Design Optimization of Cast-Resin Transformer Using Nature-Inspired Algorithms

摘要

In the recent years, cast-resin dry-type transformers (CRT) are considerably used in various areas of the electrical power industry. Themajor properties of these transformers include inflammability and moisture protection. This is the reason that they are an appropriate alternative for oilimmersed transformers in residential, hospitals, ships and other areas. Large number and discontinuity of design parameters cause the design optimization and automation in the CRT to be complicated. In this paper, structure of the CRT and its design parameters are proposed. Three methods based on genetic algorithm, particle swarm optimization (PSO) and artificial bee colony are introduced for optimization and design automation of the CRT. These methods are applied to design a typical transformer, and the results are compared and discussed. It is shown that PSO is the best optimization technique which converges to optimal solution quickly.