Search

Measured force-displacement sensing data are analyzed to create nonlinear hysteretic models for elastic sliding bearings and steel dampers. These data were collected during full-scale experimental tests of a base-isolated structure at Japan's National Research Institute for Earth Science and Disaster Resilience (NIED) "E-Defense" Hyogo Earthquake Engineering Research Center in 2013 that employed historical and synthetic earthquake inputs. Several models are explored for each isolation-layer device, including Coulomb friction models for the elastic sliding bearings and Bouc-Wen hysteresis models for both devices. Parameters for each device model are estimated via constrained nonlinear optimization to minimize model prediction error using testing data. The resulting models are applied to data from other seismic tests to evaluate the models' ability to provide accurate force predictions. The model for the steel dampers demonstrates the capacity to be generalized across seismic excitations and devices, whereas the model for the elastic sliding bearing reproduces well its behavior during other seismic excitations but is limited to the device to which it was calibrated. The residual error between the model predictions and measurements does not contribute a significant degree of variance in the model parameter estimates. [ABSTRACT FROM AUTHOR]

Summary: Full‐scale dynamic testing and earthquake engineering experiments are often costly, yet they are vital for advancing the understanding of the complex interactions between seismic protective systems and structural responses. E‐Defense (a part of the Japan's National Research Institute for Earth Science and Disaster Resilience) constructed and tested a four‐story reinforced‐concrete base‐isolated building to study, among other things, how different anti‐seismic devices in the isolation layer of a building improve its response. This structure was subjected in August 2013 to a series of tests, including random ground motions and simulated earthquakes. Analyzing the acceleration responses from these random excitations forms the first step in identifying the dynamic characteristics of this test structure. The advanced numerical algorithm for subspace state‐space system identification (N4SID), from the family of subspace identification methods, is used to estimate the nominal modal parameters of this structure for both input–output and output‐only models. These models are then evaluated for their generalizability by using them to predict the response of the structure to both random excitations and seismic inputs. Additional identification is performed using responses recorded during simulated earthquakes, and the recovered modal parameters are compared against the nominal parameters to analyze how the building's isolation system impacts its response. The results indicate that the interaction between seismic excitation and the base‐isolation layer has both local and global structural effects. Locally, the modes most directly associated with isolation layer displacements experience, as expected, significant decreases in frequency and increases in damping; globally, the modes with more substantial displacements across the superstructure also experience decreases in frequency and increases in damping but at a greatly reduced scale. Further, even in the presence of seismic forces large enough to induce nonlinear restoring forces in the isolation layer, the relative changes in identified modal parameters for the superstructure associated modes remained modest. [ABSTRACT FROM AUTHOR]

A data-based procedure is presented to develop spectra-compatible time-history records that are based on the dominant probabilistic features of an ensemble of records corresponding to the non-stationary stochastic phenomena of interest (e.g., earthquakes, wind loads, etc). The method requires a statistically significant collection of time-history records that are used to construct the associated covariance kernel of the random process. Subsequently, orthogonal decomposition approaches are used to determine the dominant eigenvectors of the covariance matrix, and these vectors are then linearly combined, with an adjustable amplitude-scale and phase-shift, to determine, via a nonlinear optimization scheme (employing a combination of stochastic and deterministic approaches), a time-history record that matches the target spectrum within a specified error bound. The utility of this approach is demonstrated with several collections of earthquake records from different regions of the world (Japan, Los Angeles, and San Francisco) that are then used to match various spectra widely used in seismic design applications. Issues that impact the selection of the bases vectors to construct the optimum spectra-matching record are discussed, and guidelines are provided for successful implementation of the proposed methodology. [ABSTRACT FROM AUTHOR]

Models are used to represent and characterize physical phenomena. When there are many plausible models for a particular phenomenon, the modeler can exploit the computational tool called model falsification to systematically eliminate models that do not reasonably fit measured data. Model falsification typically compares measurements and their predictions by different models, and rejects a model if some metric of the difference between them is outside some prescribed bounds. This paper compares two model falsification approaches: a conventional bounds on residual errors and a proposed bounds on a model's prediction of the likelihood of the residual errors. The bounds in both approaches are selected based on two error control criteria: the more commonly used familywise error rate (FWER) and—proposed herein for model falsification—the false discovery rate (FDR). Because FDR control significantly increases the likelihood of rejecting an invalid model when there are many measurements, FDR provides advantages over FWER in exploratory studies. A variant of the second approach, using likelihood bounds specified by a constant probability mass contained within those bounds, is also investigated. Unlike many model falsification studies, the focus herein is on systems with many measurements, spread across spatial and/or temporal dimensions, such as dynamical systems. An elementary example is used to show the principles of each approach. A second example considers a series of four-degree-of-freedom models of a structure subjected to an earthquake excitation. The results from these examples show that FDR does indeed increase the number of falsified models, whereas the use of likelihood bounds additionally gives unfalsified models confidence values, which can also be used for maximum likelihood parameter estimation. [ABSTRACT FROM AUTHOR]

