Parametric studies of a stiffened suspension structure

Abstract

The behaviors of various stiffened suspension structures were studied under various loading conditions. The properties and dimensions of the structural members were changed in order to study how these variations affected the behavior of the structures. Each structure was a synclastic suspended two-way orthogonal cable network with stiffening gridwork. The responses of the structures under load were analyzed by a three-dimensional elastic frame computer program with nonlinear capabilities, which was developed at Rice University. The structures were analyzed and designed to meet present structural requirements. For ease of presentation and applicability to other structures, the results of the analyses have been presented with respect to two dimensionless parameters. The two parameters were general measurements of the stiffhess of the structure and the finenesses of the stiffening gridwork and cable network. The structural responses studied were cable tensions, structure deflections, and stiffening gridwork member forces. The design and weights of the supporting structure, foundations, columns, etc., were not considered. The structures were analyzed for six loading conditions. The stiffening gridwork for each was composed of steel trusses and open-web bar joists. For the gridwork members in one structure nine flexural stiffnesses were computed based upon nine span-to-depth ratios. With the stiffening gridwork having each of the above flexural stiffnesses, the structure was analyzed under each loading condition. In the various structures, the center-to-center spacing of the cables and gridwork members varied in a range from fifty feet to one hundred feet. A structure with a span-to-depth ratio of fifty was scaled according to the laws of dimensional analysis so as to be one and one-half times its original size and was analyzed using the same loading conditions. Within the limits of the results obtained by the computer analyses, the structures studied appeared to be feasible and to meet AISC allowables. However, since the analyses were performed assuming the loads to have been concentrated at the joints, the analysis did not account for local bending produced by the application of loads between joints. Assuming the loads to have been uniformly distributed, the gridwork members to have been continuous, and the node points to have been unyielding pinned supports, the distribution of moment produced under these assumptions was superimposed onto the moment distribution obtained from the computer analyses. These combined moments were used as approximations of the true moments which might have occurred in the gridwork. Under the limitations of this study, one attractive solution for a structure of this type would have a fairly fine gridwork (5 to 8 bays) with a span-to-depth ratio of about sixty.