Earthquake Engineering and Structural Dynamics

Capecchi, Danilo; Vestroni, Fabrizio

doi: 10.1002/(SICI)1096-9845(199905)28:5<447::AID-EQE812>3.0.CO;2-2pmid: N/A

The present work evaluates the possibility of using dynamic data to assess structural integrity. It addresses the problem of understanding when it is sufficient to measure and use only natural frequencies, thus avoiding the need to measure modal shapes. The classic problem of detecting damage in beams, or beam assemblies, due to concentrated cracks, or damage spread over a structural member is dealt with. Damage is represented as a decrease in stiffness and linear behaviour before and after the event assumed to have caused damage is considered. Damage is restricted to a few unknown sections or elements, so that only the modification of few parameters of the system need to be determined. This study thus rejects assumptions unrelated to the physical aspects of the problem, in contrast to many papers on the subject. The amount of data to locate and quantify damage correctly is discussed; general considerations lead to the conclusion that a unique and reliable estimate of the damage can be obtained using only few additional frequency data with respect to the number of damaged zones. Continuous and discrete (finite element) models are examined. Finally the paper considers the applications to both analytical and experimental data of the procedure developed, which takes account of the peculiar characteristics of damage detection problem. Copyright © 1999 John Wiley & Sons, Ltd.

Rodriguez, Mario E.; Aristizabal, Juan C.

doi: 10.1002/(SICI)1096-9845(199905)28:5<463::AID-EQE818>3.0.CO;2-Vpmid: N/A

Several parameters have been proposed in the literature for the evaluation of seismic damage. However, in most cases the correlation between results obtained using these parameters and observed damage in structures has not been satisfactory. A parameter for measuring seismic damage previously proposed by the first author is used in this study to analyse a set of 15 accelerograms recorded in 11 earthquakes experienced in different countries. Results using this parameter are compared to global building damage observed during these earthquakes. The use of the parameter proposed here yields results which are consistent with building damage observed in the earthquakes studied. Copyright © 1999 John Wiley & Sons, Ltd.

Kongoli, Xhafer; Minami, Tadao; Sakai, Yuki

doi: 10.1002/(SICI)1096-9845(199905)28:5<479::AID-EQE826>3.0.CO;2-Hpmid: N/A

Effects of structural walls on the elastic–plastic earthquake response of short‐ to medium‐height reinforced concrete buildings were investigated. The analytical model consists of independent lumped mass systems representing walls and frames connected at each floor. The wall structure undergoes flexural as well as shear deformation and fails in shear at relatively small story drifts, the frames deforming only in shear. As a measure of structural damage, the ductility factor responses of frame structures were calculated for different combinations of base shear coefficients for the frames and walls. In buildings with relatively weak frames, the installation of structural walls did not improve the large plastic response of the frames up to the point where the walls were unfailed in shear and the ductility factors of the frame structure were suddenly reduced to a very small number. For relatively strong frames, however, the response displacements decreased gradually as the number of walls increased, whether or not the walls failed. Empirical formulas for the required base shear coefficients of the walls and frames which gave a target ductility factor response also were derived for two particular groups of accelerograms. These equations should be of practical use in designing frame‐wall type buildings and in retrofitting damaged buildings. Copyright © 1999 John Wiley & Sons, Ltd.

Panagiotakos, T. B.; Fardis, M. N.

doi: 10.1002/(SICI)1096-9845(199905)28:5<501::AID-EQE827>3.0.CO;2-5pmid: N/A

Estimation of peak inelastic deformation demands is a key component of any displacement‐based procedure for earthquake‐resistant design of new structures or for seismic evaluation of existing structures. On the basis of the results of over a thousand non‐linear dynamic analyses, rules are developed for the estimation of mean and upper‐characteristic peak inelastic interstorey drifts and member chord rotations in multistorey RC frame buildings, either bare or infilled in all storeys but the first. For bare frame structures, mean inelastic deformation demands can be estimated from a linear, equivalent static, or preferably multimodal response spectrum analysis with 5 per cent damping and with the RC members considered with their secant stiffness at yielding. 95 per cent characteristic values can be estimated as multiples of the mean deformations. For open‐first‐storey buildings, the linear analysis can be equivalent static, with the infills modelled as rigid bidiagonal struts and all RC members considered with their secant stiffness to yielding. Copyright © 1999 John Wiley & Sons, Ltd.

Yamada, Kazuhiko

doi: 10.1002/(SICI)1096-9845(199905)28:5<529::AID-EQE828>3.0.CO;2-Dpmid: N/A

This paper proposes a non‐linear control law for a variable damping device (VDD) aimed at reducing structural seismic responses. The VDD is attached to the structure by an auxiliary spring element composing a non‐linear Maxwell element. The VDD's damping coefficient is adjusted to control the reactive internal force in the non‐linear Maxwell element. A large controlled force is thus produced with little external power required to adjust the VDD's damping coefficient. The proposed control law defines the rate or increment of the VDD's damping coefficient at a certain moment by a differential equation or its discretized form. The controlled force vs. deformation relation plots parallelogram‐like hysteretic curves, which indicates quick action and energy dissipation. Fundamental characteristics of an SDOF model with the VDD controlled by the proposed law are examined for impulse, sin and seismic excitations. The law for the SDOF model is extended to one for an MDOF model. The control effect for a 3DOF model is examined by numerical experiments. Copyright © 1999 John Wiley & Sons, Ltd.

Pagnini, Luisa Carlotta; Solari, Giovanni

doi: 10.1002/(SICI)1096-9845(199905)28:5<543::AID-EQE829>3.0.CO;2-Qpmid: N/A

The dynamic response of bridge piers with aseismic devices to earthquake excitation is evaluated by the stochastic equivalent linearization technique. The seismic acceleration is schematized through a Gaussian stationary random process. The pier is considered linear elastic, the span is idealized as a rigid mass, the restoring force of the device is represented through a non‐linear differential model. The study of the complex modes of the linearized system gives an interpretation of the mechanical behaviour, leads to a formally elementary solution and highlights some phenomena which are typical of the hysteretic systems, particularly of those marked by weak hardening. Even though the solution is limited to the stationary field, it brings out several noteworthy considerations about the effective non stationary behaviour of the structure. Copyright © 1999 John Wiley & Sons, Ltd.