Poorandokht Ilani-Kashkouli

A chemical structure-based model for estimating speed of sound in liquids

Quantitative structure property relationship models for the speed of sound in liquids are developed based on molecular descriptors. A large dataset of 1,470 experimental data of speed of sound in 73 liquids is applied to derive the model. Twelve descriptors are selected by genetic function approximation to relate the speed of sound in liquids to their corresponding chemical structures. To capture the nonlinear nature of speed of sound in liquids, a model based on least-squared supported vector machine is also developed. The derived models are authenticated with several statistical validation techniques.