Title: Estimation of dynamic design parameters for buildings with multiple sliding non-structural elements using machine learning
Authors: S.P. Challagulla; Chandu Parimi; S. Pradeep; Ehsan Noroozinejad Farsangi
Addresses: Department of Civil Engineering, BITS-Pilani, Hyderabad, 500078, India ' Department of Civil Engineering, BITS-Pilani, Hyderabad, 500078, India ' Department of Civil Engineering, BITS-Pilani, Hyderabad, 500078, India ' Faculty of Civil and Surveying Engineering, Graduate University of Advanced Technology, Kerman, Iran
Abstract: The seismic behaviour of a primary structure (PS) with multiple sliding non-structural elements (NSEs)/secondary bodies (SBs) is investigated in this paper. Equations governing the motion of PS with SBs have been developed considering the Coulomb's friction model. Spectrum compatible ground motions associated with the two Indian seismic hazard levels were considered. A parametric study was performed to analyse the variation in the displacement of the structure by varying the structural period (Tp), the mass ratio (αi), and coefficients of friction (μsi, μki). The results of the parametric study demonstrate that the sliding blocks behave as rigidly attached bodies to the structure for higher structural periods and frictional constants. A novel method is proposed to calculate the modified structural period (Tnew) with multiple sliding rigid blocks. Finally, design equations for Tnew are proposed by utilising the machine learning technique like artificial neural network (ANN).
Keywords: primary structure; secondary bodies; Coulomb friction; sliding element; seismic hazard levels; non-structural element; NSE; artificial neural network; ANN.
International Journal of Structural Engineering, 2021 Vol.11 No.2, pp.147 - 172
Received: 30 Apr 2020
Accepted: 08 Jun 2020
Published online: 15 Apr 2021 *