Hello Ioannis Machaliotis,
Generally, initial battery energy of the node is set with respect to the application and the expected node capability.
For example, if you use a wireless node which is assumed to uses continuous power source, then setting any higher value for initial energy will not have any implication - because, in such case, you only try to measure the total consumed energy at the end of the simulation.
But, if you are using the same wireless node, for example, on a drone, then you can not assume a very high value for initial energy.
Similarly, if you are using a wireless node as a tiny sensor node, then its initial battery energy is based on the allowed maximum size of the node, expected life of the node/network, as well as the financial budget of that project.
On the other hand, in a simulation, you are interested to show the effect of link failure due to exhausting batteries, with in a short duration of simulations, then you may intentionally set very low initial energy to make that node 'dead' as early as possible.
So, simply, it depends on your application.
Charles Pandian.