The identification of water status is the foundation for fuel cell water management, which is helpful to fuel cell reliability and longevity. In this article, a novel and reliable method for diagnosing the hydration condition of proton exchange membrane fuel cells (PEMFCs) was presented using a fractional-order model (FOM) to represent the PEMFCs impedance. The results show that the mean root-mean-squared error (RMSE) and mean absolute percentage error (MAPE) between the proposed model and experimental data (in normal, drying, or flooding cases) are about 0.034 and 0.473, respectively. The fast Fourier transform–electrochemical impedance spectroscopy technique (FFT-EIS) was used as an alternative technique that is simple and efficient to electrochemical impedance spectroscopy (EIS). The PEMFCs hydration state is monitored by observing the changing effect of the physical resistor values (membrane resistance, polarization, and diffusion resistances) of the proposed model. These resistors, characterized by their high sensitivity to the drying and flooding of PEMFCs, affect the Nyquist impedance spectra and frequency spectrum amplitudes at low and high frequencies. Based on the obtained results, it is concluded that the proposed strategy can be used to develop new domains in which the PEMFCs’ hydration states can be properly predicted.