how to caculate the bond length distribution and bond angle distribution?

Please reply to the list, thanks.

Since you used a many-body potential where bonds and angles are “estimated” on the fly based on the instantaneous bond order and no true bonds and angles exist, you wouldn’t be able to use compute bond/local, angle/local, nor property/local commands. You are best off writing a small post-processing program to analyze your dump files – this is what I did for my paper ~6 years ago….

Hope this helps.