9th GRACM International Congress on Computational Mechanics Chania, 4-6 June 2018
DYNAMIC RESPONSE OF MASONRY BUILDINGS CONSIDERING THE TIMEDEPENDENT SOIL SATURATION CONDITIONS 1 2* Alexandros Liratzakis , Yiannis Tsompanakis 1
School of Environmental Engineering Technical University of Crete Chania, GR-73100, Greece e-mail:
[email protected]
2
School of Environmental Engineering Technical University of Crete Chania, GR-73100, Greece e-mail:
[email protected]
Keywords: Masonry Buildings, Seismic Hazard, Soil-Structure Interaction, Soil Saturation, Incremental Dynamic Analysis, Limit States, Rehabilitation Measures. Abstract. The main aim of this work is to investigate the seismic vulnerability of traditional masonry buildings, taking into account the impact of dynamic soil-structure interaction (SSI). More specifically, the dynamic response of a typical unreinforced masonry (URM) building constructed over a silty sand layer is examined. The main novelty of the present study is that it considers time-varying soil mechanical properties, i.e., depending on the soil saturation level, which usually varies with time. In addition, a new structural assessment approach, which aims to accurately assess the performance levels (Limit States) of historic buildings and monuments after performing certain seismic rehabilitation measures has been applied. Under this perspective, a quite simple and efficient -in terms of time, cost and effectiveness- intervention was considered, in which blocks of expanded polystyrene (EPS) geofoam are placed at the perimeter of the foundation of theURM building in order to improve its dynamic response and reduce its seismic vulnerability under the examined circumstances. Subsequently, the calculation of building’s nominal life is performed in a realistic manner by taking into account the annual changes in the soil saturation level. 1 INTRODUCTION Modern regulations for the evaluation of existing structures [1-3] are based on performance-based assessment, which aims to implement a number of limit states in relation to predetermined seismic actions scenarios [4]. On the other hand, the challenge of balancing safety versus maintenance of the architectural and artistic features of historic structures remains a crucial issue to address, usually on a case-by-case basis. Τhere is a lack of a unified approach for the assessment of seismic risk of monuments and historic structures. Ensuring the monumental buildings’ integrity in long terms is an issue that needs further attention and improvements. The most important difficulty, in creating a holistic methodology for assessing the seismic hazard of cultural heritage structures, is the limitations that are imposed on structural interventions. The constraints on the implementation of interventions in cultural heritage buildings are arising from the internationally accepted guidelines and the conceptual differences compared to design of new structures [5]. The best retrofitting practice in monumental structures is considered the application of reversible interventions in order to limit their vulnerability in a less intrusive manner. Performance through Limited Duration Rehabilitation Interventions (LDRI) is a new methodology, which aims to assess the seismic risk of monumental structures [6, 7]. This methodology attempts to provide a framework that quantifies the “safe” duration (i.e., the nominal life) of an intervention that upgrades structural integrity in a specified manner. The nominal life of an intervention is defined as the period for which this action ensures that the structure fulfills selected performance level(s) for a certain seismic scenario (e.g., probability of exceedance 10% and 20% in 50 years, respectively for Significant Damage and Damage Limitation Levels). A typical two-storey URM building, recently presented by the authors [8], was selected as a case study to perform the LDRI methodology, introducing also a new retrofitting scheme as it will be presented in the sequence. In the previous work [8], the impact of silty sand soil saturation level on the dynamic response of a typical masonry building was presented. Additionally, a slight retrofitting scheme was also examined, in which reinforced concrete (RC) friezes were placed at the floor levels and wooden lintels were replaced with RC lintels. The results indicated that the examined building presented higher drifts when the structure was constructed on relatively dry soil, while the slight strengthening with RC elements substantially improved its
Alexandros Liratzakis and Yiannis Tsompanakis
response and also reduced the impact of saturation conditions. As an extension, the present study proposes the application of a new mitigation method capable of also improving the dynamic structural response, depending on the soil saturation level. More specifically, a layer of expanded polystyrene (EPS) geofoam is placed between the surrounding silty sand layer and the external side at the perimeter of building’s foundation, acting as a compressible “shield”. EPS blocks are commonly used as a lightweight filling material in many civil engineering applications, e.g., embankments, retaining walls, pipelines, ground vibration isolations, etc. This simple, economic and fast intervention, which does not cause any structural or functional disturbance, aims to isolate the structure from ground shaking and absorb most of the seismic energy. In addition, this paper attempts to determine the nominal life of the URM building after this slight intervention considering either constant or annually varying soil saturation conditions.
