This work adopts a probabilistic evaluation approach to investigate the effectiveness of isolation devices for bridges in terms of seismic performance, vulnerability and expected life-cycle cost-benefit. A novel procedure to evaluate a reliable structure-based IM for isolated bridges and an improved life-cycle cost analysis formulation with respect to the existing ones are the two main original contributions. This dissertation has an assessment approach, so for each step some assumptions on the design of intervention, types of modelling and analysis have been introduced. No design optimization is carried out since it was not the purpose of this work, however the assumptions are based on the state of the art and practice and they will be clearly explained together with the limitations that eventually result from them. In order to achieve the purposes, an existing bridge has been selected as case study to take into account the complexity of a real structure. Even though a single case study bridge can restrict the generality of the numerical results, the main contributions mentioned previously consist in procedures that are not conditioned on the case study and that can be readily applied to other bridges. Damage to bridges during an earthquake event can lead to significant service breaks in the transportation system, causing primarily difficulties to the emergency operations. The main consequences due to bridge failure are a potential huge human's life loss and in addiction a wide economic impact on the transportation network, represented mainly by direct repair costs of intervention and indirect costs due to the loss of functionality of the bridge during repair. With specific reference to the Italian transportation network, the majority of the bridges was built between 1960 and 1980, consequently these structures are to date suffering structural deterioration and a large number of them was built following antiquated design standards with deficient or missing design criteria against seismic actions, therefore the issue of retrofitting of bridges assumes a key role, and it needs to be addressed with also reference to the Life-Cycle Cost (LCC) analyses. Between all the several design and retrofit strategies for improving the resistance of bridges to earthquakes, the seismic isolation is nowadays an effective choice for the protection of bridges that has been adopted in bridge design or retrofit for over 35 years in the United States and more recently it has been increasingly adopted also in Italy, especially towards the application of elastomeric bearings and friction-pendulum devices. The modern design philosophies, based on probabilistic performance-based earthquake engineering (PBEE) approaches, provides useful tools to identify the best retrofits for non-seismically designed bridges not only in terms of vulnerability assessment but also in order to achieve goals such as risk mitigation or minimization of economic loss. A primary objective of this work is the effectiveness evaluation of seismic protection devices for bridges following the probabilistic Intensity Measure (IM) based approach developed by the Pacific Earthquake Engineering Research (PEER). In fact, if IM-based approaches are well established and widely studied for bridges and buildings, there has been a very limited research to date regarding the performance assessment of bridges for evaluating the effectiveness of seismic isolation devices. This matter is then considered an actual topic implying a number of additional issues with respect to the case of non-isolated bridges. The elastomeric bearings (ERB) and the friction pendulum system (FPS) are here considered as isolation solutions, and they are applied to an existing railway bridge as case study. The bridge has a continuous five-span steel truss deck with a total length of about 500 m carried by four concrete piers with height ranging from 50 to 130 m. Geometry, loads, structural materials and existing bearings are investigated in order to design and estimate the retrofit interventions in an executable manner which can actually be put into practice. The structural modelling is conducted by developing three-dimensional finite element (FE) models of three bridge configurations (as-built, with ERB and with FPS) and subjecting them to a suite of 80 recorded ground motions with a wide range of spectral properties that are appropriate for isolated bridges. The FE models are developed in OpenSees employing fibre beam column elements for bridge piers and bilinear hysteretic elements for isolation devices. The influence of isolation on the demand for various critical bridge elements is evaluated through the development Probabilistic Seismic Demand Models (PSDMs) with 'cloud' approach to derive analytical fragility functions by nonlinear time history analysis (NLTHA) of the models. Peak ground acceleration (PGA) and Spectral Acceleration (Sa) calculated at different periods are adopted and compared as intensity measures (IMs) in terms of efficiency and sufficiency. To deal with the issue of adopting a reliable structure-based IM for isolated bridges, a novel procedure is introduced for the evaluation of the most appropriate period Ts which makes Sa(Ts) a reliable IM by maximizing its correlation to different components of a complex structure. The proposal of a new property for the IM, additional to efficiency and sufficiency, is addressed. Moreover, after the definition of appropriate limit states, the analysis of vulnerability at component level is addressed, followed by the evaluation of the effectiveness of isolation in terms of total probabilities of failure after the convolution with a seismic hazard coherently evaluated with respect to the selected ground motion set. To prevent high level of damage, both isolation systems give better protection in small piers than high ones, while they give more benefits in high pier for slight level of damage. The ERB results to be the most efficient to reduce the expected damage in the piers' base, however in terms of probability of damage at 1/3 height of the pier the effect of the two isolation systems are comparable. The FPS isolation is more efficient for the small piers than higher ones, for all the limit states. Moreover, the ERB provides a more uniform effect on the piers and better results on high piers than FPS. As mentioned above, life-cycle cost (LCC) analysis for bridges has gained widespread interest in recent years. Nevertheless, the effect of adopting seismic isolation devices for existing or new bridges, needs to be correctly addressed in terms of costs. The lack of knowledge regarding the correct estimation of LCC in presence of these kinds of devices needs to be covered by research, since in literature there are only few examples on how the isolation systems are producing cost-effective solutions for bridge owners. In order to give a contribution in this direction, this work provides an insight regarding the damage restoration of bridge components. The nominal retrofitting costs (initial cost in case of intervention), restoration costs (due to the possible damages in bearings and piers) and indirect costs (due to the loss of functionality of the bridge during repair) are estimated in an executable manner. Finally statistical moments of seismic losses, such as the expected value and variance, are calculated for the three examined bridge configurations by different life-cycle cost formulations (Wen and Kang, 2001, Beck et al., 2002, Wen et al., 2003, Ghosh and Padgett, 2011). The proposal of an improved LCC formulation with particular attention to the issue of a correct evaluation of discount functions to commutate future costs into present values is presented. The benefits of isolation in terms of expected costs are calculated comparing the different solutions.

