In case of ideal piezoelectric you did not put into account the dielectric and conduction losses. In real system your piezoelectric it's naturally saturated. Moreover, generally needed to match the output impedance of piezoelectric and active load impedance.

Regards

please check the impedance mismatch. Also Since check the time constant of the circuit.

Note that time constant of Rload x Capacitance_piezo may have impact on the signal shape. Also, are you certain it is saturation - is signal shape more sinusodal-like at lower excitation?

Sergejus Borodinas

Thanks

Do you mean the loss factors in COMSOL for PZT material shown in the picture?

Do you know how I find the loss factors matrixes for PZT-5H material?

Yes, in Comsol such parameters. I recommended for you paper below.

Regards

Also, a picture of a typical piezoelectric cantilever has been added [1].

[1] H. Li, C. Tian, and Z. D. Deng, "Energy harvesting from low-frequency applications using piezoelectric materials," Applied physics reviews, vol. 1, p. 041301, 2014.