Dielectric Frequency Response and temperature dependence of power factor

Modern technology and developments in signal acquisition and analysis techniques have provided new tools for transformer diagnostics. Of particular interest are dielectric response measurements where insulation properties of oil-paper systems can be investigated. Dielectric Frequency Response, DFR (also known as Frequency Domain Spectroscopy, FDS), was introduced more than a decade ago and has been thoroughly evaluated in a number of research projects and field tests with good results. DFR data in combination with mathematical modeling of the oil-paper insulation is proven as an excellent tool for moisture assessment. Since the modeling theory contains influence of temperature, DFR and modeling can be used to calculate the temperature dependence of the insulation system. This paper gives a background to DFR, insulation modeling and how these tools can be utilized to improve understanding of power factor temperature dependence and how this can be used for decisions on maintenance and/or replacement.