Objective Increasingly stringent safety and environmental requirements for large tank farms drive the widespread adoption of assembled full-contact internal floating roofs with honeycomb reinforcement. Structural features such as numerous buoyancy modules, diverse layouts, and closely spaced bolted joints create complex in-service stress states in these floating roofs. Understanding how structural configuration affects the mechanical behavior of assembled floating roofs under typical operating conditions is therefore essential for floating-roof design optimization, safety assessment, and accident analysis.

Methods Stress in an assembled stainless steel floating roof with a grid-shaped beam layout is measured during hydrostatic testing of a 1,000 m³ storage tank. The equivalent theory for honeycomb panels and a simplified modeling method for bolted connections are used to develop a full-scale finite element model of an assembled full-contact floating roof. Stress and deformation distributions under normal operation and extreme roof jamming are analyzed to reveal how buoyancy-module layout and beam structure affect the mechanical behavior of the floating roof.

Results During the hydrostatic test of the storage tank, field measurements yield stress-response data at different locations on the assembled floating roof. These data validate both the simplified modeling method and the finite element results. Under normal operation, stresses in the roof with a grid-shaped beam layout and in the beamless roofs with staggered and herringbone layouts are all well below the yield strength of stainless steel. Structural deformation varies with the layout direction. Under full jamming, the maximum stress in every roof configuration occurs along the vertical edges at the bolted connections. The beam system carries part of the load and effectively reduces the maximum von Mises stress in the shells of the buoyancy modules. Under diagonal jamming, the buoyancy-module layout determines the anisotropic mechanical behavior of the floating roof. Stress responses differ markedly with the direction of jamming.

Conclusion The findings reveal how structural configuration affects the mechanical behavior of assembled full-contact internal floating roofs. Mechanical performance should be analyzed and evaluated during the design or retrofit of floating roofs for on-site storage tanks. Areas prone to stress concentration should also be identified. The overall aim is to improve the inherent safety and operational reliability of floating-roof tanks.