%0 Journal Article %J Bulletin of Earthquake Engineering %D 2015 %T Analysis of first building retrofitted to EN-Eurocode 8 versus performance under near-design-level earthquake %A Fardis, MichaelN %A Liosatou, Eftychia %A Kosmopoulos, Antonis %K Eurocode 8 %K Nonlinear dynamic analysis %K Seismic assessment %K Seismic damage %K Seismic evaluation %K Seismic rehabilitation %K Seismic retrofitting %X A ground motion with a PGA of 0.39 g, above the 475-year design earthquake specified in the Greek seismic code and the National Annex to Eurocode 8, was recorded about 100 m from the municipal theatre of Argostoli, the capital of the Ionian island of Kefalonia. The building, the largest in volume in the town, was designed and constructed to the 1959 seismic code for lateral loads (about 30 % of those specified today). Suffering also from severe reinforcement corrosion and poor concrete quality, it provided the ground to apply for the first time (in 2005) Part 3 of Eurocode 8 for seismic assessment and retrofitting. The assessment based on nonlinear dynamic analysis showed that the buidling was deficient even under ground motions with PGA less than 0.05 g. Alongside the need to tackle corrosion, the design of the retrofitting had to face several constraints and requirements: minimal disruption of the facility’s operation, accessibility to the foundation only along the building perimeter, no change to the façade and minimal intervention to the other lateral sides. A very cost-effective retrofitting was designed, employing: (a) one-sided RC or fibre-reinforced polymer (FRP) overlays and two new walls at the perimeter, (b) connection of the two structurally independent and torsionally imbalanced units of the building in a single structure, and (c) one-sided FRP jacketing of selected interior elements. Less than 7 years after the works were completed, the building went unscathed through a Magnitude 6.1 earthquake. Serious damage was limited to four nonstructural masonry infill panels at the penthouse, which, according to the retrofit design, could be sacrificed, so as to protect vulnerable structural elements that could not be retrofitted in the first place. Nonlinear dynamic analyses of the response of the retrofitted building to the recorded ground motions are in agreement with the observed performance. They show also that, had the as-built facility not been retrofitted, it might have collapsed in the earthquake. %B Bulletin of Earthquake Engineering %V 13 %P 2590 %G eng %U http://dx.doi.org/10.1007/s10518-015-9740-3 %N 9 %& 2567 %R 10.1007/s10518-015-9740-3 %0 Conference Paper %B 10th Asia-Oceania Symposium on Fire Science and Technology %D 2015 %T An Analytical Method for the Estimation of Radiation Heat Flux from Open pool Fires and pool Fires Im-pinging on Ceilings %A Vouros, Α %A Delichatsios, ΜΑ %A Zhang, J %K Ceiling %K Simplified modelling %K Thermal radiation %X

This paper presents a new integral method for the estimation of the thermal radiation flux from open pool in air fires and pool fires having flames sometimes impinging under extensive ceilings. For open fires, a uniformly distributed emissive power to the cylinder that represents the flame envelop is used together with a radiant fraction that depends on the source diameter. Estimations and comparisons are produced using the mean as well as the continuous (stoichiometric) flame heights which is calculated and correlated numerically including flames impinging on the ceiling. The applicability of the method and comparison with previous analytical methods is validated for a range of heat release rates (1 to 300 kW) using experimental results of the literature. For fires under ceilings, the gas temperature distribution under the ceiling is used to estimate the thermal radiation from a disc that represents the flames if flames impinge on the ceiling as well the hot gases in all the ceiling layer. Radiation flux from the ceiling jet is significant in this case, while the lower and higher values are estimated for the complicated view factor, assuming that the flame cylinder under the ceiling is opaque or transparent. Estimations are compared with experimental results of the literature, indicating that the adoption of a transparent flame cylinder produces conservative results for all the cases examined.

