Download (3MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Lingua: English
Miano, Andreaandrea.miano@unina.it
Date: 10 December 2017
Number of Pages: 197
Institution: Università degli Studi di Napoli Federico II
Department: dep25
Dottorato: phd048
Ciclo di dottorato: 30
Coordinatore del Corso di dottorato:
Rosati, Lucianorosati@unina.it
Jalayer, FatemehUNSPECIFIED
Date: 10 December 2017
Number of Pages: 197
Uncontrolled Keywords: Performance base earthquake engineering; Seismic fragility; Cloud Analysis; Cloud to IDA; Structural modeling uncertainties; Retrofit; Life cycle cost analysis
Settori scientifico-disciplinari del MIUR: Area 08 - Ingegneria civile e Architettura > ICAR/09 - Tecnica delle costruzioni
Additional Information: Mail alternativa: andrea.miano@live.it
Date Deposited: 01 Jan 2018 11:47
Last Modified: 02 Apr 2019 10:42
URI: http://www.fedoa.unina.it/id/eprint/12126


The scope of this thesis is to propose a journey through probabilistic performance based assessment and retrofit design based on nonlinear dynamic analysis tools. The thesis aims to address the performance-based assessment paradigm by developing seismic fragilities and earthquake loss estimation. The “Performance-based earthquake engineering” (PBEE) for design, assessment and retrofit of building structures seeks to enhance seismic risk decision-making through assessment and design methods that have a strong scientific basis and support the stakeholders in making informed decisions. The PBEE is based on a consistent probabilistic methodological framework in which the various sources of uncertainty in seismic performance assessment of structures can be represented. The methodology can be used directly for performance assessment, or can be implemented for establishing efficient performance criteria for performance-based design. In particular, the PBEE aims to maximize the utility for a building by minimizing the expected total cost due to seismic risk, including the costs of construction and the incurred losses due to future earthquakes. The PBEE advocates substituting the traditional single-tier design against collapse and its prescriptive rules, with a transparent multi-tier seismic design, meeting more than one discrete “performance objective” by satisfying the corresponding “performance level” (referred to as the “limit state” in the European code) expressed in terms of the physical condition of the building as a consequence of an earthquake whose intensity would be exceeded by a mean annual rate quantified as the “seismic hazard level”. In other words, PBEE distinguishes itself from the prescriptive requirements of the traditional building codes by envisioning explicit verification of satisfying various performance objectives. Last but not least, the PBEE is fundamental to seismic assessment of existing buildings, seen as an indispensable step in the seismic retrofit design process. The principal elements of the PBEE procedure can be summarized as description, definition, and quantification of earthquake intensity measures, engineering demand parameters, damage measures, and decision variables. The process encompasses the following steps: (1) calculation of ground motion hazard by representing the uncertainty in ground motion with a probabilistic model for a parameter (or vector of parameters) related to ground motion and known as the intensity measure (IM); (2) estimation of the uncertainty in structural response expressed as a group of engineering demand parameters EDP (e.g., force and deformation-related engineering parameters) conditioned on each IM level; (3) estimation of the uncertainty in damage measure DM (i.e., physical states of damage, that describes the condition of the structure and its components) conditioned on the EDP and IM; (4) estimation of the uncertainty in the decision variable DV expressing the decision-related consequences (e.g., financial losses, fatalities, business interruption, etc.) given DM, EDP and IM. One interesting and useful characteristic of the PBEE procedure is that any of the above-mentioned intermediate steps can be collapsed. For example, the damage measure DM can be conditioned directly on intensity measure IM by collapsing the intermediate step related to the engineering demand parameter EDP. It is important to note that the performance levels should ideally be described in terms of the decision variable(s) DV. However, many modern codes and guidelines express the various discrete performance levels in terms of the incurred damage (i.e., DM). An important focus in this thesis is dedicated to the estimation of the conditional probability of exceeding a damage measure DM expressed as the critical demand to capacity ratio throughout the structure and a given ground motion time-history and relating it directly to IM (by collapsing the intermediate EDP step). The conditional probability of exceeding a given level of DM given IM can be expressed as the structural fragility for a given performance level. In fact, the assessment of analytic structural fragility for existing buildings is one of the fundamental steps in the modern performance-based engineering. In