Detection of stator core faults in large turbo-generators

Many large turbo-generators today are now operating beyond their design lifetime, with careful condition-based maintenance and operation having extended their life expectancy. Monitoring the health of large turbo-generators is now an integral part of their operation. This paper describes the use of non-invasive electromagnetic sensors to detect core inter-lamination insulation faults in the stator cores of large turbo-generators before they propagate to a level that can and has lead to catastrophic failure. The work develops a deeper understanding of the electromagnetic behavior of core faults so that condition-monitoring sensors can be more specific about the location and severity of the fault. The paper develops a new analytical model for predicting the fault current distribution in a stator core fault. This is verified using detailed 2D FE models which additionally predict the electromagnetic sensor signals so that a clearer understanding of the location and severity of the fault is possible. These are validated on a specially constructed experimental test core that can impose controlled fault currents at various fault locations around the core. The paper importantly demonstrates how small core faults will escalate then self-limit radially, but can propagate axially into longer more damaging faults, and also demonstrates the ability of the electromagnetic sensor to detect small core faults inside the core before they become severe.