Critical buckling load prediction of axially compressed cylindrical shell based on non-destructive probing method. (June 2019)

• Record Type: Journal Article
• Authors: Fan, Haigui
• Is Part Of: Thin-walled structures. Volume 139(2019)
• Page Start: 91
• Page End: 104
• Publication Date: 2019-06
• Subjects: Critical load -- Buckling -- Cylindrical shell -- Imperfection -- Non-destructive; Thin-walled structures -- Periodicals; 690.1
• DOI: 10.1016/j.tws.2019.02.034
• Languages: English
• ISSNs: 0263-8231

Abstract

Abstract: Critical buckling load prediction of axially compressed cylindrical shell is investigated based on the non-destructive probing method in this paper. Finite element model of the cylindrical shell under combined axial load and radial probe is established using ABAQUS and the static modified Newton-Raphson method that uses artificial damping is chosen in the geometrically and materially nonlinear buckling analysis. By the means of repeatedly probing the shell under different prescribed axial loads, the three-dimensional representation of probe force, probe displacement and prescribed axial load is established. Based on that, the critical buckling load of the shell is predicted by the fitting curve that reflecting the relationship between the maximum probe force and the prescribed axial load. Applying this prediction method, the perfect cylindrical shell, the cylindrical shells with dimple-shape imperfections and the cylindrical shell with measured imperfections have been studied. All of the predicted results are compared with the real critical loads in buckling behavior of the shells under axial compression. Effects of poker size and probing location on the prediction results are analyzed. Results show that the critical loads for the first buckling pattern of the cylindrical shells with notable local imperfections can be predicted accurately when the shell is probed at the location with the largest imperfection amplitude. If the shell is probed away from that area, the predicted result will become much larger. Compared to the probing location, the poker size has little effect on the prediction results. For the cylindrical shell with measured imperfections, probing at the location with largest imperfection amplitude has achieved the most accurate prediction result. Besides, the predicted critical buckling loads obtained by probing at other locations are also acceptable. It is believed that for a general cylindrical shell without notable local defects, the smallest one of different predicted results obtained by probing a series of representative locations on the cylindrical shell could be regarded as the critical buckling load. Highlights: Critical load of axially compressed cylindrical shell is non-destructively predicted. First buckling pattern of the shell with local defect can be predicted efficiently. The shell should be probed at the most deleterious defect for accurate prediction. The shell with measured defects can be predicted by probing a series of locations.