An Artificial Neural Network model is proposed for the calculation and prediction of the surface tension of alcohols. A total amount of 4316 data for 147 alcohols was used for training, validating and testing the network model. After considering different architectures, the one giving better results includes an input layer that uses four independent variables (temperature, critical point temperature, critical density, and radius of gyration), two hidden layers with 21 neurons each one, and one neuron in the output layer was found to give the best results. Overall mean absolute percentage deviation of 1.04% was found, whereas models based on corresponding-states principle give mean deviations higher than 11.3%.
An Artificial Neural Network for the surface tension of alcohols
Pierantozzi M.;
2017-01-01
Abstract
An Artificial Neural Network model is proposed for the calculation and prediction of the surface tension of alcohols. A total amount of 4316 data for 147 alcohols was used for training, validating and testing the network model. After considering different architectures, the one giving better results includes an input layer that uses four independent variables (temperature, critical point temperature, critical density, and radius of gyration), two hidden layers with 21 neurons each one, and one neuron in the output layer was found to give the best results. Overall mean absolute percentage deviation of 1.04% was found, whereas models based on corresponding-states principle give mean deviations higher than 11.3%.File | Dimensione | Formato | |
---|---|---|---|
2017_ST-Alcohols.pdf
Solo gestori archivio
Tipologia:
PDF editoriale
Dimensione
2.09 MB
Formato
Adobe PDF
|
2.09 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.