stiffness of a body

huangyhg

stiffness of a body
the problem that i am having is being able to justify that the member that i am analyzing is connected to a rigid body.  i am in the process of design highway vessels and they require rollover protection.  my analysis is based on the assumption that the tank with stiffener rings are considered rigid in order for me to validate the design of the rollover protection.
i have considered a case where the rollover protection is connected directly to the shell (with a repad) and that only that plate material will resist the forces and moments translated through from the forces applied to the upper part of the rollover protection.   calculating the moment of inertia for the shell and the pad i get a certain value. then i have considered a case where the rollover protection is mounted around the stiffener rings (flat bar rolled on edge and welded to the tank).  recaculating the moment of intertia, we find that with similar sized repads, and only the addition of the stiffener ring, that same section of shell is now has a moment of inertia that is 16x greater than the previous case (both m.o.i's considered about the n.a. of the geometry)  does this constitute rigid if the shells are to be considered non rigid?
also, how would this apply if the stiffener ring is close by where "edge effects" with the proximity of the stiffener ring may play a role.  if it does play a role, how far away until these effects can be neglected.
obviously in the real world there is no such thing as a perfectly stiff/rigid structure, but i need to learn how i can determine " the degree of stiffeness" in order for me to understand the risks in making a rigid base assumption.
thanks
as you told, there is no perfectly stiff structure, sufficient rigidity may only be a relative concept.
this means that you must be able to compare the stiffness (or deflections) of your tank to the deflections in the rollover protection that are acceptable in order to confirm the assumption of rigid support.
of course by using a stiffened vessel wall there is an impressive increase in stiffness, but, again, for this to be considered sufficient (or possibly oversized) you must also analyze the requirements for the protection.
you need also to consider the resistance of the vessel wall: even if it can be considered as perfectly rigid with respect to the protection, it will have to be able to sustain the load transmitted to it.
prex
simplifying assumptions are only useful up to a point.
possibly in the case your are analysing, with the interconnection of curved shell and beam elements, it may be better to do a full finite element model.