2 THEORETICAL BACKGROUND 2.1 Performance-based design/assessment As already mentioned, modern seismic design norms/guidelines for the seismic design of new structures and for the assessment of interventions in existing buildings have included state-of-the-art methodologies for assessing the structural response based on performance-based assessment for certain limit states (design levels). 2.1.1. Greek Code for Structural Interventions The Greek norm for structural interventions for existing reinforced concrete structures (Greek Code for Structural Interventions (CSI) [9]) has adopted two seismic hazard levels: - Seismic excitation with exceedance probability 50% in 50 years. - Seismic excitation with exceedance probability 10% in 50 years. In addition, CSI defines three performance levels, namely: Damage Limitation, Significant Damage, and Near Collapse, for structures with a conventional lifetime of TL=50 years. Accepting that the cultural heritage structures belong to importance classes III and IV, CSI defines three performance levels: - A1: Limited damage for seismic excitation with exceedance probability 10% in 50 years. - A2: Limited damage for seismic excitation with exceedance probability 50% in 50 years. - B1: Important damage for seismic excitation with exceedance probability 10% in 50 years. It is noteworthy that CSI does not consider the performance level "Near Collapse" as acceptable for important monumental structures. Greek Earthquake Planning and Protection Organization (EPPO) has more recently released a draft regulation: Code for the Assessment and Interventions of Masonry Structures (CASIM) aiming to establish criteria for the assessment of the bearing capacity of existing masonry structures [10], and a draft with specialized guidelines for monuments [11]. In general, CASIM follows the same principles and performance levels as CSI. 2.1.2. Eurocode EC8 Eurocode 8 -Part 1 [12] and Part 3 [2]- follows similar principles as CSI, while it provides an additional seismic hazard level: - Seismic excitation with exceedance probability 20% in 50 years. - Seismic excitation with exceedance probability 50% in 50 years. - Seismic excitation with exceedance probability 10% in 50 years. The target performance level results from the combination of acceptable damage level and seismic risk scenario, as well as the importance class of the structure. It has to be noted that EC8 does not refer with specific guidelines for the cases of high historical or artistic value monumental structures [7]. However, its principles can be followed for structural assessment and retrofitting in such cases as well. 2.1.3. FEMA 349 According to US guidelines FEMA 349, the following four performance levels are defined for masonry structures [4]: - Slight Damage State. - Moderate Damage State. - Extensive Damage State. - Complete Collapse.
Alexandros Liratzakis and Yiannis Tsompanakis
Similarly, to EC8, FEMA 349 adopts three seismic hazard levels [4]: - Seismic excitation with exceedance probability of 50% in 50 years. - Seismic excitation with exceedance probability of 10% in 50 years. - Seismic excitation with exceedance probability of 2% in 50 years. In addition, FEMA 349 [4] proposes limit drift values for each performance level. The proposed values in Table 1 change according to the construction materials and the norm under which the structure was designed. Note that for historic structures, the limit values of the URM buildings correspond to design level "Low-Code”.
Performance Level Special High - Code High – Code Moderate – Code Low – Code Pre – Code
Average Inter-Story Drift Ratio Structural Damage State Thresholds Capacity Curve Control Points (Fragility Medians) Yield Plastic Slight Moderate Extensive Complete 0.0057 0.1371 0.005 0.015 0.05 0.125 0.0038 0.0913 0.004 0.012 0.04 0.1 0.0029 0.0514 0.004 0.0099 0.0306 0.75 0.0019 0.0343 0.004 0.0099 0.0306 0.75 0.0019 0.0343 0.0032 0.0079 0.0245 0.06
Table 1 : Structural Damage State thresholds per Performance Level [4] 2.2 Limited Duration Rehabilitation Interventions Improvement of dynamic structural response by applying the so-called “Limited Duration Rehabilitation Interventions” (LDRI) [6, 7] aims to implement mitigation measures for a specified period and for a predefined limit state, after which a re-assessment of the building must be performed and depending on the results to revise the mitigation measures. According to this conceptual methodology, the time for which the operation ensures a predetermined performance level is defined as the nominal life of an intervention (Τ Δ). This methodology uses the following Equations for each of the three seismic hazard zones in Greece (Z1, Z2, Z3) for the calculation of the return period (T RL) with respect to reference peak ground acceleration (agR): (1) (2) (3) The code-imposed acceleration values agRL for of the three Greek seismic hazard zones are 0.16g, 0.24g and 0.36g, respectively [13]. The return period TRL related to the corresponding a gR is calculated using the proper attenuation relationship among (1) - (3), for which a 20% reduction, i.e., agR = 0.8agRL, is also considered [7]. Adopting a Poissonian distribution for the occurrence of seismic events, TΔ is related to the return period T RL and to the probability of occurrence PR as follows: (4)
