It all depends on you up to what length you want an impact because when accident occur you don't decide how much it has to crush , it just happens and depends on impact energy. More fluctuations may occur as the folding initiates from the top of the tube and the whole geometry will no longer be of uniform section.

Generally, the exerted force on the end of the tube (usually un-impacted end) is called the crushing force. So to plot force-displacement curve you don't need the force of the longitudinal elements. In fact, in case of axial impact, the crushing force of tube is independent of its length. It depends on cross sectional area, material property etc.

Moreover, to plot the force-displacement curve, you have to obtain the crushing force (vs time) of the cross sectional area at one end and the displacement of a node located at the crushing end of the tube to track the shortening of the tube (vs time).

if the time intervals of the force-time and displacement-time curves are the same, then you can have the force displacement curve.

Also note that the fluctuations are related to formation of the folds. When a fold starts to form, a peak appears in force-time curve. Usually the maximum peak appears at the beginning of the process and is considered as the maximum crushing force which is an important factor in design of thin-walled tubes as energy absorption systems.

You should cut your f-d graph before the densification part. Peak and plateu parts are enough for crushing.