obtain necessary experimental data with, which to check the predictions of 

 theory. 



Further requirements may dictate the use of other than circular 

 cylindrical pressure hulls, such as those possessing either deep-draft or 

 wide-beam cross sections. To meet these needs and retain the same 

 degree of rationality that exists for the structural analysis and design of 

 circular cylindrical pressure hulls, experimental and analytical studies 

 are required. Under joint sponsorship by the Bureau of Ships and the 

 Office of Naval Research, the group at the Polytechnic Institute of 

 Brooklyn has been conducting theoretical studies of the elastic behavior 

 of noncircular cylindrical shells under hydrostatic pressure. Kempner, 

 Romano, and Vafakos have already reported some of their findings in a 

 series of PIBAL reports. One of the more significant results they 



have found so far is that for the case of a short cylinder having a quasi- 

 elliptic cross section and either simply supported or clamped edges, it 

 appears that one can use the local radius of curvature in the equations for 

 the right-circular cylinder to get excellent prediction of the longitudinal 

 and circumferential stresses in the shell. Although one might argue that 

 intuitively this result is not at all unexpected, it also turns out that no such 

 simple analogy can be deduced for the case of the elastically supported 

 (ring-stiffened) quasi-elliptic cylinder and recourse must be had to the 

 exact solution which is based on a Fourier series approach. As an 

 adjunct to these analytical studies, model tests are required to obtain 



131 



