On-line combined use of neural networks and genetic algorithms to the solution of transformer iron loss reduction problem

Abstract

A new approach using neural networks and genetic algorithms to solve the transformer iron loss reduction problem is proposed in this paper. Neural networks are used to predict iron losses of wound core distribution transformers at the early stages of transformer construction. Moreover, genetic algorithms are combined with neural networks in order to improve the grouping process of the individual cores by reducing iron losses of assembled transformers. Results from the application of the proposed method on a transformer industry demonstrate the feasibility and practicality of this approach.