Key points are not available for this paper at this time.
Statistical studies have shown that failures of bushings, winding insulation, and online tap changers are the main causes for long-duration outages of transformers. This article investigates the development of an instrument for supervising the conditions of transformer units. The use of sectional winding transfer functions (SWTFs) for online PD evaluation in power transformers has several advantages: localization of PD sources, discrimination between PDs inside or outside of the transformer, and evaluation of actual PD amplitudes along the winding. Direct measurement of SWTFs-when different points along the coil are accessible-and use of these SWTFs for PD evaluation and localization has shown excellent results. In order to use this method at transformers on site, an appropriate modeling method is needed for the computation of the SWTFs using only measurements at the transformer terminals. A detailed modeling approach based on discrete RLC circuit elements has been studied and different algorithms used for parameter estimation and optimization. This model is applicable in practical cases for a limited frequency range if genetic algorithms (GAs) are used for parameter optimization. A new method based on traveling wave theory has been investigated using genetic algorithms to search for the optimum parameters of a partial differential equation that describes the transient behavior of the coil. This method has shown potential in solving SWTF calculation problems for different transformers.
Akbari et al. (Sun,) studied this question.