A simple concat and sort_values could work:
concat
sort_values
pd.concat((dtrijen, dtvolkel)).sort_values(by='datetime')
But can you explain what you mean (no pun intended) by "keeping the average"?