Prediction of sound velocity for selected ionic liquids using a multilayer feed-forward neural network

Ionic liquids (ILs) have a great potential as the new hydraulic fluids and the high-pressure heat transfer media. The speed of sound data represent the basis for determining practically important thermoelastic and thermodynamic properties. Since the speed of sound velocity measurements at elevated pressures are expensive and time consuming, their accurate predictions become particularly valuable. This study proposes a structurally straightforward neural network for speed of sound and adiabatic compressibility data in ILs. Its results are compared with the predictions of CP-PC-SAFT Equation of State. It is shown that both models are accurate, while each of them has advantages and disadvantages. Predictions of the adiabatic compressibility coefficients by CP-PC-SAFT and modelling framework coupling CP-PC-SAFT and neural network are also discussed.