Benchmark of the OpenSC2 model in transients with significant current redistribution in superconducting cables for nuclear fusion

Superconductivity is a critical enabling technology for magnetic confinement nuclear fusion. Different types of cables are under study in this field, with much effort in the scientific community focused on the use of high-temperature superconducting (HTS) cables, which enable the generation of more powerful magnetic fields. Numerous designs have been proposed and studied using various modeling software to predict the behavior of these cables under operational conditions. For LTS cables, various software programs (such as 4C, THEA, and VENECIA) can model both the electrical and thermohydraulic profiles. However, they are not recommended for the modeling of HTS cables for which the H4C code was introduced some years ago. To address this need, OpenSC2 was developed by the MAHTEP group at Politecnico di Torino. This open-source program facilitates multi-physics analysis for both LTS and HTS cables. The thermal-hydraulic model available in OpenSC2 has already undergone a preliminary phase of verification and validation. The modeling capability of OpenSC2 was improved by the possibility of coupling the thermal-hydraulic model to a lumped electric model based on a network of resistances, inductances and conductances, which is described in this thesis. The objective of this work is to benchmark the coupling between the thermal-hydraulic and the electric models in OpenSC2 against the results obtained with the H4C tool for a quench study of the 132 m long ENEA HTS cable. It carries a nominal current of 32.1 kA and is placed in a magnetic field of 17 T. Cooling is achieved using ~5 g/s of supercritical helium at 4.5 K and 0.6 MPa, allowing it to be introduced downstream of the same cryostat as the LTS cable, but with higher temperature margins in case of a tape quench. The quench of the cable is induced with a heat pulse 0.1-s-long of 150 J in 10 cm of the cable. After 0.25 s from the quench detection, the current is dumped according to an exponential law. Temperature peaks as well as current time evolution are compared to the outcomes of the H4C tool, based on the discharge rate of the current and the delay time in which this occurs. To achieve this goal, it was necessary to suitably discretize the cable in sub-components; by exploiting the symmetry conditions present in the case study, it was found that 1/12 of the total geometry was representative of the cable. The choice of materials used to model the different sub-components was as accurate as possible, considering the real composition of the ENEA cable. However, assumptions are always necessary to simplify the analysis. Finally, the quality of the results is guaranteed by space and time convergence analysis. Overall, the outcome of this analysis has shown that, duly considered the difference between the two software, OpenSC2 achieves the same results predicted with 4C within a 8-10% accuracy. Therefore, for this test case, it can be claimed that the coupling of the thermal-hydraulic and of the electric model in OpenSC2 is consistent: during the transient as the electric resistance of the superconductor increases, the current is progressively redistributed on the stabilizer according to their electrical properties becoming the main driver for the thermal-hydraulic problem.