Prediction of AC Breakdown Voltage of Mineral Oil Nanofluid

Abstract

AC breakdown strength is a critical parameter in the performance of an insulating liquid. Developing a method to predict the breakdown strength of dielectric liquid makes it possible to predetermine the voltage withstand capability of materials. Prediction of breakdown voltage of nanofluids (NFs) has great significance to the optimization of filler loading concentration of NFs. Based on the electric field distribution inside the dielectric liquid samples, this paper proposes a prediction method for AC breakdown voltage (BDV) using curve fitting technique. By analyzing various statistical parameters of electric field distribution such as average, standard deviation, variance, skewness, and kurtosis along with AC breakdown voltage values of NFs, a relationship connecting these parameters and breakdown voltage is formed. Using this relation and statistical parameters of electric field distribution obtained from simulation model, breakdown voltage is predicted for another set of NFs. TiO 2 NFs are used for framing the equations and breakdown voltage is predicted for Al 2 O 3 NFs. It is observed that AC breakdown strength is predicted with least error by using kurtosis as the statistical parameter. Test results show that the relative errors between predicted values and actual values are all less than 8%, which indicates the accuracy and reliability of this method.