%B 10th Asia-Oceania Symposium on Fire Science and Technology %C Tsukuba, Japan %G eng %0 Conference Paper %B fib Symposium: “Innovation and Design” %D 2015 %T First building retrofitted to EN-Eurocode 8 tested by near-design-level-earthquake %A Fardis, ΜΝ %A Liossatou, E %A Kosmopoulos, A %X

The theatre building in the main town of Kefalonia, designed to the 1959 seismic code for lateral loads about 30% of those specified today and suffering from reinforcement corrosion, was apparently the first application of Part 3 of Eurocode 8 for seismic assessment and retrofitting. Nonlinear dynamic analyses have shown that it could not survive earthquakes with peak ground acceleration (PGA) of about 0.05g. The retrofitting faced several constraints: minimal disruption of operation, access to the foundation only at the perimeter, no change to the façade and minimal intervention to the other sides, by employing at the perimeter one-sided RC or FRP overlays and two new walls, connection of the two structurally independent and torsionally imbalanced units of the building into a single one and one-sided FRP jackets at selected interior elements. Less than seven years later the building was subjected to an earthquake with a PGA of 0.39g. Damage was serious only in nonstructural masonry at the penthouse. Nonlinear dynamic analyses of the retrofitted or the original building agree with the observed performance and show that, had the building not been retrofitted, it would have collapsed.

%B fib Symposium: “Innovation and Design” %C Copenhagen, Denmark %G eng %0 Journal Article %J Bulletin of Earthquake Engineering %D 2015 %T Hybrid Simulation of Bridge Pier Uplifting %A Stathas, N %A Skafida, S %A Bousias, SN %A Fardis, MN %A Digenis, S %A Palios, X %K bridge piers %K distributed testing %K Foundation uplifting %K Hybrid simulation %X

Substructured Pseudodynamic (PsD) testing is used to study the transverse seismic response of a bridge with box-girder deck monolithically connected
to circular piers, but free to move laterally at the abutments. The tests are of two types, each one with two variants: with the pier footing rocking on elastic soil and
uplifting from it, or fixed to rigid ground. The first type concerns a two-span deck connected monolithically to a central pier and supported at each end on the abutment
through a pair of elastomeric bearings, which are stiff and strong in compression but very soft and weak in tension. The second one concerns the interior
piers of multi-span bridges having all spans equal and all piers similar. The pier was physically tested at 1:2 scale; everything else was numerically simulated
online with Opensees. The deck, the soil and the bearings were modeled as elastic, but not allowing tension to develop at the soil-footing interface, and the bearings
two orders of magnitude more flexible in tension than in compression. In the first type of tests, at a peak ground acceleration (PGA) of 0.15g, the footing uplifted,
the bearings developed tension, but the pier stayed elastic. In the second type of tests, with a PGA of 0.4g, uplifting of the footing did not prevent plastic hinging at
the pier base, but reduced significantly the pier damage. A cyclic quasi-static test in the end shows the deformation capacity margins of the pier.

%B Bulletin of Earthquake Engineering %G eng %0 Conference Paper %B MCS-2015: Ninth Mediterranean Combustion Symposium %D 2015 %T Numerical and experimental validation study of flame extent of a pool fire under the ceiling %A Wang, Y %A Vouros, A %A Zhang, J %A Delichatsios, M %X

The purpose of this work is to establish dimensionless correlations for the flame extent under a ceiling to be used for large scale calculations of radiative heat fluxes for which numerical simulations may be prohibitive and probably, not credible. Towards this objective, this paper presents a numerical study of the flame height in an open pool fire and the subsequent flame extent under the ceiling. Fire dynamics simulator (FDS, version 6) was applied. The stoichiometric mixture faction was used to define the continuous flame height which also coincides with the maximum mean temperature on the axis of the fire. The size of the pool fire was 0.1m × 0.1 m having heat release rates (HRRs) from 2 to 100 kW for two heights 30 and 50 cm. The predicted free flame height and horizontal flame extent are plotted as functions of HRR and compared to correlations and experimental data in the literature. Because most of the correlations and measurements are based on the mean (visual) flame height, the predicted continuous flame height is found to be 3/5 of the experimentally mean flame height. It is found that the predicted flame height and horizontal flame extent agree well with existing experimental relations. However, some difference was noted in the horizontal flame extent indicating that the relation between the mean and continuous flame heights derived for an open free pool fire may no longer apply due to the presence of the ceiling affecting the entrainment and the turbulence of the ceiling flame jet.

