I see a large variation between min and max values. Perhaps multiplying the "average" flux by the overall area is too approximate. Can you use the Probe="Reaction" to get the total wattage at the interface? It will work on boundary conditions and contact. Total Heat In should balance Total Heat Out, but sometimes these reaction probes do not match exactly if there are boundary conditions that share entities (edges and areas) on the surface(s) in question.
You might have looked at this already, but if not, please refer to Mechanical Users Guide => Using Results => 'Understanding Averaged and Unaveraged Contour Results'. Refer also to MAPDL Theory Manual 6.3 on mathematical expressions used for Heat Flow Evaluations and Section 17.6.2 on error approximation techniques for temperature-based problems.