Estimation of dynamic design parameters for buildings with multiple sliding non-structural elements using machine learning

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).