Investigation of the aging of the stress grading coatings of hydro-generators

Stress grading coating (SGC) are essential components for the proper functioning of preformed windings in large rotating machines. Indeed, the SGC present at the end of the semiconductor coating of the hydroelectric generator bars serve to control the electrical field to avoid the appearance of corona discharges. Premature degradation was observed on certain hydro-generators on the Hydro-Québec network during inspection five years after commissioning. In addition, corona discharges have been observed on generator bars presenting a high temperature in endwinding areas. In the long term, this degradation can lead to premature failure of stator bars and coils, generating significant repair costs and long-term shutdowns in a context of increasing power demand. The study of the aging of the SGC is part of a larger Hydro-Québec Research Center (IREQ) project.

This project will aim to study the evolution of the electrical properties of modern SCG (epoxy/SiC composites) as a function of aging and the impact of this evolution on the distribution of the electrical field in the enwinding areas as well as on the density current, and the resulting heating, circulating in the enwinding areas of the stator winding. The project will involve an experimental part as well as a multi-physics simulation part (mainly electrical and thermal). This study will not be limited to the behavior at power frequency but also at higher frequencies. SGCs are designed for power frequency operation and appear ill-suited to pulses and high frequency exposure, a growing phenomenon due to the penetration of renewable energy sources into the grid. Indeed, this type of excitation leads to higher current densities, concentrating on a few centimeters at the junction of the SGC and the slot semiconductive coating, causing an increase of several tens of degrees in the temperature at this location, which increase is likely to lead to accelerated aging of the coating. The increase in temperature can be mitigated by reducing the conductivity of the SGCs but this reduction comes at the expense of increasing the local electric field, which increases the risk of partial discharges occurrence. In addition, SGC suppliers have poor control over the exact electrical properties of their ribbon due to the variability in the electrical properties of silicon carbide from suppliers whose main market is abrasives.

The main deliverable of this project will consist of a model making possible to predict the aging behavior of SGCs as a function of the operating conditions. This model will take into consideration equipment wear and maintenance planning. Several papers, both in journals and in conference proceedings, will be produced during this work.