Fnu Tabish

Date of Award

January 2023

Document Type


Degree Name

Master of Science (MS)


Civil Engineering

First Advisor

Iraj H.P D. Mamaghani


This thesis aims to estimate and improve the buckling strength of the cylindrical storage tanks under static loadings. Initially, for the verification of the overall performance of the numerical modeling approach; a computational analysis was conducted to calculate the linear buckling behaviour of empty cylindrical shells with different H/D and D/t ratios using ANSYS workbench 2021. Results revealed that the FE models accurately predict static critical buckling stress which is mainly depends on the D/t ratio. The solution of the buckling analysis provides multiple buckling mode shapes and critically buckling load values. Those mode shapes (eigenvectors) can indicate the expected buckling modes during the nonlinear analysis.

For steel made cylindrical specimens subjected to external pressure; varying R/t and H/R ratios strongly influence the critical buckling pressure. The buckling pressure remarkably increases with the decrease of both R/t and H/R ratios; however, the effect of the R/t ratio is more dominant than the H/R ratio. The Results revealed that the geometric imperfections have little influence on the overall buckling capacity, especially for tanks with large H/R ratios and smaller R/t ratios. Numerical results show good agreement with experimental and theoretical results; however, FEA gave higher results, especially for cylinders with smaller R/t ratios might be due to neglecting imperfections that are probably created in the construction process.

For Aluminium made thin-walled ring stiffened cylindrical specimens subjected to the external pressure; A comprehensive finite element (FE) numerical study investigated the influence of external ring stiffeners varying from 3 to 17 on a thin-walled, stiffened aluminium cylindrical shell buckling strength. Ten ring-stiffened cylindrical specimens were modeled using an ANSYS workbench 2021 whose stiffener dimensions varied so that all specimens' overall weight remained constant. FE linear and nonlinear buckling results were compared with the experimental work and the theoretical formulas in the literature. The failure mode shapes and number of circumferential lobes at failure for all specimens obtained from the linear analysis closely matched the experimental failure pattern. The linear buckling pressures were lower than the corresponding experimental critical pressures; however, they compare well with the buckling pressure obtained from the theoretical equations. The nonlinear buckling pressures for perfect geometries are lesser than the experimental pressures, and specimens with nine or fewer stiffeners were crushed instead of buckling at failure. For nonlinear analysis of imperfect geometries based on the eigenmode shape, results revealed that the 5 % imperfection giving the failure mode shapes similar to the experimental buckling shapes for most of the specimens, and local shell buckling pressures were closer to the experimental buckling pressures compared to the overall flexural buckling results. The overall FE results indicate that the failure mode types shifted from shell local buckling mode to the flexural buckling mode while increasing the number of ring stiffeners by keeping the specimen’s overall weight constant. Parametric study reveals that linear and nonlinear buckling strength remarkably improved by keeping a constant stiffener height compared to the FE buckling strength for specimen dimensions obtained from experiments, especially for specimens that failed with overall flexural buckling mode. The experimental, theoretical, and finite element (FE) results proved that the ring stiffener’s optimum size and spacing could improve the stiffened cylinder buckling strength since critical buckling pressure and failure mode shape were influenced by the ring stiffener’s size and spacing.

Available for download on Sunday, February 25, 2024