Direct Strength Analysis - Integral Asphalt Carriers

Dear Members,

I am currently working on my thesis project, which involves Finite Element Analysis (FEA) of a cargo hold in an integral tank asphalt carrier operating at 180°C. I have a few questions regarding the setup of my FEA model, as I am using BV 2024 rules for guidance.

Questions:

1. Corrugated Bulkheads and Dummy Rod Elements: According to the BV rules, when a corrugated bulkhead is subjected to liquid cargo or ballast, dummy rod elements with a cross-sectional area of 1 mm² should be modeled at the corrugation knuckle between the flange and the web. Does anyone know the purpose of inserting these dummy rods? Is it primarily to increase stiffness, or is there another reason?
   
   
2. Modeling Ordinary Stiffeners: The rules specify that ordinary stiffeners should be modeled as "beam elements." However, can these stiffeners be modeled as shell elements instead? The process of correctly inserting stiffener eccentricities, directions, and other details can be quite tedious.
   
   
3. Mesh Convergence Test Necessity: Is it necessary to perform a mesh convergence test for ship FEA models? The BV rules provide guidelines on mesh dimensions, such as coarse and standard mesh. I am considering using a 50 mm mesh throughout my cargo hold model since I have sufficient computational power. Can I proceed without conducting a mesh convergence test, given that the mesh dimensions are within the specified limits?

For my analysis, I plan to perform both linear static structural and linear buckling analysis using ANSYS. My project ship has a LOA of less than 150 m.

Below, I am sharing some of the initial FEA analysis results. Please note that this is not the final model—I have made several adjustments since then, and the initial setup had some inaccuracies. This was my first attempt at ship FEA, but it should give you a general idea of the model.

THERMAL GRADIENT 180ºC:


FULLY LOADED CONDITION:



ALTERNATE LOADING:


Thank you for your time and assistance. I look forward to your guidance on these matters.

Best regards,

 

Welcome to the forum Alex.T

It is likely to ensure shear flow and provide a ‘degree’ of fixity that differs between hand calculations (there are several options) and FEA.

You could, but I would not.
Beam elements are the correct elements to use and will provide 100% correct results, unlike the shell elements, which require high order elements and increased modelling time.

You still need to demonstrate that whatever mesh size you have, or started with, that it does converge.
Thus your initial mesh size, is not relevant, in that sense.

 

Well noted.

Thank you for the welcome, Ad Hoc, and for your quick response .