Optimal design of a furnace transformer by intelligent evolutionary methods

Abstract: This paper presents three intelligent evolutionary optimization techniques to investigate the optimal design parameters of a 3-phase furnace transformer. The transformer rating is derived from the operating conditions of a medium size direct arc furnace. Scatter Search (SS), Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) techniques are employed on the developed nonlinear mathematical model of the transformer for constrained optimization minimizing the cost. The design and analysis programs of the furnace transformer are developed using codes written in C++/C language. The optimal design data results validated by an example show the efficacy of the three intelligent techniques. Among the three methods, the optimal results obtained by GA and PSO techniques show the potential for implementing as efficient search techniques for design optimization of furnace transformers. [Copyright &y& Elsevier]