SSI on Geodesic Dome Using RSM and Comparison with ANN

In the present study, design of the steel truss of the existing geodesic dome is carried out by using the codebook IS 800:2007 (code of practice for general construction in steel). The dynamic analysis is also carried out by using response spectrum method (RSM) as per IS 1893(Part 1):2016 (the criteria for earthquake resistant design of structures). The MAT foundation is designed for geodesic dome models for 30 m diameter. To incorporate the soil-structure interaction (SSI), the standard soil mechanical properties are obtained by soil test. Finite element analysis (FEA) of the geodesic dome is carried out using SAP 2000 V22. The artificial neural network (ANN) model is developed by using MATLAB 2021Ra.119 and soil spring values are calculated by varying the standard penetration number (SPT), shear wave velocity, density, and Poisson’s ratio for different soil conditions. ANN model is developed with four inputs as soil spring values and four outputs as base shear values for zone II, III, IV, and V using the Levenberg–Marquardt algorithm. It is observed from the study that the ANN model can predict the dynamic characteristics and seismic response of the soil-structure system with 98% accuracy compared to the results obtained by FEA.