Hi dear

In my opinion, you must right TCL Code in order to model non-prismatic beam and in order to nonlinear analysis the beam model you can use section Fiber.

Dear Vasudeo,

The common way is to divide the beam into smaller segments and assign different sections (and/or materials) to each smaller element. Most common situations can be modeled using from 5 to 10 segments. You can define a number of Fiber Section for which you should assume a suitable length.

Best of luck,

Masoud

Dear Rafael and Masoud

Thank you so much for your replay, I shall try to make model as per your suggestions.

Vasudeo

Dear Vasudeo,

I would like to advise you that the approach proposed by Masoud (and often used both in research and engineering practice) may lead to coarse results.

Indeed, as discussed in the paper Article Non-prismatic Beams: a Simple and Effective Timoshenko-like Model

The paper mentioned above provides a more accurate modeling strategy that you can easily implement. In any case, I recommend to verify the accuracy of the results with some highly refined 2D or 3D FE analysis.

All the best,

Giuseppe

Dear Giuseppe

Thanks for sharing the details

Vasudeo

Dear Vasudeo

I think in addition to discretizing in length, you can also use displacement based beam column element, by which you can assign linear changing cross section

Pooya