Abstract
In this communication, a new approach is presented which combines a group-contribution (GC) method approach with genetic function approximation (GFA) for the prediction of liquid heat capacities at constant pressure (C pL) for ionic liquids at atmospheric pressure. The proposed method can be used instead of complicated nonlinear modeling approaches like artificial neural networks and support vector machine. The NIST standard reference database was used to prepare a dataset for C pL data. The dataset comprised 82 ionic liquids and consisted of 3,726 experimental data points. The dataset was divided such that 80 % of the data were used as a training set, and 20 % as a validation and test set. GFA was used to select functional groups, from which the GC based model was developed. Statistical analysis of the model shows that it has an overall average absolute relative deviation of 1.68 %, coefficient of determination (R 2) of 0.990, and root mean square of error (RMSE) of 18.42 J mol−1 K−1.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call