Predicting the Superconducting Transition Temperature Tc of BiPbSrCaCuOF Superconductors by Using Support Vector Regression

According to an experimental data set on the superconducting transition temperature ( T c ) of 21 BiPbSrCaCuOF superconductors under different process parameters including the amount of bismuth ( n (Bi)), amount of oxygen ( n (O)), sintering time ( t ) and sintering temperature ( T ), support vector...

Full description

Saved in:
Bibliographic Details
Published inJournal of superconductivity and novel magnetism Vol. 23; no. 5; pp. 737 - 740
Main Authors Cai, C. Z., Zhu, X. J., Wen, Y. F., Pei, J. F., Wang, G. L., Zhuang, W. P.
Format Journal Article Conference Proceeding
LanguageEnglish
Published Boston Springer US 2010
Springer
Subjects
Characterization and Evaluation of Materials
Condensed Matter Physics
Condensed matter: electronic structure, electrical, magnetic, and optical properties
Exact sciences and technology
Magnetic Materials
Magnetism
Original Paper
Physics
Physics and Astronomy
Strongly Correlated Systems
Superconductivity
Theory and models of superconducting state
Support vector regression
Regression analysis
BiPbSrCaCuOF
Prediction
Superconducting transition temperature
Inorganic compounds
Superconducting transitions
Linear model
Oxyfluorides
Regression model
Neural networks
Sintering
High-Tc superconductors
Bismuth compounds
Online AccessGet full text

Cover

More Information
Summary:According to an experimental data set on the superconducting transition temperature ( T c ) of 21 BiPbSrCaCuOF superconductors under different process parameters including the amount of bismuth ( n (Bi)), amount of oxygen ( n (O)), sintering time ( t ) and sintering temperature ( T ), support vector regression (SVR) combined with particle swarm optimization (PSO) for its parameter optimization, was proposed to establish a model for prediction of the T c of BiPbSrCaCuOF superconductors. The performance of SVR model was compared with that of back-propagation neural network (BPNN) and multivariable linear regression (MLR) model. The results show that the mean absolute error ( MAE ) and mean absolute percentage error ( MAPE ) of test samples achieved by SVR are smaller than those achieved by MLR or BPNN. This study suggests that SVR as a novel approach has a theoretical significance and potential practical value in development of high- T c superconductor via guiding experiment.
ISSN:1557-1939
1557-1947
DOI:10.1007/s10948-010-0732-x