Power Transformer Characterization and Design Optimization Environment

Abstract

Power Transformers are essential part of all AC Power Grids. This work proposes a characterization and design optimization environment for Power Transformers to increase efficiency while decreasing core volume. It utilizes a Finite-Element State-Space models in conjunction with Artificial Neural Networks and Particle Swarm Optimization. A case study of a 42 MVA, 118/13.8kV distribution power transformer was conducted, and transformer performance characteristics were compared to test data for verification. Next, the optimization environment was used to decrease the transformer volume and improve its efficiency. In addition, the Taguchi orthogonal arrays method was used to reduce computational costs.