Health Diagnosis of Structural Systems Using a Repetitive Model Updating Approach

Author(s):  
Jeng-Wen Lin ◽  
Chong-Shien Tsai ◽  
Chih-Wei Huang

This paper proposes a statistical confidence interval based model updating approach for the health diagnosis of structural systems subjected to seismic excitations. The proposed model updating approach uses the 95% confidence interval of the estimated structural parameters to determine their statistical significance in a least-squares regression setting. When the parameters’ confidence interval covers the “null” value, it is statistically sustainable to truncate such parameters. The remaining parameters will repetitively undergo such parameter sifting process for model updating until all the parameters’ statistical significance cannot be further improved. This newly developed model updating approach is implemented for the developed series models of multivariable polynomial expansions: the linear, the Taylor series, and the power series model, leading to a more accurate identification as well as a more controllable design for system vibration control.

2009 ◽  
Vol 16 (3) ◽  
pp. 229-240 ◽  
Author(s):  
Jeng-Wen Lin ◽  
Hung-Jen Chen

This paper proposes a statistical confidence interval based nonlinear model parameter refinement approach for the health monitoring of structural systems subjected to seismic excitations. The developed model refinement approach uses the 95% confidence interval of the estimated structural parameters to determine their statistical significance in a least-squares regression setting. When the parameters' confidence interval covers the zero value, it is statistically sustainable to truncate such parameters. The remaining parameters will repetitively undergo such parameter sifting process for model refinement until all the parameters' statistical significance cannot be further improved. This newly developed model refinement approach is implemented for the series models of multivariable polynomial expansions: the linear, the Taylor series, and the power series model, leading to a more accurate identification as well as a more controllable design for system vibration control. Because the statistical regression based model refinement approach is intrinsically used to process a “batch” of data and obtain an ensemble average estimation such as the structural stiffness, the Kalman filter and one of its extended versions is introduced to the refined power series model for structural health monitoring.


2020 ◽  
pp. 147592172092697
Author(s):  
Zhao Chen ◽  
Hao Sun

Identifying damage of structural systems is typically characterized as an inverse problem which might be ill-conditioned due to aleatory and epistemic uncertainties induced by measurement noise and modeling error. Sparse representation can be used to perform inverse analysis for the case of sparse damage. In this article, we propose a novel two-stage sensitivity analysis–based framework for both model updating and sparse damage identification. Specifically, an [Formula: see text] Bayesian learning method is first developed for updating the intact model and uncertainty quantification so as to set forward a baseline for damage detection. A sparse representation pipeline built on a quasi-[Formula: see text] method, for example, sequential threshold least squares regression, is then presented for damage localization and quantification. In addition, Bayesian optimization together with cross-validation is developed to heuristically learn hyperparameters from data, which saves the computational cost of hyperparameter tuning and produces more reliable identification result. The proposed framework is verified by three examples, including a 10-story shear-type building, a complex truss structure, and a shake-table test of an eight-story steel frame. Results show that the proposed approach is capable of both localizing and quantifying structural damage with high accuracy.


Author(s):  
Jeng-Wen Lin ◽  
Chih-Wei Huang

System identification for economic health monitoring in modern economy is moving to the forefront of worldwide research activities. Such monitoring for economic development and management is important for the public welfare related to technological breakthrough such as the Industrial Revolution, leading to new construction technology. In economy practice, however, there are many situations in which a feedback identification system is given model uncertainties and uncertainty of measurement. Aiming for accurate economy model updating, this paper presents the diagnosis of the GDP per capita trend through an automatic repetitive sifting process. It shows how a statistical confidence interval based model updating approach can be applied to the health evaluation of economic development via prediction of GDP per capita over time. The model updating approach uses the confidence interval of the estimated economic parameters to determine their statistical significance in order to mitigate the model uncertainties and measurement errors. If the parameters’ confidence interval covers the “null” value, it is statistically sustainable to truncate such parameters. The remaining parameters will repetitively undergo such parameter sifting process until all the parameters’ statistical significance cannot be further improved. Consequently, the proposed repetitive identification approach promotes the accuracy of the prediction of GDP per capita, assisting the assessment of modern economic trend.


