Magnetic properties prediction of NdFeB magnets by using support vector regression

Autor:	Wende Cheng
Rok vydání:	2014
Předmět:	Support vector machine Neodymium magnet Materials science Nuclear magnetic resonance Mean absolute percentage error Magnetic energy Remanence Magnet Applied mathematics Particle swarm optimization Statistical and Nonlinear Physics Coercivity Condensed Matter Physics
Zdroj:	Modern Physics Letters B. 28:1450177
ISSN:	1793-6640 0217-9849
DOI:	10.1142/s0217984914501772
Popis:	A novel model using support vector regression (SVR) combined with particle swarm optimization (PSO) was employed to construct mathematical model for prediction of the magnetic properties of the NdFeB magnets. The leave-one-out cross-validation (LOOCV) test results strongly supports that the generalization ability of SVR is high enough. Predicted results show that the mean absolute percentage error for magnetic remanence Br, coercivity Hcj and maximum magnetic energy product (BH) max are 0.53%, 3.90%, 1.73%, and the correlation coefficient (R2) is as high as 0.839, 0.967 and 0.940, respectively. This investigation suggests that the PSO-SVR is not only an effective and practical method to simulate the properties of NdFeB , but also a powerful tool to optimatize designing or controlling the experimental process.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::ead383ad6b3713e6c56cc0807644362c https://doi.org/10.1142/s0217984914501772 Zobrazit plný text záznamu