If the seismic action is defined in terms of the reference peak ground acceleration a gR, the value of the importance factor γI multiplying the reference seismic action to achieve the same probability of exceedance in T Δ years as in the TΔR years for which the reference seismic action is defined, can be computed by: (5)
where exponential parameter k is in the order of 3 [12] and relates nominal life of the examined intervention with the importance class [7]. 3
CASE STUDY
Details for the examined URM building can be found in [8, 14], herein due to space limitations only a brief description is given. The mechanical characteristics of masonry walls were calculated according to EC6 [15], while the parabolic Drucker-Prager yield criterion [16] was used for the description of the inelastic behavior of the masonry walls. The structure is constructed on unsaturated silty sand, the mechanical properties of which
Alexandros Liratzakis and Yiannis Tsompanakis
with respect to the degree of saturation are taken from the study of Byun et al. [17]. Regarding the inelastic behavior of the soil, the Cam-Clay yield criterion according to the Critical State theory for unsaturated soils was used [18]. Numerical analyses were performed utilizing general purpose finite element software MSC Marc [19]. For the dynamic nonlinear analyses, the multiple-stripe dynamic analysis (MSDA) procedure was repeated for eight different soil saturation conditions (8%, 12%, 16%, 20%, 32%, 54%, 63% and 80%) and ten seismic intensity levels [8]. The twenty seismic records (Table 2) which have been used in this study were selected from PEER database [20] and were scaled utilizing EC8 [12] guidelines as implemented in ISSARS software [21]. Νο 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Region
Station name
Imperial Valley Mammoth Lakes Coalinga N. Palm Springs Chalfant Valley Loma Prieta Cape Mendocino Big Bear
Northridge
Magnitude
Bonds Corner El Centro Array #5 El Centro Array #7 El Centro Array #8 Convict Creek Pleasant Valley P.P North Palm Springs Whitewater Trout Farm Zack Brothers Capitola Gilroy Array #3 Rio Dell Overpass Big Bear Lake - Civic Beverly Hills Canyon Country LA Obregon Park Newhall - Fire Sta Pardee – SCE Rinaldi Receiving S. Monica City Hall
6.53 6.06 6.36 6.06 6.19 6.93 7.01 6.46
6.69
Epicentral distance (km) 6.2 27.8 27.64 28.09 1.43 9.98 10.57 4.24 14.33 9.78 31.4 22.64 10.15 13.39 26.49 39.39 20.27 25.65 10.91 22.45
PGA (g) 0.686 0.448 0.42 0.538 0.419 0.571 0.59 0.602 0.425 0.48 0.462 0.424 0.503 0.459 0.436 0.467 0.698 0.505 0.634 0.591
Table 2 : Characteristics of the ground motion records
a)
b) Figure 1. a) Model in its initial state; b) Retrofitted model with EPS geofoam
As a continuation of the previous study [8], the original building was slightly retrofitted with EPS blocks, with height equal to 1m and width 0.50m (Figure 1), which are placed at the exterior of the foundation, aiming to improve the seismic response of the building and to minimize the impact of soil saturation conditions. This costeffective intervention can enhance the dynamic behavior of the building, as this EPS layer acts as a damper (due to its high compressibility), absorbing most of the dynamic distress, thus, protecting the structure, especially for lower Sr(32%) the response of the structure is not improved in the first steps of record scaling. On the other hand, for higher seismic intensity levels, in all cases the application of this retrofitting scheme drastically improves the response of the structure, depending on the saturation level. In particular, the improvement of the response of the retrofitted structure is even more pronounced at lower saturation levels. In addition, the results of the original building seem to be grouped, i.e., to have slight variations for low (Sr=8% to 20%) and high (Sr=32% to 80%) saturation levels. This is due to the variation of soil stiffness for these soil conditions, since according to the experimental data [17], the impact of saturation level substantially affects the basic mechanical parameters of the soil in this specific manner. In contrast, the application of EPS geofoam at foundation level alleviates this scattering and groups the curves in a more uniform way irrespective of the saturation conditions.
Initial building
Retrofitted building
Figure 3. Comparison of initial and retrofitted model’s IDA curves. Regarding the calculations of nominal life of the retrofitted and initial models utilizing LDRI approach, it is assumed that the building is located at seismic zone Z2, thus, TRL return period is calculated via Equation (2). The nominal life for all examined models is calculated using Equation (4) and the results are presented in Table 3. As a consequence of the building’s response variation according to the degree of saturation, the nominal life of the structure varies considerably depending on the soil conditions. Τhe response of the initial building has proven to be directly dependent on the degree of soil saturation. More specifically, it was observed that the increase of soil saturation (Sr >32%) contributes to the increase of the building’s nominal life. When the building is founded on soil with a Sr