Performance-based seismic assessment for life-cycle cost analysis of existing bridges retrofitted with seismic isolation / Sebastiani, PAOLO EMIDIO. - ELETTRONICO. - (2016).

Performance-based seismic assessment for life-cycle cost analysis of existing bridges retrofitted with seismic isolation

SEBASTIANI, PAOLO EMIDIO
01/01/2016

Abstract

This work adopts a probabilistic evaluation approach to investigate the effectiveness of isolation devices for bridges in terms of seismic performance, vulnerability and expected life-cycle cost-benefit. A novel procedure to evaluate a reliable structure-based IM for isolated bridges and an improved life-cycle cost analysis formulation with respect to the existing ones are the two main original contributions. This dissertation has an assessment approach, so for each step some assumptions on the design of intervention, types of modelling and analysis have been introduced. No design optimization is carried out since it was not the purpose of this work, however the assumptions are based on the state of the art and practice and they will be clearly explained together with the limitations that eventually result from them. In order to achieve the purposes, an existing bridge has been selected as case study to take into account the complexity of a real structure. Even though a single case study bridge can restrict the generality of the numerical results, the main contributions mentioned previously consist in procedures that are not conditioned on the case study and that can be readily applied to other bridges. Damage to bridges during an earthquake event can lead to significant service breaks in the transportation system, causing primarily difficulties to the emergency operations. The main consequences due to bridge failure are a potential huge human's life loss and in addiction a wide economic impact on the transportation network, represented mainly by direct repair costs of intervention and indirect costs due to the loss of functionality of the bridge during repair. With specific reference to the Italian transportation network, the majority of the bridges was built between 1960 and 1980, consequently these structures are to date suffering structural deterioration and a large number of them was built following antiquated design standards with deficient or missing design criteria against seismic actions, therefore the issue of retrofitting of bridges assumes a key role, and it needs to be addressed with also reference to the Life-Cycle Cost (LCC) analyses. Between all the several design and retrofit strategies for improving the resistance of bridges to earthquakes, the seismic isolation is nowadays an effective choice for the protection of bridges that has been adopted in bridge design or retrofit for over 35 years in the United States and more recently it has been increasingly adopted also in Italy, especially towards the application of elastomeric bearings and friction-pendulum devices. The modern design philosophies, based on probabilistic performance-based earthquake engineering (PBEE) approaches, provides useful tools to identify the best retrofits for non-seismically designed bridges not only in terms of vulnerability assessment but also in order to achieve goals such as risk mitigation or minimization of economic loss. A primary objective of this work is the effectiveness evaluation of seismic protection devices for bridges following the probabilistic Intensity Measure (IM) based approach developed by the Pacific Earthquake Engineering Research (PEER). In fact, if IM-based approaches are well established and widely studied for bridges and buildings, there has been a very limited research to date regarding the performance assessment of bridges for evaluating the effectiveness of seismic isolation devices. This matter is then considered an actual topic implying a number of additional issues with respect to the case of non-isolated bridges. The elastomeric bearings (ERB) and the friction pendulum system (FPS) are here considered as isolation solutions, and they are applied to an existing railway bridge as case study. The bridge has a continuous five-span steel truss deck with