Gibbs-Duhem-Informed Neural Networks for Binary Activity Coefficient Prediction

Abstract

We propose Gibbs-Duhem-informed neural networks for the prediction of binary activity coefficients at varying compositions. That is, we include the Gibbs-Duhem equation explicitly in the loss function for training neural networks, which is straightforward in standard machine learning (ML) frameworks enabling automatic differentiation. In contrast to recent hybrid ML approaches, our approach does not rely on embedding a specific thermodynamic model inside the neural network and corresponding prediction limitations. Rather, Gibbs-Duhem consistency serves as regularization, with the flexibility of ML models being preserved. Our results show increased thermodynamic consistency and generalization capabilities for activity coefficient predictions by Gibbs-Duhem-informed graph neural networks and matrix completion methods. We also find that the model architecture, particularly the activation function, can have a strong influence on the prediction quality. The approach can be easily extended to account for other thermodynamic consistency conditions.

References in corpus (12)

• Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
• Thermodynamics-based Artificial Neural Networks for constitutive modeling
• Transfer learning for solvation free energies: from quantum chemistry to experiments
• Machine learning materials physics: Integrable deep neural networks enable scale bridging by learning free energy functions
• Machine Learning in Thermodynamics: Prediction of Activity Coefficients by Matrix Completion
• Graph Neural Networks for Temperature-Dependent Activity Coefficient Prediction of Solutes in Ionic Liquids
• A smile is all you need: Predicting limiting activity coefficients from SMILES with natural language processing
• SPT-NRTL: A physics-guided machine learning model to predict thermodynamically consistent activity coefficients
• Thermodynamics-informed graph neural networks
• Gibbs-Helmholtz Graph Neural Network: capturing the temperature dependency of activity coefficients at infinite dilution
• Machine Learning of consistent thermodynamic models using automatic differentiation
• Graph neural networks for the prediction of molecular structure-property relationships

Cited by in corpus (7)

• HANNA: Hard-constraint Neural Network for Consistent Activity Coefficient Prediction
• Thermodynamics-Consistent Graph Neural Networks
• Graph Neural Networks for Surfactant Multi-Property Prediction
• Predicting the Temperature Dependence of Surfactant CMCs Using Graph Neural Networks
• GRAPPA -- A Hybrid Graph Neural Network for Predicting Pure Component Vapor Pressures
• GraphXForm: Graph transformer for computer-aided molecular design
• Prediction of Activity Coefficients by Similarity-Based Imputation using Quantum-Chemical Descriptors