Dear Hessein, for free vibration simulation are you talking about modal analysis? If yes I would assume it is not possible. As far as I know, for solving the modal analysis Abaqus uses the principle of superposition (which requires linear systems).

Best Regards,

Bruno

Thanks Claudio Pedrazzi and Bruno Luchini for your valuable answers. I recently tried to give a predefined velocity to my beam but It still does not work well. I think it is kind of impossible to do so.

You can do this using techniques usually employed for identifying the Volterra kernel of non-linear time invariant systems. Essentially, you apply a well-defined force or velocity input of varying amplitude to the system, calculate the system's response to each excitiation signal, and then decompose these results to obtain the non--linear transfer functions. There is plenty of literature on this subject.

You could make a very crude start by applying white noise of a given amplitude and look at the system's response. Resonances and apparent damping can be identified from the Fourier transform of the system's response.

Thanks all for your responses.

I was able to do free vibration in Abaqus via the idea of given a predefined velocity (initial step) and then let the system go. I get the responses; however, it takes forever to see some results. The converge issue was apparent too but at least it worked!!!

Hello Hessein,

I guess using ADINA it's possible. In any case you have do add a artificial soft spring to initiate the nonlinear iteration.