Hilmi Luş

Probabilistic Benefit-Cost Analysis for Earthquake Damage Mitigation: Evaluating Measures for Apartment Houses in Turkey

In the wake of the 1999 earthquake destruction in Turkey, the urgent need has arisen to evaluate the benefits of loss mitigation measures that could be undertaken to strengthen the existing housing stock. In this study, a benefit-cost analysis methodology is introduced for the comparative evaluation of several seismic retrofitting measures applied to a representative apartment building located in Istanbul. The analysis is performed probabilistically through the development of fragility curves of the structure in its different retrofitted configurations. By incorporating the probabilistic seismic hazard for the region, expected direct losses can be estimated for arbitrary time horizons. By establishing realistic cost estimates of the retrofitting schemes and costs of direct losses, one can then estimate the net present value of the various retrofitting measures. The analysis in this work implies that, even when considering only direct losses, all of the retrofitting measures considered are desirable for all but the very shortest time horizons. This conclusion is valid for a wide range of estimates regarding costs of mitigation, discount rates, number of fatalities, and cost of human life. The general methodology developed here for a single building can be extended to an entire region by incorporating additional structural types, soil types, retrofitting measures, more precise space- and time-dependent seismic hazard estimates, etc. It is hoped that this work can serve as a benchmark for more realistic and systematic benefit-cost analyses for earthquake damage mitigation.

Ambient and Forced Vibration Testing of a Reinforced Concrete Building before and after Its Seismic Retrofitting

AbstractThis paper investigates the effects of seismic retrofitting on the modal characteristics of a 6-story RC building located in Istanbul, Turkey. Ambient vibration surveys were carried out before, during, and after the retrofitting work, which took place between June and December 2010. The building was retrofitted via jacketing of columns, addition of structural walls, and construction of a mat foundation. These studies were complemented with data from forced vibration tests performed with an eccentric-mass shaker after the retrofitting work was completed. During retrofitting, partitions were demolished; as a result, the first modal frequency of the building decreased by 11%, based on the results of the ambient vibration survey. The ambient vibration survey also showed that the modal frequencies after the seismic retrofitting increased by almost 96%. During the forced vibration tests, the building was excited around its modal frequencies using an eccentric-mass shaker. It was found that the modal dam...

El Farol revisited

This article is concerned with the global behavior of agents in the El Farol bar problem. In particular, we discuss global attendance in terms of its mean and variance, and show that there is a strong dependence of both on the externally imposed comfort level. We present a possible interpretation for the observed behavior, and propose that the mean attendance converges to the perceived threshold value as opposed to the actually imposed one.

Damage Detection Using Flexibility Proportional Coordinate Modal Assurance Criterion

In the recent years, vibration-based identification techniques have attracted the attention of the civil engineering community, as these methods can be naturally incorporated into automated continuous structural health monitoring procedures. It is a generally accepted approach to model the damage and deterioration of a structural element through stiffness reduction. For this reason, a feature tailored so as to be well correlated to the expected differences between the undamaged and damaged flexibility matrices, such as the recently proposed Flexibility Proportional Coordinate Modal Assurance Criterion (FPCOMAC), is ideally suited to be exploited as damage sensitive feature. We present a statistical pattern recognition based damage detection method that employs FPCOMAC as damage sensitive feature. The proposed methodology is executed according to the training and testing phases typical of the pattern recognition framework. Particular effort is devoted to test the ability of the method to correctly identify the damage when response time histories used in the training are measured in different environmental conditions. The formulation is derived considering a shear-type structural system. Results obtained by considering a 7 DOFs shear-type system prove the efficiency of the method in detecting and locating the damage, irrespective of damage severity and environmental effects, under the conditions that the damage amount is greater than the structural variations caused by the external factors and the amount of data is reasonably large.

Output Only Structural Identification with Minimal Instrumentation

It is of interest to the modal testing and structural health monitoring community to be able to identify the mass and stiffness parameters of a system from its vibration response measurements. However, incomplete instrumentation of the monitored system, and the more general ambient vibration situations, result in measured mode shapes which are incomplete and not normalized. In this paper, we consider the problem of mode shape expansion, normalization, and subsequent physical parameter identification, for shear-type structural systems with output-only (ambient vibration) measurements. It is shown that only two sensors, located at the first and last degrees of freedom of the system, are sufficient for parametric identification of the system. A mode shape expansion – physical parameter identification algorithm, starting from the incomplete modal parameters identified using a stochastic subspace based method, is discussed. The algorithm incorporates the information available from the structural topology of the physical system in terms of its modal parameters. The advantage of the algorithm lies in its ability to obtain a reliably accurate identification using the minimal necessary instrumentation, and the minimal necessary a-priori information about the system. The efficiency of the proposed algorithm is finally validated through numerical simulations of ambient vibration experiments on a 7 story shear-type structure, including the effect of white/colored measurement noise.

