Interpretable Machine Learning-Based QSPR Modeling for Predicting Hydrogen Adsorption Performance of Ester Oil Impregnating Agents
- 1Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences , , ,
Abstract
Understanding the relationship between molecular structures and hydrogen adsorption performance is essential for the rational screening of ester oil impregnating agents for high-voltage capacitor applications. Herein, an interpretable machine learning-based quantitative structure–property relationship (QSPR) framework was developed to predict hydrogen adsorption performance of ester oils. A data set of 56 ester molecules was established through experimental characterization, and robust molecular descriptors were identified from a high-dimensional descriptor space using a bootstrap-assisted least absolute shrinkage and selection operator (LASSO) stability selection strategy. Multiple machine learning models were evaluated, with the optimized XGBoost model achieving excellent predictive performance (R2 = 0.940, RMSE = 10.316, and MAE = 8.707). Explainable analyses based on SHapley Additive exPlanations (SHAP) and partial dependence plots revealed key molecular features and nonlinear structure–property relationships governing hydrogen adsorption behavior. This study provides an interpretable data-driven strategy for screening ester-based functional liquids and offers molecular insights for optimizing capacitor impregnating materials.
Related
- Descriptor-guided thermodynamic screening of H2 adsorption on single-atom-doped anatase TiO2 nanoparticles with interpretable machine learning
- Predicting Glass Transition Temperature of Polyimide Homopolymers and Copolymers Using Interpretable QSPR-Based Machine Learning: The MFKit Framework
- Predicting Blood–Brain Barrier Permeability from Experimental Data: An Interpretable and Externally Validated Machine Learning Framework
- Interpretable Machine Learning for Predicting Cytochrome P450 Mechanism-Based Inhibition
- Double-network silica–polyacrylamide composite aerogels with enhanced mechanical integrity and multifunctional adsorption performance
BibTeX
@article{wang2026interpretable,
title = {{Interpretable Machine Learning-Based QSPR Modeling for Predicting Hydrogen Adsorption Performance of Ester Oil Impregnating Agents}},
author = {Xiaoyu Wang and Chao Ju and Huaigang Su and Qin Zhao and Haichao Liu and Qilong Zhao and Wenjing Lou},
journal = {ACS Omega},
year = {2026},
month = oct,
doi = {10.1021/acsomega.6c11253},
url = {https://pubs.acs.org/acsodf/article/doi/10.1021/acsomega.6c11253/5456910/Interpretable-Machine-Learning-Based-QSPR-Modeling},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - Interpretable Machine Learning-Based QSPR Modeling for Predicting Hydrogen Adsorption Performance of Ester Oil Impregnating Agents
AU - Xiaoyu Wang
AU - Chao Ju
AU - Huaigang Su
AU - Qin Zhao
AU - Haichao Liu
AU - Qilong Zhao
AU - Wenjing Lou
PY - 2026
DA - 2026/10/23
JO - ACS Omega
DO - 10.1021/acsomega.6c11253
UR - https://pubs.acs.org/acsodf/article/doi/10.1021/acsomega.6c11253/5456910/Interpretable-Machine-Learning-Based-QSPR-Modeling
AB - Understanding the relationship between molecular structures and hydrogen adsorption performance is essential for the rational screening of ester oil impregnating agents for high-voltage capacitor applications. Herein, an interpretable machine learning-based quantitative structure–property relationship (QSPR) framework was developed to predict hydrogen adsorption performance of ester oils. A data set of 56 ester molecules was established through experimental characterization, and robust molecular descriptors were identified from a high-dimensional descriptor space using a bootstrap-assisted least absolute shrinkage and selection operator (LASSO) stability selection strategy. Multiple machine learning models were evaluated, with the optimized XGBoost model achieving excellent predictive performance (R2 = 0.940, RMSE = 10.316, and MAE = 8.707). Explainable analyses based on SHapley Additive exPlanations (SHAP) and partial dependence plots revealed key molecular features and nonlinear structure–property relationships governing hydrogen adsorption behavior. This study provides an interpretable data-driven strategy for screening ester-based functional liquids and offers molecular insights for optimizing capacitor impregnating materials.
SN - 2470-1343
ER -