Contents I. Introduction
Electrical transformer is a significant device for converting voltage level for the power system. Conventionally electrical transformer is designed by using trial and error technique, however some problem may be occurred occasionally with either the expensive cost or the unexpected performance. Generally optimization design of transformer is focused to either minimum manufacturing cost or maximum transformer efficiency. Recently for developing transformer performance the finite element method and the artificial intelligent (AI) technique have been mentioned in many literatures [1]–[6]. For example application of AI technique for evaluating transformer loss can be achieved [4]–[5] that function of core design parameters are predicted by means of artificial neural networks (ANNs). Georgilakis and colleague also have used artificial neural networks to reduce the core loss of assembled transformers while the production process of individual cores is optimized by using the Taguchi method [5]–[6].