Probabilistic Structural Health Assessment with Identified Physical Parameters from Incomplete Measurements

AbstractThis paper studies the performance of identified physical parameters in structural health-monitoring applications. A probabilistic method is discussed to assess the location and severity of structural damage. This method attempts to account for the variability in both the baseline (healthy) and unknown (damaged or healthy) states of the monitored system through empirical distributions modeling the ratios of stiffness estimates from different tests. The presence and severity of damage at any location are detected by comparing the distribution in the unknown state with the baseline distribution; damage severity is expressed through damage probability versus severity curves corresponding to different confidence levels of the baseline state. Experimental data from a 3-story sliding base frame, modeled as a free–free system, and different damaged versions of the frame, are considered as applications. The structural identification is performed in a situation of highly-incomplete measured and a priori as...

Identification of structural damage using dynamic input-output measurements

This paper presents a variety of methodologies that are used to detect the location and amount of structural damage using dynamic measurements of the input and of the structural response. One approach (and its variations) starts from an identified first order model of a structural system and obtain estimation of the structures mass, damping and stiffness matrices. For these approaches, both the full instrumentation option and the partial instrumentation option are presented. An alternative approach for the identification of the dynamic characteristics of the structure is based on Evolution Strategies. Once these dynamic characteristics have been determined, structural damage is assessed by comparing the undamaged and damaged estimation of such parameters. Both these methodologies are tested on simulated numerical results and their effectiveness in determining structural damage is evaluated.

Physical Parameter Estimation from State-Space Models for Systems with Missing Input Information

AbstractIn this article, the transformation of linear first-order models to second-order models is investigated for cases in which the system is excited by unknown inputs and ground excitations. It is shown that the second-order coefficients lie in the null space of a matrix, which is constructed from the partitions of the state matrix using the Kronecker product. The conditions that determine the size of this null space are discussed and solutions are developed for various cases.

A risk based PML estimation method for single-storey reinforced concrete industrial buildings and its impact on earthquake insurance rates

Estimating earthquake losses is an important issue for many private and public bodies. As a major stakeholder, insurers need realistic probable maximum loss (PML) values to foresee the possible losses they would face after a major earthquake and also to calculate optimal insurance premiums. Insurers generally use fragility curves to manage their portfolio by calculating overall PML values. There are, however, serious impacts of risk based PML estimation on earthquake insurance rates, and in this respect fragility curves, which represent regional losses rather than individual losses, could lead to suboptimal decisions. In this study, a rapid earthquake loss estimation methodology, which can be used even by the non-experts in earthquake engineering without conducting comprehensive structural analyses, is proposed for single-storey reinforced concrete industrial buildings based on parameters determined after investigating more than 80 industrial building projects in Turkey. 384 analytical structural loss estimation curves were obtained via the non-linear structural performance analysis method proposed in the 2007 Turkish Seismic Design Code. To provide a detailed evaluation of the proposed methodology’s performance, fragility curves representative of the structural types and the design levels of the buildings investigated were also developed. Finally, total insurance premiums corresponding to PML values of the inventory buildings were calculated, using the two aforementioned estimation methods and others previously published, by addressing issues such as reinsurance cost, capital cost and profit. Results reveal considerable differences in PML values and eventually earthquake insurance rates for the buildings investigated between the risk based structural loss estimation method and the existing methods, indicating possibilities for improved portfolio analysis and management tools.

Structural Identification Using Response Measurements under Base Excitation

In vibration-based structural identification, experimentally obtained modal parameters from measured structural responses are often used, along with some information about the structural model, for identifying the physical parameters, i.e. mass and stiffness matrices of the structural system. In this study, we consider this problem of physical parameter identification for building structures subjected to base excitation, and attempt to address the issues of (a) unknown scaling of experimental mode shapes, (b) incomplete instrumentation, and (c) incomplete information of the physical parameters of the structural system prior to identification. A mode shape normalization and expansion approach, which incorporates the information available from the structural topology of the physical system in terms of its modal parameters, is discussed. Using this proposed approach, along with the modal orthogonality relations, the mass and stiffness matrices of the system can be estimated. The performance of the algorithm is finally evaluated through numerical simulations of base acceleration induced vibrations of a 4-story shear-type frame, as well as using experimental data collected from a 4-story frame subjected to base excitation on a shake table facility. The use of the modal-and-physical parameter identification method for the purpose of structural damage detection is also investigated using the experimental data, with the damage being represented by a reduction in the cross-sectional area of two columns of the “healthy” frame.