Help with FEA related topic for 4th year project

Hi there, I'm currently working on my interim report for my 4th year project on the structural analysis of naval ships as a result of underwater explosions. My original plan for the modelling stage was to create a detailed model of the midship section in Rhino and then import it into ANSYS Mechanical APDL to carry out the simulations. I was wondering if there was any general advice anybody could offer regarding this approach and if doing this in ANSYS is feasible, I was told that I could simulate the water acting on the structure by way of applying general loading conditions opposed to actual simulation of the fluid.

I do have some concerns whether I can find sufficient information online in order to create a detailed model of the midship section however, I was advised to go about this approach by my adviser but as expected, arrangements of naval ships aren't easy to find online. This has resulted in me questioning whether its worthwhile doing a detailed analysis of the midship section, with the stiffeners, scantlings, etc modeled, or to just create a full ship model, excluding such specific details but offering a broader look at the overall effects of this shock loading being applied.

 

Explosions have to be modeled differently than a regular load. The velocity at which the load is applied has to be considered. The toughness of the material and structure (ability to absorb energy without breaking) is the limiting factor. Can you use any ship for the calculations? Older designs that are de-classified are available.

 

Yeah just really any naval vessel, ideally not one would be entirely obsolete by today’s standard would suit. Do you have any websites that are particularly good for getting arrangements?

Do you also know if modelling explosions in ANSYS is feasible?

 

Are you familiar with the framing systems employed on these sorts of vessels? Can be longitudinal, transverse or combination system. Could just model the mid-body section that is of sufficient length for your analysis assuming the top sides are wall sided. Which type of hull form did you prefer?

Which element are you using in FEA? I’d suggest using shell elements. That way it’s easier to adjust scantlings during the analysis

 

Way way too much.

Firstly the amount of detailing required would take forever to model. Secondly, how/where will you be able to verify the results?

Thus I would model a "generic hull" between WTBs. This reduces the model. It will also be quicker to run!
Then you should be able to find data on similar generic WTB-WTB hulls of various loading scenarios.

Bottom line is...You need to model 'something' that has been presented to verify and validate your own model and assumptions. Since you must validate your work...without which, it is just a bunch of pretty colour pictures!

 

Thanks for the advice, so to be clear are you saying to just model the hull between the bulkheads and keep the internal framing as simple as possible in order to reduce the time to run it?

I've not really been given much reference in terms of how time consuming this project will be, so any guidance is greatly appreciated!

 

Probably a combination of the framing systems, although might have to simplify it a bit to save time. As for the hull form, I am not entirely sure at the moment, it's really just a case of if I can find sufficient general arrangements online of past naval ships, if not then a generic slender hull will most likely be used.

Yeah I plan to use shell elements for FEA, although do you happen to know if importing models from Rhino into ANSYS works well enough without breaking the geometry?

 

Yes.

The accuracy of the hull form is not in question.
It is the data of frame spacing v stiffness and how the scantlings of each influence the results.

Since is a plate thickness of say, 10mm, with 400x6mm web with 100x10mm frames spaced at say 1250mm apart, better than....XXX..or YY...etc....thus apply sensitivity analysis of different frame spacings, stiffener spacings and the scantlings of each. This will allow you to arrive at a conclusion of stiffness v weight for example. .

 

It works fine if done correctly. There is little room for error and all surfaces must align perfectly with no duplicate lines or stray keypoints. I might have a tutorial that will help with correct file format for shell elements and a few other tricks. I’ll try dig it up if it’ll help

 

Okay I see what you mean. Thanks a lot for the help!

 

Yeah that would be great, thanks a lot.