Oxidation Stability Enhancement Of Natural Ester Insulation Oil: Optimizing The Antioxidants Mixtures By Two-Level Factorial Design

• Main Authors: Ab Ghani, Sharin, Muhamad, Nor Asiah, Noorden, Zulkarnain Ahmad, Zainuddin, Hidayat, Talib, Mohd Aizam
• Format: Article
• Language: English
• Published: Asian Research Publishing Network (ARPN) 2017
• Subjects: T Technology (General); TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utem.edu.my/id/eprint/20829/2/jeas_0317_5811.pdf; http://eprints.utem.edu.my/id/eprint/20829/; http://www.arpnjournals.org/jeas/research_papers/rp_2017/jeas_0317_5811.pdf
• Institution: Universiti Teknikal Malaysia Melaka

Natural ester insulation (NEI) oil is known as an alternative solution for mineral-based insulation (MI) oil used in power transformer applications. Majority of NEI oil properties are reported to have better performance than MI oil. However, NEI oil still has limited properties especially in oxidation stability, pour point, viscosity and lightning impulse event. Recent studies have shown that the addition of selected mixed antioxidants into NEI oil has improves the oxidation stability properties. However, previous researchers only implemented one-factor-at-a-time (OFAT) method as their experimental design approach and might overlooked any possibility of mixed antioxidants at optimum ratios that will acquire more preferable result. Furthermore, a wide number of test runs are necessary to examine the effect of these antioxidant mixtures on the performance of NEI oil. Therefore, in this study, two-level (2k) factorial design of experiments is used to define the optimum concentration of propyl gallate (PG) and citric acid (CA) antioxidants which will amplify the oxidation stability of the oil. Oxidation induction time (OIT) is used to assess the oxidation stability of NEI oil according to ASTM E1858. The results show that the optimum concentration of PG and CA that yields the highest OIT of NEI oils is0.25 and 0.25 wt.%, respectively. A regression model is also established to estimate the OIT of NEI oil as a function of PG and CA concentrations.