Structural analysis of ancient stupa in Sri Lanka

Abstract

Ancient stupas in Sri Lanka hold immense historical, archaeological, and religious significance. These monumental structures, dating back to ancient times, serve as tangible links to the country's rich cultural heritage. Therefore, it is imperative to restore and conserve these structures with great care so that future generations can witness the remarkable engineering technology of the past. However, it is essential to acknowledge that some restoration attempts in the past have resulted in catastrophic failures. These failures were mainly due to the complex structural behaviour of these massive structures due to complex geometries and different loading conditions. Consequently, these misguided restoration efforts have not only failed to preserve the historical value of the stupas but have also led to their degradation. The purpose of this research is to tackle the limitations found in current finite models and understand how they affect the restoration work. To achieve this aim, the study has two main objectives. Firstly, it investigates whether shifting from a 2D axisymmetric analysis to a more detailed 3D analysis is necessary. This change could significantly impact how successful restoration efforts are. Secondly, the research examines the lasting effects on stupas caused by moisture-induced expansion and the impact of temperature changes. Both of these factors play a crucial role in the restoration of stupas. Deegawapi stupa was taken as the case study to achieve the objectives of the research. Then, 2D axisymmetric and 3D non-axisymmetric models of Deegawapi stupa were analysed under self-weight using the finite element package ABAQUS. Results indicated minimal divergence in maximum stress values between the two models. Notably, comparing full 3D analysis to partial (half and quarter) stupa analyses reveals negligible differences in outcomes, accompanied by significantly reduced computational time in wedge analyses. Accounting for material non-homogeneity between the original and new components of the stupa, there was stress variation in the structure compared to the homogeneous model. Notably, there was a high stress concentration between the interface of new and old material. However, stress concentrations at this interface were within material strength limits. Given that ancient stupas are predominantly constructed from clay bricks and a butter-like plaster, both of which are highly susceptible to moisture-induced expansion, this study conducted an analysis to assess the implications of this phenomenon. The analysis outcomes revealed a notable concentration of high stress at the interface between the new and old components, which exceeded the strength of both the new and old materials. Considering that these ancient structures are primarily situated in arid regions, they are subject to elevated thermal loads arising from intense solar radiation. To replicate the impact of such conditions, a thermal load ranging from 35°C to 65°C was applied to the stupa's exposed surface. The subsequent analysis of stress distribution revealed that the recorded stresses remained within the material strength.