OPTIMIZATION OF THE MIXING PROCESS PARAMETERS TO ENHANCE THE DIELECTRIC STRENGTH OF MINERAL AND PALM FATTY ACID ESTER INSULATING OIL BLENDS

S.N. Norhan, H. Zainuddin, S.A. Ghani, I.S. Chairul
2017 Journal of Electrical Systems  
Mineral insulating (MI) oils are well-established as insulating oils in power transformers for almost 100 years owing to their low cost, wide availability and good dielectric performance. However, a large number of studies have been carried out over the years to develop substitutes for MI oils since these oils are non-biodegradable and moreover, they are derived from petroleum, which is a non-renewable source. Blending MI oils with natural ester insulating (NEI) oils is an innovative approach
more » ... ich fulfils this purpose since this approach has been proven to enhance the dielectric properties of insulating oils. However, there is a dearth of studies concerning the effects of mixing process parameters on the dielectric strength of MI-NEI oil blends. This information is important in order to produce insulating oils with favorable dielectric properties. In this study, the two-level (2k ) factorial design is used to determine the significance of mixing process parameters (specifically stirring speed and temperature) in influencing the dielectric strength (AC breakdown voltage) of an MI-NEI oil blend. The oil blend is produced by mixing Nytro Libra MI oil with palm fatty acid ester (PFAE) oil at a volume ratio of 80:20 using a hot plate magnetic stirrer. The stirring speed and temperature is varied from 250 to 550 rpm and 30 to 90˚C, respectively. AC breakdown voltage test is performed using Megger OTS60PB portable oil tester in accordance with the ASTM D1816 standard test method to determine the dielectric strength of the oil blends, whereby the gap distance between the electrodes is 1 mm. Based on the results of the 2k factorial design, it is found that the combination of stirring speed and temperature has the most significant effect on the AC breakdown voltage, with a percentage contribution of 60.45%. A regression model is also developed to predict the AC breakdown voltage as a function of the stirring speed and temperature. Response surface methodology (RSM) is used to optimize the mixing process parameters in orde [...]
doaj:92e3d2e9b06c4423a17d123c8d1f379a fatcat:kzkmwff7o5dhzh2zqfvug34mui