A Novel Method to Enhance the Core Design of Power Transformers Using Particle Swarm Optimization (PSO) Technique

Omar Sharaf Alyozbaky,M. Kadir,M. Izadi
DOI: https://doi.org/10.5013/IJSSST.A.19.06.42
2019-02-03
Abstract:The reduction of losses in a power transformer is a significant topic in a power transmission grids, with the design of the core being one of the important factors related to the losses. In this study, an intelligent algorithm is developed using Particle Swarm Optimization (PSO) technique to find the optimum core design. Using a smart algorithm we consider a novel method to reduce the losses. To achieve this target, a nonlinear B-H curve with a 3D Finite-Element Method (FEM) is used to simulate the proposed model of core coupling with the PSO technique. Steady state analysis is considered in the simulation and the losses of the core and windings were considered a heat source inside the transformer. The results show core losses are reduced by 11 % and 5 % was achieved by the proposed model compared with the 90° butt-lap and 45° mitred designs respectively. The results were validated against the manufacturer’s experimental measurements. Our proposed optimization techniques can be used to improve performance of electrical power transformers and can be applied to various transformers of different sizes and power rating.
Engineering,Computer Science
What problem does this paper attempt to address?