An Alternative Approach to the Buckling Resistance Assessment of Steel, Pressurised Spherical Shells

Abstract

Provisions leading to the assessment of the buckling resistance of pressurised spherical shells are available since 2008 when they were published first time as the European Design Recommendations (EDR) (cf. Rotter and Schmidt in Buckling of Steel Shells: European Design Recommendations. ECCS, 2008 [13], Rotter and Schmidt in Buckling of Steel Shells: European Design Recommendations. ECCS, 2013 [14]). This collection of recommendations comprises rules which refer to the buckling resistance of steel shells of different shapes. In the first step of the general procedure, the calculation of two reference quantities: the elastic critical buckling reference p_Rcr and the plastic reference resistance p_Rpl is required. These quantities should be determined in the linear buckling analysis (LBA) and in the materially nonlinear analysis (MNA) respectively. Only in the case of spherical shells the existing procedure has exceptional character. It is based on the geometrically nonlinear analysis (GNA) and on the geometrically and materially nonlinear analysis (GMNA), respectively. From this reason, in this particular case there was a need to change the existing provisions. The first version of a new procedure was presented in the work of Błażejewski and Marcinowski (Buckling capacity curves for pressurized spherical shells. Taylor & Francis Group, London, pp. 401–406, 2016 [4]). All steps of the procedure leading to the assessment of buckling resistance of pressurized steel, spherical shells were presented in that work. The elaborated procedure is consistent with provisions of Eurocode EN1993-1-6 (cf. Błażejewski and Marcinowski in The worst geometrical imperfections of steel spherical shells, pp. 219–226, 2014 [3]) and with general recommendations inserted in Europeans Design Recommendations. In the present work the proposed capacity curves were compared with the existing provisions of ECCS for three different fabrication quality classes predicted. Comparisons of the author’s proposal with some experimental results obtained by other authors are presented as well. They have confirmed that the proposed procedure is less conservative than the existing one but it is still safe.