Machine learning investigations and ab-initio study of thermoelectric properties of Half-Heusler compounds / Ahmed Lamin Ben Kamri

Public ISBD Titre : Machine learning investigations and ab-initio study of thermoelectric properties of Half-Heusler compounds Type de document : document multimédia Auteurs : Ahmed Lamin Ben Kamri, Auteur ; Ibn Khaldoun Lefkaier, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département des sciences de la matière Année de publication : 2025 Importance : 136 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Materials physics and nanostructures Langues : Anglais Catégories : THESES :17 physique Mots-clés : Machine learning  Ensemble learning  Thermoelectricity  Seebeck coefficient  Half-Heusler compounds Résumé : This dissertation explores the application of machine learning (Ensemble Learning)to predict the Seebeck coefficient in Half-Heusler compounds, focusing on both n-type and p-type compounds. Six machine learning models were developed and compared, utilizing only the chemical formulas of compounds. Through advanced techniques such as data preprocessing, feature selection, and hyperparameter optimization, the models demonstrated robust and accurate predictions. Performance metrics including R2, MAE, and RMSE confirmed the strong predictive power and generalizability of the models. Notably, LightGBM was most effective for n-type materials, while GBoost excelled for p-type compounds. The findings emphasize the importance of charge carrier concentration in predicting thermoelectric performance and highlight the models’ potential in screening materials for thermoelectric applications. Predictions made for approximately 60 half-Heusler compounds further validated the models, showing strong alignment with first-principles calculations and other studies. These results underscore the potential of machine learning to accelerate the discovery of high-performance thermoelectric materials. note de thèses : Thèse de doctorat en physique Machine learning investigations and ab-initio study of thermoelectric properties of Half-Heusler compounds [document multimédia] / Ahmed Lamin Ben Kamri, Auteur ; Ibn Khaldoun Lefkaier, Directeur de thèse . - Laghouat : Université Amar Telidji - Département des sciences de la matière, 2025 . - 136 p. + 1 disque optique numérique (CD-ROM). Option : Materials physics and nanostructures Langues : Anglais Catégories : THESES :17 physique Mots-clés : Machine learning  Ensemble learning  Thermoelectricity  Seebeck coefficient  Half-Heusler compounds Résumé : This dissertation explores the application of machine learning (Ensemble Learning)to predict the Seebeck coefficient in Half-Heusler compounds, focusing on both n-type and p-type compounds. Six machine learning models were developed and compared, utilizing only the chemical formulas of compounds. Through advanced techniques such as data preprocessing, feature selection, and hyperparameter optimization, the models demonstrated robust and accurate predictions. Performance metrics including R2, MAE, and RMSE confirmed the strong predictive power and generalizability of the models. Notably, LightGBM was most effective for n-type materials, while GBoost excelled for p-type compounds. The findings emphasize the importance of charge carrier concentration in predicting thermoelectric performance and highlight the models’ potential in screening materials for thermoelectric applications. Predictions made for approximately 60 half-Heusler compounds further validated the models, showing strong alignment with first-principles calculations and other studies. These results underscore the potential of machine learning to accelerate the discovery of high-performance thermoelectric materials. note de thèses : Thèse de doctorat en physique

SU(2) ×U(1) Formulation of GINZBURG-LANDAU équations / Ahmed Lamin Ben Kamri

Public ISBD Titre : SU(2) ×U(1) Formulation of GINZBURG-LANDAU équations Type de document : texte manuscrit Auteurs : Ahmed Lamin Ben Kamri, Auteur ; Salah Khenchoul, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département des sciences de la matière Année de publication : 2020 Importance : 34p. Format : 30cm. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Materials physics Langues : Anglais Mots-clés : Ginzburg-Landau equations  Breaking symmetry  Spin superconductivity  Abelian and non-abelian electromagnetic potential Résumé : Abstract Spin superconductivity is analogue of conventional charge superconductivity. Here, we derive a framework for describing the SU(2)-U(I) breaking symmetry in Ginzburg-Landau equations. We have found that it is not possible to directly insert or change an abelian electromagnetic potential with a non-abelian other one. The existence of the temporal components in the expression of breaking symmetry potential constant, indicates that the critical temperature can be controlled by applying an electric potential. note de thèses : Mémoire de Master en physique SU(2) ×U(1) Formulation of GINZBURG-LANDAU équations [texte manuscrit] / Ahmed Lamin Ben Kamri, Auteur ; Salah Khenchoul, Directeur de thèse . - Laghouat : Université Amar Telidji - Département des sciences de la matière, 2020 . - 34p. ; 30cm. + 1 disque optique numérique (CD-ROM). Option : Materials physics Langues : Anglais Mots-clés : Ginzburg-Landau equations  Breaking symmetry  Spin superconductivity  Abelian and non-abelian electromagnetic potential Résumé : Abstract Spin superconductivity is analogue of conventional charge superconductivity. Here, we derive a framework for describing the SU(2)-U(I) breaking symmetry in Ginzburg-Landau equations. We have found that it is not possible to directly insert or change an abelian electromagnetic potential with a non-abelian other one. The existence of the temporal components in the expression of breaking symmetry potential constant, indicates that the critical temperature can be controlled by applying an electric potential. note de thèses : Mémoire de Master en physique

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MP 01-77	MP 01-77	Thése	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

---

page 1/1