Optimization Procedure to Design a Low-Losses MgB2 Wire

The EU-funded MARES project aims to develop a novel design of superconducting generator integrated into a wave energy converter (WEC). The superconducting generator consists of two parts, one assembled using REBCO tapes and the other using magnesium diboride (MgB2) wires, with the latter offering a more cost-effective alternative. Assessing the optimal design of MgB2 wires for this application is not straightforward, since AC losses represent a major technical issue of this device. In this work, a genetic algorithm is applied to determine the optimal design of the MgB2 wires. The algorithm is coupled with a finite element method (FEM) electrodynamic model that computes the AC losses in the MgB2 wires. The goal of the algorithm is to identify the key parameters of the wire configuration (filaments number and diameter, fill-factor, wire diameter, twist pitch) minimizing the AC losses in the device operating conditions.