a total length of about 500 m carried by four concrete piers with height ranging from 50 to 130 m. Geometry, loads, structural materials and existing bearings are investigated in order to design and estimate the retrofit interventions in an executable manner which can actually be put into practice. The structural modelling is conducted by developing three-dimensional finite element (FE) models of three bridge configurations (as-built, with ERB and with FPS) and subjecting them to a suite of 80 recorded ground motions with a wide range of spectral properties that are appropriate for isolated bridges. The FE models are developed in OpenSees employing fibre beam column elements for bridge piers and bilinear hysteretic elements for isolation devices. The influence of isolation on the demand for various critical bridge elements is evaluated through the development Probabilistic Seismic Demand Models (PSDMs) with 'cloud' approach to derive analytical fragility functions by nonlinear time history analysis (NLTHA) of the models. Peak ground acceleration (PGA) and Spectral Acceleration (Sa) calculated at different periods are adopted and compared as intensity measures (IMs) in terms of efficiency and sufficiency. To deal with the issue of adopting a reliable structure-based IM for isolated bridges, a novel procedure is introduced for the evaluation of the most appropriate period Ts which makes Sa(Ts) a reliable IM by maximizing its correlation to different components of a complex structure. The proposal of a new property for the IM, additional to efficiency and sufficiency, is addressed. Moreover, after the definition of appropriate limit states, the analysis of vulnerability at component level is addressed, followed by the evaluation of the effectiveness of isolation in terms of total probabilities of failure after the convolution with a seismic hazard coherently evaluated with respect to the selected ground motion set. To prevent high level of damage, both isolation systems give better protection in small piers than high ones, while they give more benefits in high pier for slight level of damage. The ERB results to be the most efficient to reduce the expected damage in the piers' base, however in terms of probability of damage at 1/3 height of the pier the effect of the two isolation systems are comparable. The FPS isolation is more efficient for the small piers than higher ones, for all the limit states. Moreover, the ERB provides a more uniform effect on the piers and better results on high piers than FPS. As mentioned above, life-cycle cost (LCC) analysis for bridges has gained widespread interest in recent years. Nevertheless, the effect of adopting seismic isolation devices for existing or new bridges, needs to be correctly addressed in terms of costs. The lack of knowledge regarding the correct estimation of LCC in presence of these kinds of devices needs to be covered by research, since in literature there are only few examples on how the isolation systems are producing cost-effective solutions for bridge owners. In order to give a contribution in this direction, this work provides an insight regarding the damage restoration of bridge components. The nominal retrofitting costs (initial cost in case of intervention), restoration costs (due to the possible damages in bearings and piers) and indirect costs (due to the loss of functionality of the bridge during repair) are estimated in an executable manner. Finally statistical moments of seismic losses, such as the expected value and variance, are calculated for the three examined bridge configurations by different life-cycle cost formulations (Wen and Kang, 2001, Beck et al., 2002, Wen et al., 2003, Ghosh and Padgett, 2011). The proposal of an improved LCC formulation with particular attention to the issue of a correct evaluation of discount functions to commutate future costs into present values is presented. The benefits of isolation in terms of expected costs are calculated comparing the different solutions.
2016
File allegati a questo prodotto
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/874444
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact