Prediction of <scp>CO<sub>2</sub></scp> solubility in ionic liquids via convolutional autoencoder based on molecular structure encoding
Tianxiong Liu, Dingchao Fan, Yusen Chen, Yasen Dai, Yuyang Jiao, Peizhe Cui, Yinglong Wang, Zhaoyou Zhu
Abstract
Abstract In this study, novel molecular structure encoding descriptors composed of feature encoding and one‐hot encoding was developed and then convolutional autoencoder was used to denoise based on the structure of ionic liquids (ILs). It could be used to predict the CO 2 solubility in ILs at different temperatures and pressures, when combined with three different machine learning algorithms (multilayer perceptron [MLP], random forest [RF], and support vector machine [SVM]). Statistics of the prediction results show that the newly proposed molecular structure‐based coding has better regression prediction performance than the conventional molecular cheminformatics descriptors. SE‐MLP model with R 2 of 0.9873 and mean square error of 0.0007 has the best performance in predicting the CO 2 solubility in ILs. In addition, the relationship between features and dissolved CO 2 capacity was analyzed through model interpretation to retrieve physical insights for the underlying system. This work provided a new predictive tool for enriching and refining data on CO 2 solubility in ILs and for solving phase equilibrium problems.