A question on the co-simulation by ADS and HFSS of DRO

A question on the co-simulation by ADS and HFSS of DRO

I am working on a DIELECTRIC RESONATOR OSCILLATOR(DRO) at 5.8GHz.
And I plan to simulate the DR structure with Ansoft HFSS to get the S2P file,
and then import it to Agilent ADS to co-simulate with the transistor.
However, simulation results show that the loop gain is less than 1, which means that the circuit cannot oscillate.
Actually, experimental results show that it oscillates quite well, and the output power is between -1dBm and 0dBm.
Could anyone tell me the reason why the simulation does not coincide with the experiment?
Thanks very much!

Attachment:
Figure 1--ADS Schematic
Figure 2--Loop Gain, Note that the loop gain is quite near 1(more than 0.99), and the phase is close to 0.

Please forgive me for my poor English! I am from Chinese.

申明：网友回复良莠不齐，仅供参考。如需专业帮助，请学习易迪拓培训专家讲授的ADS视频培训课程。