A Practical Approach for Static Eccentricity Fault Diagnosis of Hydro-Generators

There has been an increased recognition that more attention needs to be paid to the condition monitoring of hydro-generators to provide a more reliable and sustainable power production. The static eccentricity (SE) fault is one of the prevalent types of failure in the wound rotor synchronous generator that leads to the outage of the power unit. Many different solutions have been proposed for SE fault detection. The techniques are normally based on an offline approach which is not an industry favorite. In this paper, a novel statistical approach is applied to the air gap magnetic field that can detect the fault in its early stage. Furthermore, the severity of the failure on the proposed index in different loading conditions is studied. A polar diagram of the air gap magnetic field is used to detect the location of SE. Finally, the ability of the statistical indices to detect the SE fault in under load variation are verified by experimental results.