1. Study of high temperature superconducting tapes for applications in high-field magnets and power systems
- Author
-
Tian, Mengyuan and Coombs, Timothy
- Subjects
AC loss ,high temperature superconductor ,HTS cables ,Superconducting tapes - Abstract
Nowadays, the application of fusion in achieving large-scale, sustainable and carbon-free form of energy is urgently needed. In the design of fusion machines, the construction of superconducting magnets with high field and low inductance is very important. This requires superconducting magnets to be able to carry very high current even with a small conductor curvature (less than 50 mm in some special cases). Due to the high current density and high tolerance to axial tensile and compressive strain of RE − Ba2Cu3O7−σ (REBCO) coated conductors, the CORC® cable is designed as a powerful superconducting cable wound with REBCO coated conductor tapes in multiple layers which can meet these challenging requirements. The Conductor on Round Core (CORC) cable is one important type of superconducting cabling concept which is capable of high current densities. The special structure of CORC cable determines that it is a high-performance, practical and robust conductor with mechanical and electrical isotropy, which is suitable for high-field magnets and compact superconducting power cables. Although it has these advantages, further optimization and research of CORC cable is critical to push its technology to higher performance levels. The spiral geometry of CORC and high aspect ratio of superconducting tapes form challenges for exploring its electromagnetic characteristics, such as current density distribution, AC loss and magnetic field shielding effect. This study is therefore set out to investigate the electromagnetic characteristics of CORC cables by providing comprehensive analyses through experiments and multi-layer numerical models. The main contribution of this thesis is its unique and innovative approach of a new two-dimensional (2D) simulation model, which is applied for the first time to model CORC cables with real spiral structure. In addition, the application of deep-learning neural network into AC loss calculation of CORC cables sets the first attempt to calculate the loss using a statistical method. The simulation and statistical models proposed in the thesis are verified by experimental results. A systematic understanding of how the winding angle, diameter of core, number of superconducting layers and the external applied field influence the electromagnetic behaviour of CORC cables is presented. First of all, a novel 2-D simulation method which can dramatically reduce the simulation time without loss of accuracy has been proposed. The simulation results are summarized and discussed based on both H-formulation and T-A formulation under changing magnetic fields and transport currents. Both models have been validated by experimental measurements, proving their feasibility for quickly calculating the AC loss of CORC cables. Furthermore, this newly proposed 2-D method is able to model the rotating structure of CORC while saving calculation time and simulating the magnetic field shielding effect. This simplified model can also be applied to study the electromagnetic performance of other HTS applications with similar geometries where the traditional 2-D or 2-D axisymmetric models are not applicable. Secondly, a comprehensive comparison of CORC cables up to four layers is presented, which have been modelled using both T-A formulation and H-formulation in three-dimensional (3-D) coordinates. The computation efficiency, magnetization loss, current density and mag- netic flux distribution are studied and compared. The AC loss study of CORC cables with different thicknesses of superconducting layers and with or without copper former has been conducted in order to understand how the AC loss changes when the magnitude and frequency of applied external fields vary. The simulation data and the results gained from experiments show a great agreement, demonstrating that the 3D CORC cable model is accurate and reliable. Next, based on the data obtained from the simulations done in the previous chapters, this thesis has conducted regression analysis using a variety of machine learning methods, and has shown that the AC loss of a spiral structured CORC cable can be accurately and quickly calculated through deep-learning neural network. This provides a new approach for future AC loss calculations and a compelling design approach for industrial applications such as HTS magnets. Typical design considerations, including airgap, the amplitude of the external magnetic field, winding angle and the number of layers, which have covered the typical design considerations, have been investigated. Finally, high-temperature superconducting capacitors are investigated using simulation method to test its performance in a superconducting wireless power transfer system. In particular, the component simulation models of capacitor and coil have been developed using finite element simulation. This process verified that the transfer efficiency is highest when we use the HTS coil together with HTS capacitor instead of traditional components.
- Published
- 2023
- Full Text
- View/download PDF