%B MCS-2015: Ninth Mediterranean Combustion Symposium %C Rhodes, Greece %G eng %0 Conference Paper %B SMAR-2015 3rd Conference on Smart Monitoring, Assessment and Rehabilitation of Civil Structures %D 2015 %T Properties of RC walls produced by infilling a frame with concrete for seismic rehabilitation %A Biskinis, D %A Psaros-Andriopoulos, A %A Fardis, MN %X

In seismic retrofitting of concrete buildings, frame bays are converted to reinforced concrete (RC) walls by infilling the space between the frame members with RC of a thickness of not more than their width. The cyclic behavior of the resulting wall depends on the connection between the RC infill and the surrounding RC members. The paper uses results from 48 cyclic tests of such composite walls to express their properties in terms of the geometry, the reinforcement and the connection. Properties addressed are: a) the yield moment at the story base; b) the secant-to-yield-point stiffness over the shear span of the wall in a story; c) the deflection at flexural failure in cyclic loading; d) the cyclic shear resistance, including sliding shear. Separate models are given for squat walls failing in shear. The models are modifications of the ones developed in the past for monolithic RC walls from several hundred cyclic tests.

%B SMAR-2015 3rd Conference on Smart Monitoring, Assessment and Rehabilitation of Civil Structures %C Antalya, Turkey %G eng %0 Journal Article %J Bulletin of Earthquake Engineering %D 2015 %T Strength, deformation capacity and failure modes of RC walls under cyclic loading %A Grammatikou, S %A Biskinis, D %A Fardis, MN %K Concrete walls %K Cyclic loading %K Deformation capacity %K Ductility %K Eurocode 8 %K RC walls %K Seismic assessment %K Seismic design %K Seismic evaluation %K Shear strength %K Ultimate deformation %X

Past models for the cyclic strength and deformation capacity of reinforced concrete walls are evaluated and new ones are developed using results from 866 wall tests. On the basis of the observed damage, the failure mode is classified into the following categories: in flexure, in diagonal tension, in diagonal compression, or in sliding shear. Among the past models evaluated are those proposed by two of the authors: (a) for the cyclic strength in diagonal tension after flexural yielding or in diagonal compression and (b) for the ultimate chord rotation capacity under cyclic flexure, which were adopted in Eurocode 8-Part 3 and in fib Model Code 2010. Walls with height-to-length ratio\1.2 (‘‘squat’’) are considered separately: past models are evaluated and two new ones are proposed on the basis of 321 cyclic tests. Past models for sliding shear strength are modified, using the results from 55 cyclic tests. The predictions for the failure mode and the cyclic strength and/or deformation capacity agree well with the tests.

%B Bulletin of Earthquake Engineering %G eng %U http://link.springer.com/article/10.1007/s10518-015-9762-x %R 10.1007/s10518-015-9762-x %0 Journal Article %J Earthquake Engineering and Structural Dynamics %D 2015 %T Uplift of deck or footings in bridges with distributed mass subjected to transverse earthquake %A Fardis, MN %K bridges %K continuous mass %K distributed mass %K eigenvalues %K uplifting %X

The eigenvalue problem is analytically formulated in symmetric bridges with distributed mass and moment of inertia under transverse earthquake. The piers are elastically supported on the ground. The deck is monolithically connected to one or two piers for all degrees of freedom and restrained or transversely free at the abutments. The characteristic equation, symmetric normal modes, modal participation factors, and participating mass ratios are given analytically. The problem is expressed in terms of few dimensionless parameters: (i) the radius of gyration of the deck mass divided by the pier height; (ii) the ratio of the rotational stiffness of a footing to that of the pier at the base; (iii) the ratio of flexural stiffness of the outer spans to those of the pier; (iv) the ratio of torsional stiffness of side spans to the rotational stiffness of the pier top; (v) for two piers, the side-to-central-span ratio. Modal response spectrum analysis gives the moment at the base of the footings and the torque in the deck at its supports on the abutments as ratios to the values at incipient uplifting from the ground or the bearings. The peak ground acceleration of the motion at the onset of either one of these two types of nonlinearity is depicted as a function of the dimensionless parameters and the fundamental period of an elastic deck supported only at the abutments, or of a rigid deck on piers fixed against rotation at top and bottom.

