Intelligent and Application-Oriented Optimal Design of Travelling Field Flux Pumps

Abstract

Flux pumping based on traveling field is a promising technology, potentially able to produce breakthrough innovation in the supply of HTS magnets, which offers a contactless, low-voltage, and high-current alternative to power electronics exciters and current leads solutions. However, their engineering process has proved to present major challenges. Previous studies have empirically investigated, either numerically or experimentally, the impact of individual design parameters on the outputs and performance of flux pumps, but they were only able to provide qualitative relations that are not suitable for proper designing actions. In this study, we propose a new approach based on artificial intelligence (AI) techniques to generate effective flux pump designs. A finite element (FE) model, previously validated against experimental results, was employed in this procedure to provide a relation between the design parameters of the flux pump and the objective function of the optimization problem, that is the maximum efficiency during persistent operation. The FE model is exploited in the form of a function that is fed into AI-based optimization algorithms such as the genetic algorithm and the particle swarm optimization. The established procedure offers a “systematic” method for the design of viable and efficient flux pumps for contactless energization HTS magnets in real applications.