Author(s):  
Jeng-Wen Lin ◽  
Chong-Shien Tsai ◽  
Wen-Shin Chen

This paper presents the identification of structural systems under tri-directional seismic excitations using a statistically refined Bouc-Wen model of tri-axial interaction. Through limited vibration measurements in the National Center for Research on Earthquake Engineering in Taiwan, the Bouc-Wen model has been statistically and repetitively refined using the 95% confidence interval of the estimated structural parameters so as to determine their statistical significance in a multiple regression setting. The effectiveness of the refined model has been shown considering the effects of the sampling error, of the coupled restoring forces in tri-directions, and of the under-over-parameterization of structural systems. Sifted and estimated parameters such as the stiffness, and its corresponding natural frequency, resulting from the methodology developed in this paper are carefully observed for system vibration control.


2021 ◽  
Vol 11 (4) ◽  
pp. 1622
Author(s):  
Gun Park ◽  
Ki-Nam Hong ◽  
Hyungchul Yoon

Structural members can be damaged from earthquakes or deterioration. The finite element (FE) model of a structure should be updated to reflect the damage conditions. If the stiffness reduction is ignored, the analysis results will be unreliable. Conventional FE model updating techniques measure the structure response with accelerometers to update the FE model. However, accelerometers can measure the response only where the sensor is installed. This paper introduces a new computer-vision based method for structural FE model updating using genetic algorithm. The system measures the displacement of the structure using seven different object tracking algorithms, and optimizes the structural parameters using genetic algorithm. To validate the performance, a lab-scale test with a three-story building was conducted. The displacement of each story of the building was measured before and after reducing the stiffness of one column. Genetic algorithm automatically optimized the non-damaged state of the FE model to the damaged state. The proposed method successfully updated the FE model to the damaged state. The proposed method is expected to reduce the time and cost of FE model updating.


2013 ◽  
Vol 2013 ◽  
pp. 1-21 ◽  
Author(s):  
Rita Greco ◽  
Francesco Trentadue

Response sensitivity evaluation is an important element in reliability evaluation and design optimization of structural systems. It has been widely studied under static and dynamic forcing conditions with deterministic input data. In this paper, structural response and reliability sensitivities are determined by means of the time domain covariance analysis in both classically and nonclassically damped linear structural systems. A time integration scheme is proposed for covariance sensitivity. A modulated, filtered, white noise input process is adopted to model the stochastic nonstationary loads. The method allows for the evaluation of sensitivity statistics of different quantities of dynamic response with respect to structural parameters. Finally, numerical examples are presented regarding a multistorey shear frame building.


2013 ◽  
Vol 2013 ◽  
pp. 1-11 ◽  
Author(s):  
B. Asgari ◽  
S. A. Osman ◽  
A. Adnan

The model tuning through sensitivity analysis is a prominent procedure to assess the structural behavior and dynamic characteristics of cable-stayed bridges. Most of the previous sensitivity-based model tuning methods are automatic iterative processes; however, the results of recent studies show that the most reasonable results are achievable by applying the manual methods to update the analytical model of cable-stayed bridges. This paper presents a model updating algorithm for highly redundant cable-stayed bridges that can be used as an iterative manual procedure. The updating parameters are selected through the sensitivity analysis which helps to better understand the structural behavior of the bridge. The finite element model of Tatara Bridge is considered for the numerical studies. The results of the simulations indicate the efficiency and applicability of the presented manual tuning method for updating the finite element model of cable-stayed bridges. The new aspects regarding effective material and structural parameters and model tuning procedure presented in this paper will be useful for analyzing and model updating of cable-stayed bridges.


2018 ◽  
Vol 44 ◽  
pp. 00121
Author(s):  
Sara Nicpoń ◽  
Paula Iliaszewicz ◽  
Maciej Leoniak ◽  
Agnieszka Trusz-Zdybek

For proper enumeration of protozoa in activated sludge good methodology is required. In this paper we present some remarks on microscopic methodology of protozoa enumeration. This remarks concern number of repetitions from one sample required to obtain good statistical results as well as influence of sample aeration on number of found protozoa. Presented data shows that at last 10 repetitions are required from each sample to obtain low average confidence interval. Lower number of repetitions leads to sharp increase in average confidence interval and loss of statistical significance while higher number does not decrease average confidence interval substantially. As measurements lasts for few hours lack of sample’s aeration during measurement leads to detection of lower by 27% number of protozoa.


