A theoretical and experimental investigation of pallet rack structures under sway

A theoretical and experimental investigation into the factors affecting the accuracy of beam-end connector tests showed that the flexibility of the stub beam used in the cantilever test can induce errors of up to 4% in the experimentally determined moment–rotation relationships. Appropriate correction equations are derived in this paper. The corrected moment–curvature equation was then used in an investigation into the sway behaviour of racking structures under a combination of point loads normal to the ground and with proportionally increasing side loads. The experimental tests used a mirror arrangement of two portal frames lying horizontally. During each test three cycles of loading and unloading were performed before the frame was taken to its ultimate load. Initially the bending moments on both beam-to-column joints had the same sign. The percentage of side load applied to the structure was chosen so that, due to the effect of frame geometry change, reversal of bending moments occurred in one of the joints during the test. The beam and column components of the frame were numerically modelled using stability functions. The beam–column connections were semi-rigid and the experimental moment–rotation curves were incorporated into the connection behaviour. Three different theoretical models were tested: a multi-linear model proposed by the authors and two bi-linear models produced according to the FEM and SEMA codes. The agreement between theory and experiment, for all load cases (including the unloading parts of the experimental cycles), was excellent for the multi-linear model. The FEM code model overestimated the sway deflections occurring in the rack but gave accurate predictions of joint moments and could be conservatively used in design. The SEMA code consistently underestimated the sway deflections and was unable to predict joint moment reversal. Once the theoretical model had been validated, a study was then undertaken into the behaviour of a two-bay portal frame structure.