The experimental dynamical response of three types of nonlinear hysteretic systems is identified employing phenomenological models togheter with the Differential Evolutionary algorithm. The mass–spring–damper system is characterized by hysteretic restoring forces provided by assemblies of shape memory and steel wire ropes subject to flexure or coupled states of tension and flexure. The energy dissipation due to phase transformations and inter-wire friction and the stretching-induced geometric nonlinearities give rise to different shapes of hysteresis cycles. The mechanical device subject to strong seismic excitations is investigated in its ultimate limit state whereby inelastic strains are induced in the steel wires together with a global nonsymmetric response of the system. The targeted dynamical characterization of the hysteretic oscillator up to its ultimate limit state has a special meaning when the device is employed in the field of vibration control. The dynamical response is identified exploiting the measurements of the oscillating mass relative displacement and inertia force that must be balanced, at each time instant, by the overall restoring forces provided by the mechanism. The restoring force is assumed to be the sum of different contributions such as a cubic nonsymmetric elastic force and a nonsymmetric hysteretic force modeled according to a modified version of the Bouc–Wen model. The parameters are identified minimizing the difference between the numerical and the experimental restoring force histories. High levels of accuracy are achieved in the identification with mean square errors lower than 2%. [ABSTRACT FROM AUTHOR]

This paper proposes a state-space-based approach for computational simulation and prediction of shock response for both time histories and shock response spectra. The method operates by developing a nominal model for the system through state-space identification and then modeling the resulting residual through an artificial neural network. The model for the residual is then folded back into the nominal system via Kalman filtering, allowing for forward computational simulation without any measurement information. The proposed identification method is applied to a three-degree-of-freedom system and a high-fidelity finite element model built in Abaqus. The proposed method provides reasonable predictions for time histories of excitations not seen during training or identification and produces useful predictions of the associated shock response spectra. [ABSTRACT FROM AUTHOR]

Models, typically given by systems of mathematical equations, are built to help represent, understand, and further characterize physical phenomena. The choice of a model for a particular phenomenon is made based on user judgment, evidence from measurement data, and/or the ease of its use. Generally, many linear and nonlinear models are available to describe a particular structural dynamical system. Bayesian model selection is a probabilistic tool to help select suitable mathematical model(s) among a possible set of models using Bayes' theorem. To simplify the analysis, linear structural dynamical models are often used, regardless of whether the dynamical system behaves linearly or not. However, linear models are not always adequate to accurately compute structural responses. When the models also involve some nonlinearity, the required computation for Bayesian model selection increases significantly. An important class of nonlinear problems consists of models that are mostly linear except for some spatially localized nonlinearities. For example, in a building with base isolation, the superstructure is designed to behave essentially linearly in an earthquake excitation and only the isolation layer behaves nonlinearly. Similarly, spacecraft may be modeled with linear components that are connected by spatially localized nonlinear joints. To lessen this increased computational burden, a method is proposed in this paper, combining the senior authors' previously developed efficient dynamic response algorithm for locally nonlinear systems and an intelligent sampling algorithm, to calculate the evidence for, or marginal likelihood of, a model. The efficient dynamic response algorithm helps achieve significant gains in computational efficiency by exactly transforming the potentially high-dimensional state-space equation of the structural dynamical system into a low-order nonlinear Volterra integral equation. This algorithm is embedded into a nested sampling algorithm, which samples parameters more frequently from the high likelihood region (even if the region of higher likelihood is very different from the region where the prior parameter density function is large), resulting in a computationally efficient framework for Bayesian model selection with locally nonlinear systems; a (moderately) alternate derivation of nested sampling is developed. The approach is demonstrated using three numerical examples. The first two examples consider building models mounted on a hysteretic isolation layer that are subjected to an earthquake excitation. Both single and 99-degree-of-freedom superstructure models are investigated. Different candidate linear and nonlinear models representing the hysteretic behavior of the isolation layer are used. The third example consists of a three-dimensional wind-excited 1,623-degree-of-freedom building structure with tuned mass dampers (TMDs) attached to its roof, where different linear and nonlinear candidate damping models are considered in the TMDs. All three examples include cases in which the true model is not among the models evaluated. The application of the proposed synergistic approach, consisting of an intelligent sampling and an efficient dynamic response algorithm, demonstrates gains in Bayesian model selection computational efficiency of up to three orders of magnitude relative to conventional approaches with comparable accuracy, reducing days of computation to minutes or hours. [ABSTRACT FROM AUTHOR]

This work presents an innovative framework for the simulation of stable pit growth that accounts for the material microstructure in the computational model to investigate its effect on pit growth behavior. While many microstructural attributes can affect corrosion kinetics, the current work focuses on the effect of the crystallographic plane orientation at the corrosion front by utilizing variations in polarization behavior with orientation in a single phase material. The current modeling results show that crystallographic orientation significantly affects local pit growth rates (even with corrosion potential variations as little as 5–10% across orientations) and the resultant tortuous shapes are similar to pits observed in experimental studies. [ABSTRACT FROM AUTHOR]

