Artificial Neural Network and Principal Component Analysis Study of Excess Molar Volumes and Excess Molar Enthalpies in Ionic Liquid Mixtures
| Autor: | Fakhri Yousefi, Aboozar Kalantari |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Molar
Coefficient of determination Molar mass Enthalpy Analytical chemistry 02 engineering and technology 010402 general chemistry 021001 nanoscience & nanotechnology Mole fraction 01 natural sciences 0104 chemical sciences chemistry.chemical_compound Molar volume chemistry Principal component analysis Ionic liquid Physical and Theoretical Chemistry 0210 nano-technology |
| Zdroj: | Russian Journal of Physical Chemistry A. 93:809-821 |
| ISSN: | 1531-863X 0036-0244 |
| Popis: | This paper applies the model including back-propagation network (BPN) and principal component analysis (PCA) to estimate the excess molar volume and excess enthalpy of ionic liquid mixtures. The PCA was coupled with the BPN to optimize the BPN’s parameters and improve the accuracy of proposed model. The excess molar volume and excess enthalpy of ionic liquid mixtures are examined as a function of the temperature (T), mole fractions of compounds (x1 and x2), molar mass of pure ionic liquids (M1 and M2) and total molar mass (Mw) using artificial neural network. The obtained results by means of PCA–BPN model for excess molar volume and excess enthalpy have good agreement with the experimental data and absolute average deviations are 1.57 and 0.98%, respectively. Also, high coefficient of determination for excess molar volume and excess enthalpy are R2 = 0.9983 and 0.9999, respectively. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink