Can the standard stator bar condition assessment tests be replaced by reflectance spectroscopy?

Condition assessment is an imperative for effective power system asset management and there is a pressing need to rapidly acquire data on the operational performance of high voltage solid and liquid electrical insulation systems, both during the manufacturing process as well as in the field. These systems are governed by a number of manufacturing, operational and environmental variables which affect the time to failure in complex ways. Although forensic examination of solid insulation failures are commonly conducted and there are many electrical tests to assess the condition of the insulation system, there has been limited effort to relate remote, non-destructive analysis methods to rapidly assess chemical characteristics and physical properties, and to directly correlate these to factors that can affect failure. We demonstrate that an extension of engineers´ visual inspection methods can provide correlations to conventional test methods - the results then not only provide a predictive capability for insulation condition assessment which can support improved approaches to manufacturing, insulation life prediction and asset management but also show insights into the chemical and physical parameters that effect the desired properties. In addition, we suggest that these approaches can be “wired into” the components through embedded fiber optics and sensors, providing a continuous diagnostic check on the insulation systems.