%B Earthquake Engineering and Structural Dynamics %G eng %U http://onlinelibrary.wiley.com/doi/10.1002/eqe.2607/abstract %R 10.1002/eqe.2607 %0 Conference Paper %B 8th International Conference on Analytical Models and New Concepts in Concrete and Masonry Structures AMCM 2014 %D 2014 %T RC members in cyclic loading: strength, deformation capacity, failure modes %A Fardis, MN %A Biskinis, D %A Grammatikou, S %X

A portfolio of simple, practical models is presented for a) the resistance of a RC member in flexure or shear under cyclic loading; b) its deformation at flexural yielding and its secant-to-yield-point stiffness; and c) the member ultimate deformation in flexure or shear under cyclic loading. They are based on a databank of several thousand tests of RC members - the largest in the world. They cover beams, rectangular columns or walls, members with circular, T-, H-, U- or hollow rectangular section, with or without detailing for earthquake resistance, with bars ribbed or smooth, continuous or lap-spliced in the plastic hinge. Most of the models, or earlier versions based on a smaller data-bank, were adopted in Eurocode 8-Part 3 and/or fib MC2010, for performance- and displacement-based seismic design, assessment or retrofitting. So, the paper serves as a background document for the models adopted in these codes, improving and extending them, or presenting new ones.

%B 8th International Conference on Analytical Models and New Concepts in Concrete and Masonry Structures AMCM 2014 %C Wrocław, Poland %G eng %0 Journal Article %J Earthquake Engineering and Structural Dynamics %D 2014 %T Residual displacements of RC structures as SDOF systems %A Liossatou, Ε %A Fardis, MN %K concrete structures %K hysteresis models %K permanent seismic displacements %K pulse-type ground motion %K residual seismic displacements %K seismic response spectra %X

Residual displacements are sensitive to ground motion details, hence more random than peak inelastic displacements. Among the factors with systematic impact on residual displacements, the post-yield-stiffness-ratio has been studied thoroughly; its effects are not investigated further. Concerning another important factor, the hysteresis law, past studies have focused on the bilinear model, which does not represent concrete structures. Residual displacements from nonlinear response-history analyses of bilinear systems are compared to those from models tuned to concrete structures, conforming to modern codes, deficient or intermediate. Deficient-type structures, with their narrow, almost self-centering hysteresis loops, develop markedly smaller residual displacements than those with stable energy-dissipating behavior. A velocity pulse in the motion increases peak inelastic and residual displacements by about the same proportion. As a fraction of the peak inelastic or spectral displacement, residual displacements are on average almost independent of the period and increase when the lateral strength ratio increases, reaching a limit at a lateral strength ratio of 2 to 5. Peak inelastic displacements are a better basis for estimation of residual displacements than spectral ones: the ratio of the two is almost independent of the period, the lateral strength ratio (beyond values of 2 to 3) and velocity pulses. The spectrum of the ratio of residual displacement to peak inelastic or spectral displacement is considered as a random process of period; its mean and variance functions, marginal probability distributions and autocorrelation functions are given in terms of the lateral strength ratio, the hysteresis model and the presence of a velocity pulse.

%B Earthquake Engineering and Structural Dynamics %V 44 %P 713-734 %G eng %U http://onlinelibrary.wiley.com/doi/10.1002/eqe.2483/abstract %N 5 %R DOI: 10.1002/eqe.2483 %0 Conference Paper %B 4th International fib Congress and Exhibition %D 2014 %T Strength, deformation capacity and failure mode of RC walls in cyclic loading %A Grammatikou, S %A Biskinis, D %A Fardis, MichaelN %X A database of 621 cyclic tests of RC walls is utilized to evaluate past models for the cyclic strength and deformation capacity of the wall and to develop/calibrate new ones. From the observed damage the failure mode is classified as in flexure, diagona l tension or compression before or after flexural yielding, or in sliding shear. Past models being evaluated on the basis of the tests include models proposed by two of the authors and adopted in Eurocode 8-Part 3 and/or MC2010 for (a) flexural strength, (b) the cyclic shear strength after flexural yielding (as affected by the imposed ductility demand) and (c) the cyclic chord rotation capacity in flexure. Walls with height-to-length ratio less than 1.2 are considered separately, as their shear failure requires different models: past models are commented and a new one proposed and calibrated on the basis of 130 cyclic tests of squat walls. Past models for sliding shear strength are evaluated and modified on the basis of 29 cyclic tests with that failure mode: a new model is proposed for the possibility of uncontrolled sliding during a load reversal at a point in time when the flexural crack at the base is open throughout the section and only dowel action of the still elastic vertical bars is available to resist the shear force. There is good agreement of the predicted cyclic strength and/or deformation capacity per failure mode; the prediction of the most likely mode in the tests of the database is also satisfactory. %B 4th International fib Congress and Exhibition %C Mumbai, India %G eng %M paper no. 408