Fuzzy Measurement Algorithm for Fault Detection in the Hydrogenerator

Fault detection involves the detection, location, and identification of faults. Three types of
unbalances (mechanical, hydraulic and magnetic) can occur in hydrogenerators. There are two main
reasons for the existence of magnetic unbalance: shorted turns in the windings of the rotor poles and
geometric asymmetry of the air gap. In this study, the authors propose a novel fuzzy measurement
algorithm for fault detection in the hydrogenerator. The proposed algorithm has been developed with aim to detect the presence and cause of the magnetic unbalance and vibrations of the hydrogenerator. The study is based on fuzzy theory and three measurement methods for measuring inner and outer magnetic fluxes and measuring mechanical vibrations. The fuzzy algorithm has to ensure, in situ and real time, magnetic and vibration monitoring and fault detection and as such it is suitable for implementation in on-line and real time monitoring systems. The main benefit of the proposed solution lies in the support for the optimal decision making regarding the predictive maintenance.