Overcurrent Quench Detection of Parallel-Wound No-Insulation High Temperature Superconductor Coil Based on Digital Twin Method

Parallel-wound no-insulation (PWNI) high-temperature superconductor (HTS) coils demonstrate immense potential in fusion applications. During the charging process of PWNI coils, varying degrees of overcurrent occur within the coils due to non-uniform inductances and contact resistances, which even leads to overcurrent quench. While finite element analysis methods can effectively analyze the current characteristics within solid coils, their computational time is extensive, rendering them unsuitable for real-time applications. This paper proposes a model that combines finite element methods with artificial intelligence techniques. It can predict the maximum current amplitude within 0.039 s with a root mean square error of about 0.023, thereby preventing overcurrent quench. Through this approach, we aim to explore the digital twin method for PWNI magnets, with the goal of enhancing the safety and cost-effectiveness of superconducting magnet technology. Ultimately, this endeavor seeks to advance the prospect of utilizing nuclear fusion as a sustainable energy source.