The use of SHM (structural health monitoring) techniques has shown promising results for fault detection in rotating machines, making possible to identify various malfunctions. SHM methods provide maintainability and safe operation for these systems. The objective of the present work is to evaluate the SHM method based on the electromechanical impedance (ISHM) to detect faults in a composite rotor shaft. Composite materials present complex damage mechanisms due to their anisotropy and heterogeneity. Moreover, the process of damage detection in these materials is more challenging than in metallic structures. The ISHM approach uses piezoelectric (PZT – Lead Zirconate Titanate) patches as sensors and actuators coupled to the monitored structure. Variations in their electrical impedance are associated with changes in the mechanical integrity of the system. The electrical impedance of the PZT sensor is directly related to the mechanical impedance of the structure, which changes according to variations in the mass, stiffness, and damping properties of the structure. Damage metrics are used to quantify variations in the electrical impedance (impedance signatures) of the PZT patches. Despite the ISHM approach be able to detect incipient faults, it presents some disadvantages. For instance, the impedance signatures are susceptible to temperature variation. In the present contribution, to detect damages in the considered composite rotor shaft, the ISHM technique was implemented based on a data normalization methodology. Thus, an optimization procedure based on hybrid optimization was used to avoid false diagnostics.