DC Electro-Magneto-Mechanical Characterization of 2G HTS Tapes for Superconducting Cable in Magnet System Using Artificial Neural Networks

Characterization of the exact critical current density (Jc) and stress values in twisted superconducting tapes plays an important role for analyzing their magnetic, thermal, mechanical behaviors. In study, a model based on artificial neural network (ANN) is introduced to estimate electro-magneto-mechanical characteristic different tapes. For this purpose, magnetic flux density, temperature, strain, total thickness tape, it's width, stabilizer, substrates are used as inputs ANN model, whereas minimum normalized Jc maximum considered outputs. The required experimental data extracted from published papers literature. was trained Jc/stress estimation using by inputs. Sensitivity analysis conducted models, which were tapes, choose optimum structure models be future other researchers superconductivity community. To check reproducibility, repeatability, stability presented results, estimations with tested 500 testing runs. We found that one hidden layer Levenberg–Marquardt training method seven Comparing literature, proposed offers about 15% 1.1% higher accuracy estimations, respectively.