Calculation of volume fraction of buckypaper

I have been working on buckypaper (randomly distributed CNTs), where I’ve used a bead spring model for carbon nanotubes. I want to calculate the volume fraction of atoms in the simulation box.
I used Voronoi analysis in OVITO for atomic volume (volume of each atom). I summed it up for the total volume of atoms, but it is the same as the volume of the simulation box despite empty spaces between the beads?
Is there a command in lammps to calculate the volume of atoms and bonds only?
Please help!!

Thank You

This requires the answer to the more general question: what is the volume of an atom?
In the classical MD model atoms are points that have no volume.
Also in reality atoms are mostly empty space. The atom core and the electrons are tiny.
So where do you draw the boundary between an atom and empty space?
When you have a (dense) bulk system do atoms really occupy all the space? Or just a fraction? And how much?

Voronoi tesselation cannot work, because it divides the total volume. You need to look up how Voronoi tesselation works.

So this all comes down to the fact that the volume of an atom is an I’ll defined property. The only option is to determine an “effective volume” for your atoms, which could be based on covalent radii or by taking the average volume per atom in a dense system (one the you would consider 100% occupied.