REPOSITORY QDB RESOURCES NEWS CONTACTS

Saldana, D. A.; Starck, L.; Mougin, P.; Rousseau, B.; Pidol, L.; Jeuland, N.; Creton, B. Flash Point and Cetane Number Predictions for Fuel Compounds Using Quantitative Structure Property Relationship (QSPR) Methods. Energy Fuels 2011, 25, 9, 3900–3908.

QsarDB Repository

Saldana, D. A.; Starck, L.; Mougin, P.; Rousseau, B.; Pidol, L.; Jeuland, N.; Creton, B. Flash Point and Cetane Number Predictions for Fuel Compounds Using Quantitative Structure Property Relationship (QSPR) Methods. Energy Fuels 2011, 25, 9, 3900–3908.

QDB archive DOI: 10.15152/QDB.123   DOWNLOAD

QsarDB content

Property FP: Flash point [K]

Compounds: 631 | Models: 5 | Predictions: 20

FP_PLS-MD: PLS-MD model for flash point i

Regression model (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 445 0.936 14.387
Test set i external validation 63 0.859 21.994
Validation set i external validation 123 0.936 13.703
Prediction set i testing 528 N/A N/A
FP_SVM-GD: SVM-GD model for flash point i

Support vector machine (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 445 0.966 10.550
Test set i external validation 63 0.942 13.861
Validation set i external validation 123 0.953 11.685
Prediction set i testing 528 N/A N/A
FP_NN-MD: NN-MD model for flash point i

Neural network ensemble (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 445 0.947 13.247
Test set i external validation 63 0.938 14.201
Validation set i external validation 123 0.961 10.710
Prediction set i testing 528 N/A N/A
FP_NN-GD: NN-GD model for flash point i

Neural network ensemble (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 445 0.955 12.133
Validation set i external validation 123 0.959 10.941
Test set i external validation 63 0.938 14.152
Prediction set i testing 528 N/A N/A
FP_consensus: Consensus model for flash point i

Ensemble model (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 445 0.964 10.902
Test set i external validation 63 0.945 13.308
Validation set i external validation 123 0.970 9.429
Prediction set i testing 528 N/A N/A

Property CN: Cetane number

Compounds: 330 | Models: 4 | Predictions: 20

CN_SVM-GD: SVM-GD model for cetane number i

Support vector machine (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 208 0.960 5.401
Test set i external validation 20 0.727 14.893
Validation set i external validation 56 0.806 12.458
Unreliable experimental values i external validation 46 0.711 17.536
Prediction set i testing 829 N/A N/A
CN_SVM-MD: SVM-MD model for cetane number i

Support vector machine (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 208 0.955 5.713
Test set i external validation 20 0.936 6.615
Validation set i external validation 56 0.900 8.615
Unreliable experimental values i external validation 46 0.727 16.916
Prediction set i testing 829 N/A N/A
CN_NN-MD: NN-MD model for cetane number i

Neural network ensemble (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 208 0.963 5.356
Test set i external validation 20 0.978 4.623
Validation set i external validation 56 0.846 11.367
Unreliable experimental values i external validation 46 0.668 18.751
Prediction set i testing 829 N/A N/A
CN_GRNN-MD: GRNN-MD model for cetane number i

Neural network (regression)

Open in:QDB Explorer QDB Predictor

Name Type n

R2

σ

Training set i training 208 0.888 9.247
Test set i external validation 20 0.579 16.931
Validation set i external validation 56 0.632 16.099
Unreliable experimental values i external validation 46 0.548 22.583
Prediction set i testing 829 N/A N/A

Citing

When using this data, please cite the original article and this QDB archive:

  • Saldana, D. A.; Starck, L.; Mougin, P.; Rousseau, B.; Pidol, L.; Jeuland, N.; Creton, B. Flash Point and Cetane Number Predictions for Fuel Compounds Using Quantitative Structure Property Relationship (QSPR) Methods. Energy Fuels 2011, 25, 3900–3908. http://dx.doi.org/10.1021/ef200795j

  • Piir, G. QDB archive #123. QsarDB repository, 2014. http://dx.doi.org/10.15152/QDB.123

Metadata

Show full item record

Title: Saldana, D. A.; Starck, L.; Mougin, P.; Rousseau, B.; Pidol, L.; Jeuland, N.; Creton, B. Flash Point and Cetane Number Predictions for Fuel Compounds Using Quantitative Structure Property Relationship (QSPR) Methods. Energy Fuels 2011, 25, 9, 3900–3908.
Abstract: In the present work, we report the development of models for the prediction of two fuel properties: flash points (FPs) and cetane numbers (CNs), using quantitative structure property relationship (QSPR) approaches. Compounds inside the scope of the QSPR models are those likely to be found in alternative jet and diesel fuels, i.e., hydrocarbons, alcohols, and esters. A database containing FPs and CNs for these types of molecules has been built using experimental data available in the literature. Various approaches have been used, ranging from those leading to linear models, such as genetic function approximation and partial least squares, to those leading to nonlinear models, such as feed-forward artificial neural networks, general regression neural networks, support vector machines, and graph machines. Except for the case of the graph machine method, for which the only inputs are the simplified molecular input line entry specification (SMILES) formulas, previously listed approaches working on molecular descriptors and functional group count descriptors were used to build specific models for FPs and CNs. For each property, the predictive models return slightly different responses for each molecular structure. Thus, final models labeled as “consensus models” were built by averaging the predicted values of selected individual models. Predicted results were compared with respect to experimental data and predictions of existing models in the literature. Models were used to predict FPs and CNs of molecules for which to the best of our knowledge there is no experimental data in the literature. Using information in the database, evolutions of properties when increasing the number of carbon atoms in families of compounds were studied.
URI: http://hdl.handle.net/10967/123
http://dx.doi.org/10.15152/QDB.123
Date: 2014-11-07


Files in this item

Name Description Format Size View
2011EF3900.zip QSPR models for the flash point and cetane number application/x-zip 1.976Mb View/Open
Files associated with this item are distributed
under Creative Commons license.

This item appears in the following Collection(s)

Show full item record