Objectives This study investigates long-span, double-deck plate frame structures commonly used in ship construction. The main objectives are: to comprehensively analyze the stability of these structures under axial compression and to develop a method for accurately evaluating their ultimate load-carrying capacity. This research is of great significance as it provides essential technical support for the design and safety assessment of ship structures, ensuring the reliability and safety of ships with long-span double-deck plate frame structures.

Method A comprehensive approach is adopted to achieve these objectives. Firstly, an ultimate strength test is conducted on a specially designed long-span double-deck structure model. The test model is fabricated from Q550 structural steel with detailed dimensions. A closed-frame loading system is utilized during the experiment, along with various measurement devices, including hydraulic jacks, force sensors, and displacement gauges, to accurately record load and displacement data. Additionally, a three-directional resistance strain gauge is used to monitor stress variations at multiple locations on the model. Moreover, the initial defects of the model are measured and incorporated into the analysis, and the mechanical properties of the material are determined through tensile testing. Secondly, numerical simulations are performed using ABAQUS. A finite element model is constructed with proper boundary conditions, and initial defect data are incorporated to improve the simulation accuracy.

Results The test results reveal that global buckling involving both the upper and lower decks is the main cause of the failure of the long-span double-deck structure. Pronounced buckling is observed on the upper deck near the loading end, with the deformation gradually propagating toward the side plates. The stress analysis indicates that the axial load is unevenly distributed between the upper and lower decks, with the lower deck carrying a smaller proportion of the load. The load-displacement curve shows that the structure begins to undergo inelastic deformation when the load reaches approximately 500 t, with the ultimate load of 583.4 t. The simulation results show that the simulated load-displacement curve generally agrees well with the experimental curve, although it slightly overestimates the ultimate strength, with an error of 8.5%. Moreover, the simulation shows that both buckling in both the upper and lower decks, which differs from the experimental result of the lower deck. Further analysis reveals that eccentric compression has a significant impact on the failure mode and the ultimate load-carrying capacity of the structure.

Conclusion In conclusion, it is crucial to account for the effect of eccentric compression in practical simulations of long-span double-deck plate frame structures. Including this factor in the analysis improves the accuracy of the evaluation, thereby benefiting the design and safety assessment of ship structures. This research provides a valuable reference for future studies on similar structures and offers practical guidance for shipbuilding engineering.