general, methods for assessing the structural fragility for a given performance level or limit state range from the simplest methods based on the response of an equivalent single-degree-of-freedom (SDOF) model to complex nonlinear dynamic analysis procedures performed for a structural model subjected to a set of ground-motion records. In the past fifteen years, many research efforts have been dedicated to an in-depth study of the implementation, the nuances and the potential complications of non-linear dynamic analysis procedures. These efforts have led to different methodologies such as Incremental Dynamic Analysis (IDA), Multiple-Stripe Analysis (MSA) with conditional mean spectrum, and Cloud Analysis. This work focuses on the non-linear dynamic analysis procedure known as the Cloud Analysis. This analysis is based on fitting a linear regression model in the logarithmic scale to the pairs of structural response parameter (e.g., maximum inter-story drift) and IM (e.g., first-mode spectral acceleration) for a suite of as-recorded ground motions. This method is well-known both for the simplicity of its underlying formulation and for the relatively small number of structural analyses required. However, the Cloud Analysis is also notorious for being based on a few simplifying assumptions (fixed standard error of regression, mean response varying linearly as a function of IM in the logarithmic scale, and structural response given IM being modeled as a Lognormal distribution), and for being sensitive to the selected suite of records. A functional variation to the original Cloud Analysis is presented in order to take into account the cases leading to structural collapse. Moreover, to reduce record-selection-dependence of the results, a Bayesian version of the Cloud Analysis considering the “collapse-cases” is presented in which the uncertainty in the structural fragility model parameters is considered. This leads to a Robust Fragility estimate and a desired confidence interval defined around it. The entire method is based on the adoption of a normalized demand to capacity ratio as the damage measure/decision performance variable. Herein, as said, a normalized demand to capacity ratio coined as “critical demand to capacity ratio” and denoted as DCR, takes the structure closest to the onset of a prescribed limit state LS, is adopted. The adoption of DCRLS as performance variable is also central to a new nonlinear dynamic analysis procedure referred to as “Cloud to IDA” that exploits the Cloud Analysis to perform IDA in a more efficient manner. Evaluation of structural behaviour under seismic actions for an existing building encompasses the consideration of numerous sources of uncertainty associated with the seismic action and the structural modelling. In the past decades, significant research efforts have been carried out and substantial progress has been made towards the consideration of various sources of uncertainty into structural performance assessment and design frameworks. Several alternative methods have been proposed that combine reliability methods such as the first order second moment (FOSM and MVFOSM) methods, response surface methods, simulation-based methods (e.g., Monte Carlo, Latin Hypercube Sampling) with non-linear dynamic procedures such as IDA based on recorded ground motions in order to take into account sources of uncertainties other than record-to-records variability. This thesis aims to quantify the impact of structural modeling uncertainties on the seismic performance assessment for an existing case-study building. Herein, the proposed version of the Cloud Analysis, considering the collapse cases, is implemented to consider the record-to-record variability, the structural modeling uncertainties and also the uncertainties in the parameters of the adopted fragility model, through a Bayesian procedure. The presented procedure can lead to reliable results with a considerably lower computational effort in comparison to the methods available in literature. Finally, the PBEE methodology is implemented for the case-study building in order to choose the most appropriate seismic retrofit design that maximizes the utility (by minimizing the expected costs) and satisfies the safety-checking for three different performance levels. To this end, the non-ductile older RC frame of the case-study is retrofit designed based on different strategies aimed to improve the seismic performance of the frame. The case-study moment resisting frame is modeled using structural elements with fiber sections in order to take into account the flexural-axial interactions. Furthermore, the flexural-axial-shear interactions and the fixed end rotations due to bar slip in the columns are considered by adding zero-length springs to column ends. The performance-based safety-checking procedure is based on the Demand and Capacity Factored Design (DCFD) format. Amongst the viable retrofit designs that satisfy the risk-related safety-checking DCFD criteria, the one that corresponds to the minimum expected loss over the life cycle of the building is identified.

Actions (login required)

View Item View Item