I hope someone from the community will respond further, in case my answer regarding the impossibility of two different excitations with different phases in a harmonic analysis is false.
However theoretically (but I have never tried) it might be conceivable to perform two, three separate harmonic analyses, one for each excitation, and then do a superposition (linear combination) of them (available in Workbench).
You are navigating away from the AIS Discovery experience