شناسایی رفتار غیرخطی قاب‌های خمشی فولادی تحت بار لرزه‌ای با استفاده از تابع پاسخ فرکانسی و تبدیل هیلبرت تابع پاسخ فرکانسی

رفتار غیرخطی یکی از نشانه های تغییرات در مشخصات عملکرد سازه ها می باشد و شناسایی رفتار غیرخطی می تواند در تشخیص خرابی سیستم های سازه ای مورد استفاده قرار گیرد. در این تحقیق روشی جهت شناسایی رفتار خطی و غیرخطی در قاب های خمشی فولادی، با استفاده از تابع پاسخ فرکانسی و تبدیل هیلبرت تابع پاسخ فرکانسی استخراج شده از سازه ها طی بارگذاری لرز‌ه ای، ارائه شده است. رفتار غیرخطی، با تشکیل مفاصل پلاستیک در ستون ها و به عنوان نوعی آسیب ایجاد شده در سازه، لحاظ می گردد. در این روش با بکارگیری خصوصیات تابع پاسخ فرکانسی و تبدیل هیلبرت تابع پاسخ فرکانسی مشخصه هایی جهت تشخیص و کمی ‌سازی تشکیل مفاصل پلاستیک ارائه شده ‌اند. با استفاده از پاسخ خطی و غیرخطی سه سازه قاب خمشی فولادی چهار، هشت و دوازده طبقه، تحت بارگذاری پانزده عدد رکورد انتخابی از زمین لرزه ‌های شناخته شده، با خصوصیات متفاوت در تاریخچه زمانی و دامنه فرکانسی، مشخصه ‌ها استخراج و مورد بررسی قرار گرفته ‌اند. بر اساس نتایج حاصله، مشخصه ‌های ارائه شده، در تشخیص تشکیل مفاصل پلاستیک در قاب های خمشی فولادی، برای سه مرحله پایش سلامت سازه شامل شناسایی وجود، جانمایی و شدت آسیب، دارای دقت و حساسیت مناسب می باشند.

detection of nonlinear seismic response of steel moment frames using frequency response functions (frfs) and hilbert transforms of frfs

occurrence of the nonlinear behavior can be a sign of changes in structural parameters and the presence ofdamage in systems. this paper presents a method for detecting and quantifying of nonlinearity, as an indicationof damage, using the indicators that are extracted from the frequency response functions (frfs) and hilberttransform of frfs, for steel moment frame structural systems. using time history analysis under selectedharmonic ground motions, the results of frfs for the studied 4 story system are illustrated and discussed.nonlinear behavior is a result of formation plastic hinges under earthquake loading. frfs and hilberttransform of frfs are extracted from both the linear and nonlinear behavior of 4, 8, and 12 stories steelmoment frames under fifteen different earthquake records with different characteristics in their time histories.some near and far field well known earthquakes records have been selected for the present study as the groundmotions input in time history analysis. different levels of nonlinearity are determined based on the maximumrotation of hinges in column members of structures equal to 2θy, 4θy and 6θy, in which θy is yield limitrotation. the indicators of the studied systems are calculated and evaluated for linear and different levels ofnonlinearity based on the mathematical power of changes for frfs and hilbert transform of frfs. thepresented indicators are extracted based on the frequency response functions (frfs) and hilbert transform offrfs for the responses of absolute acceleration and relative displacement of stories. the indicators arecalculated at the location of acceleration sensors (accelerometer) in four levels of the structural systems, whilethe formation of plastic hinges in the columns of the structures will occur only at the level of the distancebetween the adjacent sensors.three aspects of identifying “existence”, “location” and “extent” of plastic hinge rotation can be investigatedfor nonlinearity detection. in the present research, it is tried to cover the three aforementioned cases in theextracted indicators and presented results. according to the illustrated results it is concluded that:1 comparing harmonic loading results showed that frequency content of ground motions and existence offree vibration data in responses have proper effects on compatibility of frfs. it is concluded thatcompatibility of frfs are related to the frequency content of ground excitations and using the presentedindicators are not acceptable for any type of optional loading.2 the proposed method and calculated indicators have enough accuracy and sensitivity in detecting the“existence”, “location” and “extent” of damage.3 to identify “existence” of damage, using the indicators with higher mathematical power of changes, 𝑰𝟏𝒋𝑷𝜷􁈺𝑗: 𝑎 𝑎𝑛𝑑 𝑑􁈻 􁈺𝛽: 2􁈻, more certainty is included.4 to identify “location” of damage, the indicators that are extracted from frfs of displacement of monitoringstories, 𝑰𝟏𝒅 𝑷𝜷, can be used. 5 to identify “extent” of damage, using the indicators with lower mathematical power of changes, 𝑰𝟏𝒋 𝑷𝜷 􁈺𝑗: 𝑎 𝑎𝑛𝑑 𝑑􁈻 􁈺𝛽: 0.5􁈻, are more accurate.