Optimization of the Safety Margins of the Major Insulation on Oil-Immersed Transformers

The performance of the insulation system is directly linked to transformer life. Therefore, achieving the best insulation system is a key objective during the transformer design. Over the years, many dielectric design criteria have been developed, such as the Weidmann<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> reference curves and safety margin calculation. This work focuses on optimizing computed safety margins of oil gaps related to the probability of inception of partial discharges in transformer oil without increasing the overall dimensions of the active part of the original transformer design. These margins must be adequate to guarantee that the equipment supports test levels and the dielectric stresses to which it will be subjected during its useful life. An optimization methodology based on genetic algorithms (GA) is developed to maximize safety margins in the major insulation arrangements of a power transformer, considering barriers, oil duct dimensions, and dielectric material properties for mineral or natural esters. The results show that the genetic algorithm yielded considerable improvements in the design of the insulation system of a 20 MVA transformer based on its insulation safety margins, calculated considering the applied potential and impulse voltage test voltages.