The authors have previously shown that many traditional approaches to operational modal analysis (OMA) struggle to properly identify the modal damping ratios for bridges under traffic loading due to the interference caused by the driving frequencies of the traffic loads. This paper presents a novel methodology for modal parameter estimation in OMA that overcomes the problems presented by driving frequencies and significantly improves the damping estimates. This methodology is based on finding the power spectral density (PSD) of a given modal coordinate, and then dividing the modal PSD into separate regions, left- and right-side spectra. The modal coordinates were found using a blind source separation (BSS) algorithm and a curve-fitting technique was developed that uses optimization to find the modal parameters that best fit each side spectra of the PSD. Specifically, a pattern-search optimization method was combined with a clustering analysis algorithm and together they were employed in a series of stages in order to improve the estimates of the modal damping ratios. This method was used to estimate the damping ratios from a simulated bridge model subjected to moving traffic loads. The results of this method were compared to other established OMA methods, such as Frequency Domain Decomposition (FDD) and BSS methods, and they were found to be more accurate and more reliable, even for modes that had their PSDs distorted or altered by driving frequencies. [ABSTRACT FROM AUTHOR]

Several different data-driven strategies for nonlinear identification are applied to experimental data exhibiting various types of hysteretic behavior. The experimental data contain displacement and restoring force information for several tests conducted using different configurations of a rheological testing device with various assemblies of nickel titanium-Naval Ordnance Laboratory (NiTiNOL) and steel wire strands. Among the different configurations, the response of the wire strands shows three distinct forms of nonlinear behavior: classical quasi-linear softening hysteresis; strongly pinched, hardening hysteresis; and slightly pinched, hardening hysteresis. The data-driven methods applied for nonlinear identification include polynomial basis functions and neural networks. The polynomial basis nonlinear identification methods are used for the construction and characterization of reduced-order models to gain insight into the physical modeling of the hysteretic phenomena. The neural network methods are found to be more useful for predictive purposes, demonstrating an ability to produce accurate results on both training and testing data. [ABSTRACT FROM AUTHOR]

Many marine structures rely upon hull monitoring systems that utilize strain information to make inferences about structural integrity. One of the most significant factors affecting structural integrity are wave loads, specifically the resultant pressures. While direct measurement is not often available, this study explores how to translate strain information into localized pressure estimates through two different low-order modeling paradigms, physics-based and surrogate modeling. Strain data was collected via strain gauges mounted on a scaled ship model that was tested in towing tank under a variety of regular and irregular wave conditions. Using a set of strain calibration data, two different models are developed, one using principal stress and strain relations and the other using a machine learning approach (Gaussian process model), for predicting pressure time histories for different wave conditions. The two modeling approaches are compared in terms of both a variety of error metrics as well as their ability to handle measurement noise and disturbances commonly encountered during marine structures testing and operations. • Model vessel monitored via strain and pressure sensors during wave tests. • Mechanics and machine learning models convert strain measurements to pressure. • Both models deliverable reasonable predictions of pressure time histories. • Measurement errors manifest as increased variance in machine learning model. [ABSTRACT FROM AUTHOR]

Accurate estimation of the damping in a structure has remained an important but challenging problem for the structural engineering community. The relative difficulty of damping estimation can be compounded when the excitation is not uniformly broadband or ambient in nature, such as when car traffic or large trains travel over a bridge. A bridge model that consisted of a series of simply supported (SS) stringers resting atop a larger girder was constructed using finite elements, and several simulations were conducted in which cars and trains crossed the bridge model. The presence of the cars and trains led to the appearance of driving forcing frequencies in the response. Driving frequencies are inherent to moving loads and proportional to the velocity of the moving loads and the length of the beam or bridge being crossed. The moving loads produced a pulse-like response in the SS stringers on the bridge model, and the resulting power spectral density (PSD) of the stringers showed that its power was concentrated at the first driving frequency and its even multiples. The vertical component of the moving load and the SS stringer responses were transferred to the girder, and owing to the continuity within the girder, each of its nodes experienced all of the car crossings. The repeated nature of the car crossings led to a PSD of the girder responses that contained peaks at the even multiples of the driving frequency. Over the course of a full simulation, the peaks produced by the different car velocities merged to form shelves of elevated power; these shelves were repeated at each of the even multiples of the driving frequency, causing sustained regions of distortion in the frequency response spectra of the girder. Attempts were made to identify the modal damping ratios from the bridge acceleration responses using the frequency domain decomposition (FDD) and blind source separation (BSS) methods, but the driving frequencies interfered with the estimates. The regions of distortion compromised the spectra for the frequency-based methods and altered the estimated modal responses recovered from the BSS method, creating problems with identification in the time domain. The driving frequencies generated by the car and train traffic on a bridge negatively impacted both the reliability and accuracy of the damping estimates found using various operational modal analysis (OMA) techniques. [ABSTRACT FROM AUTHOR]