Seismic damage diagnosis in adjacent steel and RC MRFs considering pounding effects through improved wavelet-based damage-sensitive feature

Abstract

This paper aims to propose complex Morlet (cmorfb-fc) wavelet-based refined damage-sensitive feature (rDSF) as a new and more precise damage indicator to diagnose seismic damages in adjacent steel and Reinforced Concrete (RC) Moment Resisting Frames (MRFs) assuming pounding conditions using acceleration responses. The‏ ‏considered structures ‏include 6- and 9-story steel and 4- and 8-story RC benchmark MRFs that are assumed to have ‏different values of separation distances, δMT, calculated according to the ASCE 7-10 seismic provision. For the sake of pounding modelling, ‏linear viscoelastic contact elements among the pounding structures are assumed. Furthermore, an algorithm is developed to compute the seismic collapse capacities of each pounding MRF through Incremental Dynamic Analyses (IDA) using OpenSees software. In the next step, auto-regressive moving-average with exogenous input (ARX) model together with a stabilization diagram is utilized to appraise the natural frequencies of each adjacent MRF. Shannon entropies and correlation coefficient (ρ) are used to select the best cmorfb-fc wavelet. Based on the results, damage resulting from pounding effects can be accurately detected in both cases of with and without δMT, especially for steel MRFs.