2018 ◽  
Vol 9 (2) ◽  
pp. 28-39
Author(s):  
Elena A. Zvyagina ◽  
Tatiana S. Pereyzslovets

In light of the observed global climate changes in recent decades, we studied the local climate indicators and explored the possible links between the spring and autumn phenophases and climate data changes in the Yuganskiy nature reserve (N 600 17’; E74054’ – N590 23’; E74000’) in 1982-2016. The collected climate data include daily average, maximum and minimum temperatures, daily precipitation amount and intensity, and number of days with precipitation of 0.1mm or more, monthly average of snow depths, dates of first and last occurrence of daily mean temperatures 0, +5, +10°С through the year. Timing of sap movement and leaf fall start were used as the spring and autumn indicators of birch (Bétula péndula ) phenology. The mean value of weather averages in the 30-year period of 1961-1990 was used as reference. Trends were calculated using linear least squares regression. Statistical significance was determined by calculating the standard error of the trend estimate. We found that the annual mean temperature has increased from –1.9°С (1961-1990) to –0.8° С (1982-2016), with corroborating indicators including increased temperature of the coldest night of the year from –53°С (1961-1990) to –51.3° С (1982-2016) and increased frequency of significant positive temperature anomalies from 21% (1961-1990) to 37% (1982-2016). May, June, August and October nights have become successively warmer. The air temperature increase was not accompanied by a corresponding increase in precipitation. Statistically significant trends toward earlier onset of spring and summer from 1982 to 2016 were observed. The date of the last spring freeze has been advancing by 6.1 days per decade since 1982. A freeze-free season has lengthened by 7.7 days per decade. Linear trend of the snowmelt timing was –3.7 days per decade. Permanent snow cover period has been shortening by 7.7 days per decade. The date of the first occurrence of the daily mean temperatures of +10° С has been advancing by 5.1 days per decade. However, the 0 -+5° С lag has been lengthening significantly by 9.2 days per decade, and the number of biologically effective degree days (base +5C) has not statistically changed. Sap flux and leaf fall timing of B. pendula have been advancing almost simultaneously by 4.0 and 4.2 days per decade since 1985. Sap flux beginning and last spring freeze date have been found to be linearly correlated (r=0.904). The average lag of them was 5±1 days and has been lengthening by 3 days per decade (1985–2016).


2021 ◽  
Author(s):  
jice zeng ◽  
Young Hoon Kim

Damage detection inevitably involves uncertainties originated from measurement noise and modeling error. It may cause incorrect damage detection results if not appropriately treating uncertainties. To this end, vibration-based Bayesian model updating (VBMU) is developed to utilize vibration responses or modal parameters to identify structural parameters (e.g., mass and stiffness) as probability distribution functions (PDF) and uncertainties. However, traditional VBMU often assumes that mass is well known and invariant because simultaneous identification of mass and stiffness may yield an unidentifiable problem due to the coupling effect of the mass and stiffness. In addition, the posterior PDF in VBMU is usually approximated by single-chain based Markov Chain Monte Carlo (MCMC), leading to a low convergence rate and limited capability for complex structures. This paper proposed a novel VBMU to address the coupling effect and identify mass and stiffness by adding known mass. Two vibration data sets are acquired from original and modified systems with added mass, giving the new characteristic equations. Then, the posterior PDF is reformulated by measured data and predicted counterparts from new characteristic equations. For efficiently approximating the posterior PDF, Differential Evolutionary Adaptive Metropolis (DREAM) Algorithm are adopted to draw samples by running multiple Markov chains parallelly to enhance convergence rate and sufficiently explore possible solutions. Finally, a numerical example with a ten-story shear building and a laboratory-scale three-story frame structure are utilized to demonstrate the efficacy of the proposed VBMU framework. The results show that the proposed method can successfully identify both mass and stiffness, and their uncertainties. Reliable probabilistic damage detection can also be achieved.


Sign in / Sign up

Export Citation Format

Share Document