soil Dynamics and Earthquake Engineering. 26, pp 991-1003., Sabetta, F. Bommer, JJ. (2002). Modification of the spectral shapes and subsoil conditions in Eurocode 8. Proc 12th European Conference of Earthquake Engineering, London, Paper No 518. SAP2000 (2005) Integrated Software for Structural Analysis and Design: Computers and Structures. Inc. Berkeley, California.