Multi-Objective Computational Aided Design Tool Using Pareto Analysis

Due to commercial restrictions, aiming for greater product competitiveness, uninterruptible power supplies (UPS) must exhibit optimal design in order to reduce losses, volume, and noise, as well as increase power density. However, the optimal design of a power electronic converter is a complex and repetitive task, often requiring the use of a computational tool. The main objective of this work is to present a unified computational tool capable of helping the designer to find the optimal design for a given power electronic converter. This paper will show a comparison between two power converter design techniques: the traditional industrial design approach and a multi-objective optimization approach using Pareto analysis. For this purpose, the total performance of a 10 kVA UPS was compared in terms of losses and volumes. The use of the proposed computational tool to optimize the design resulted in a UPS with better power density and efficiency