TY - JOUR
T1 - Machine learning based prediction of metal hydrides for hydrogen storage, part I
T2 - Prediction of hydrogen weight percent
AU - Rahnama, Alireza
AU - Zepon, Guilherme
AU - Sridhar, Seetharaman
N1 - Funding Information:
This work was supported by the Serrapilheira Institute (grant number Serra- 1709-17362 ).
Funding Information:
In this paper we utilise the Hydrogen Storage Materials Database, which is an openly available database that can be accessed through http://hydrogenmaterialssearch.govtools.us/ . The creation of this database was funded by the U.S. Department of Energy as a contribution to the International Energy Agency Hydrogen Implementing Agreement ( http://ieahydrogen.org/ ) [10] .
Publisher Copyright:
© 2019 Hydrogen Energy Publications LLC
PY - 2019/3/15
Y1 - 2019/3/15
N2 - The openly available database provided by the US Department of Energy on hydrides for hydrogen storage were analyzed through supervised machine learning to rank features in terms of their importance for determining hydrogen storage capacity referred to as hydrogen weight percent. In this part: we employed four models, namely linear regression, neural network, Bayesian linear regression and boosted decision tree to predict the hydrogen weight percent. For each algorithm, the scored labels were compared to the actual values of hydrogen weight percent. Our investigation showed that boosted decision tree regression performed better than the other algorithms achieving a coefficient of determination of 0.83.
AB - The openly available database provided by the US Department of Energy on hydrides for hydrogen storage were analyzed through supervised machine learning to rank features in terms of their importance for determining hydrogen storage capacity referred to as hydrogen weight percent. In this part: we employed four models, namely linear regression, neural network, Bayesian linear regression and boosted decision tree to predict the hydrogen weight percent. For each algorithm, the scored labels were compared to the actual values of hydrogen weight percent. Our investigation showed that boosted decision tree regression performed better than the other algorithms achieving a coefficient of determination of 0.83.
KW - Artificial intelligence
KW - Hydrogen storage materials
KW - Machine-learning
KW - Metal hydrides
UR - http://www.scopus.com/inward/record.url?scp=85061800108&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85061800108&partnerID=8YFLogxK
U2 - 10.1016/j.ijhydene.2019.01.261
DO - 10.1016/j.ijhydene.2019.01.261
M3 - Article
AN - SCOPUS:85061800108
SN - 0360-3199
VL - 44
SP - 7337
EP - 7344
JO - International Journal of Hydrogen Energy
JF - International Journal of Hydrogen Energy
IS - 14
ER -