دانلود ترجمه مقاله تشخیص آسیب سازه تحت تحریک زلزله با آنالیز موجک گسسته
بخشی از ترجمه:
یک شیوه‌ی قابل قبول برای شناسایی میزان صدمات وارده به سازه در اثر زمین لرزه که شیوه‌ی تبدیل موجک‌های گشته نام دارد در این مقاله معرفی شده است.
عملکرد این شیوه‌ی پیشنهادی بر اساس تغییرات ناگهانی در پاسخ به لرزش‌ها و تحلیل جا به جایی و یا واکنش‌های سرعتی با استفاده از تحلیل موجک‌هاست.
هم چنین، شیوه‌ی دیگری که عملکرد این شیوه‌ی پیشنهادی بر اساس تغییرات ناگهانی در پاسخ به لرزش‌ها و تحلیل جابه جایی و یا واکنش‌های سرعتی با استفاده از تحلیل موجک‌هاست. هم چنین، شیوه‌ی دیگری که عملکرد آن نیز بر اساس همین موجک‌هاست، ارائه شده که راه حل مربوط به مشکل شناسایی صدمات وارده است و مانع از انتشار جابه جایی‌ها و واکنش‌های مربوط به سرعت می‌شود. عملکرد شیوه‌ی پیشنهادی برای جلوگیری از انتشار امواج و کشف صدمات وارده با استفاده از مدل معیار IASC-ASCE که مدل کار گروهی بوده و سلامت سازه را بررسی می‌کند، سنجیده می‌شود. نتایج به دست آمده نشان می‌دهند که شیوه‌ی پیشنهادی می‌تواند جاوی اطلاعات سودمندی برای پیش گیری از آسیب‌های وارده باشد که با استفاده از واکنش سازه در برابر لرزه‌ها محقق می‌شود.
١)مقدمه
در دهه‌ی گذشته، توجه محققان همواره به آسیب‌های وارده بر سازه‌ها در طول عمر مهندسی سازه بوده است. در بین شیوه‌های متعدد، برخی از آنها بر اساس مشاهده‌ی رفتار دینامیکی سازه شکل گرفته‌اند (۴-١). بسیاری از این شیوه‌ها از پارامترهای مدی modal مانند شکل مد و فرکانس‌های طبیعی مربوط به تخمین و کش آسیب‌های وارده، استفاده می‌کنند.
بخشی از مقاله انگلیسی:
ABSTRACT An effective method for the damage diagnosis of structures under seismic excitation via discrete wavelet transform is proposed in this paper. The proposed method is based on the detection of abrupt changes in seismic vibration responses by the analysis of displacement or velocity responses using wavelet analysis. Also, a wavelet-based method is presented for denoising of displacement and velocity responses for the damage detection problem. The performance of the proposed method for de-noising and damage detection has been investigated using a benchmark problem provided by the IASC-ASCE Task Group on Structural Health Monitoring and a simulated shear wall model. The obtained results indicate that the proposed method can be provided useful information for the damage occurrence using the seismic response of structures. Keywords: Damage detection; de-noising; wavelet transform; seismic response; earthquake excitation 1. INTRODUCTION In the last decade, the researcher’s attentions on the detection of structural damage during the service life of engineering structures have been increased. Among numerous methods, approaches that are based on the observation of the dynamic behavior of a structure heve been developed [1-4]. Many of these techniques use the identified modal parameters like mode shapes and natural frequencies for structural damage detection and estimation. The identification of alteration in mode shapes and natural frequencies at the damaged system in comparison with the undamaged system is one of the popular methods in the structural damage detection. These changes are often small and measurements are polluted by noise * E-mail address of the corresponding author: abb46@pitt.edu (A. Bagheri) 290 A. Bagheri and S. Kourehli making this method an inefficient method. The wavelet transform is a effective method for precise signal analysis, which overcomes the problems exhibited by other signal processing techniques. Applying wavelet transform for the analysis of damaged structures responses produces satisfying results in the damage identification. Sharp changes in the wavelet coefficients near a damage exhibit the presence of damage. This method is nonparametric and allows the accurate estimation of the time of the sudden change, benefiting from the fine time resolution of the wavelet transform at small scales (i.e. high frequencies). The advantages of damage identification using wavelet analysis are nonparametric and baseline-free data. In addition, it does not depend on the change of the structural frequency, which is sensitive to soil-structure interaction [5]. Wavelet analysis has been applied in the system identification and damage detection. In addition, it has been widely implemented for various purposes, such as the characterization of non-stationary dynamic responses [6, 7]. General overview of damage detection by wavelet analysis may be found in Kim and Melhem [8] and a review on some of the wavelet application such as time-frequency analysis of signals, the fault feature extraction, the denoising and extraction of the weak signals have been done by Peng et al. [9]. In addition, the possibility of applying wavelet transform for the detection of beam cracks has been studied by Sun and Chang [10], Han et al. [11] and Poudel et al. [12]. Damage detection of frame structures via wavelet transform has been analyzed by Ovanesova and Suarez [13] and Hou et al. [14]. Rucka and Wilde [15] proposed a method for estimating damage localization in a beam and a plate by applying continuous wavelet transform. Bayissa and Haritos [16] proposed a new damage identification technique based on the statistical moments of the energy density function of vibration responses in the time-scale domain. Also, Bayissa et al. [17] offered a new damage detection technique using wavelet transform based on the vibration responses of plate. Fan and Qiao [18] developed a two-dimensional continuous wavelet transform-based damage detection algorithm using Dergauss2d wavelet for platetype structures. Also, a distributed two-dimensional continuous wavelet transform algorithm has been developed by Huang et al. [19]. They used data from discrete sets of nodes and provide spatially continuous variation in the structural response parameters to monitor structural degradation. Recently, a method has been proposed for the detection of crack-like damage in plate structures using discrete wavelet transform by Bagheri et al. [20]. In this study, the event of damages in a structure subjected to an earthquake ground motion is detected, which is related to the numbers of spikes in wavelet results. The proposed methodology is applied to the numerical examples subjected to a real earthquake. The models consist of single and multi degree of freedoms with damping. The efficiency of the presented method is shown based on the obtained results for the damage detection. 2. OVERVIEW ON WAVELET TRANSFORM Fast Fourier transform is efficient tool for finding the frequency components in a signal. The major disadvantage of fast Fourier transform is that they have only frequency resolution and no time resolution. Therefore, it is not a suitable tool for non-stationary signals. These types of signals can be processed via wavelet transform. It provides a powerful tool to characterize DAMAGE DETECTION OF STRUCTURES UNDER EARTHQUAKE EXCITATION … 291 local features of a signal. The main advantage gained by using the wavelet transform is the ability to perform the local analysis of a signal. Unlike Fourier transform, where the function used as the basis of decomposition is a sinusoidal wave, other basis functions can be selected for wavelet shape according to the features of the signal. The basis function in wavelet analysis is defined by two parameters named scale and translation. This property leads to a multi-resolution representation for non-stationary signals. As mentioned before, a basis function or mother wavelet is used in wavelet transform
