Any energy conversion devices, such as industrial motor-drives, propulsion drives of electric vehicles, pump systems, wind turbines, and others, are prone to failures. Usually, failures result in increased economic costs that come through additional energy losses, loss of production, or in a worst-case even environmental hazard. To prevent failures, energy conversion systems may be checked through particular routines developed and specified by the manufactures. However, it may be challenging due to the complex construction of energy conversion devices or devices' failure between the routine checks. Such schedule-based condition monitoring approaches provide minor information on the remaining lifetime (separate components and whole system) of the devices and do not allow proper prognostic or full exploitation. To overcome traditional two-level Boolean approaches with healthy/faulty states an Artificial Intelligence (AI)-based control techniques are used. The Fuzzy Logic approach is based on inspired by human perception processes and cognition that are often uncertain or empirical. However, Fuzzy Logic is already successfully applied in various control applications of energy conversion devices, even when the analytical models are unknown. This paper argues for developing new fault detection algorithms based on fuzzy logic methods to allow energy conversion systems designers to develop reliability factors for apparatus, which included electrical machines and power electronics subsystems.