Multi-response optimization of the properties of natural ester oil with mixed antioxidants using taguchi-based methodology

• Main Authors: Ghani, S. A., Muhamad, N. A., Noorden, Z. A., Zainuddin, H., Ahmad, A. A.
• Format: Article
• Published: Institute of Electrical and Electronics Engineers Inc. 2017
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/76249/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-85022335039&doi=10.1109%2fTDEI.2017.006589&partnerID=40&md5=795933c36da26e2cfa2c8d5b381d0d48

A Taguchi-based methodology is proposed in this study in order to determine the optimum concentration of antioxidants which will enhance the oxidative stability of natural ester insulation (NEI) oil while ensuring superior dielectric strength performance. Conventional experimental testing involves the procurement of materials and equipment, sample preparation, experimental set-up and testing the samples with a large number of test runs to determine the optimum design parameters, which is costly and time-consuming. In this regard, the Taguchi-based methodology is advantageous since it can predict the optimum design parameters with fewer test runs. This methodology can also be used for multi-response optimization by applying a desirability function. The number of test runs for the antioxidant mixtures is generated using the L9 orthogonal array with three levels and two factors. This methodology limits the need to deal with the uncertainties associated with the range of factors, unlike the one-factor-at-a-time (OFAT) method. The oxidative induction time (OIT) and AC breakdown voltage (AC BdV) are selected as the output responses in accordance with the ASTM E1858 and ASTM D1816 standard test method, respectively. Analysis of variance (ANOVA) is used to determine the percentage contribution of each factor as well as the significance level of the model developed from the Taguchi-based methodology. The desirability function is used to transform the two output responses and generate the optimum value which complements both of the responses. Repeated tests are conducted to verify the repeatability and reproducibility of the OIT and AC BdV results for the optimized NEI oil based on a 95% confidence interval. In general, there is a significant enhancement in the OIT and AC BdV for the optimized NEI oil compared to those for fresh NEI oil. Hence, it can be concluded that the Taguchi-based methodology is a feasible approach to determine the optimum concentrations of mixed antioxidants for NEI oils.