Machine Learning of consistent thermodynamic models using automatic differentiation

Abstract

We propose a data-driven method to describe consistent equations of state (EOS) for arbitrary systems. Complex EOS are traditionally obtained by fitting suitable analytical expressions to thermophysical data. A key aspect of EOS are that the relationships between state variables are given by derivatives of the system free energy. In this work, we model the free energy with an artificial neural network, and utilize automatic differentiation to directly learn the derivatives of the free energy. We demonstrate this approach on two different systems, the analytic van der Waals EOS, and published data for the Lennard-Jones fluid, and show that it is advantageous over direct learning of thermodynamic properties (i.e. not as derivatives of the free energy, but as independent properties), in terms of both accuracy and the exact preservation of the Maxwell relations. Furthermore, the method implicitly provides the free energy of a system without explicit integration.

References in corpus (2)

• Estimation of Thermodynamic Observables in Lattice Field Theories with Deep Generative Models
• BoltzmaNN: Predicting effective pair potentials and equations of state using neural networks

Cited by in corpus (5)

• Gibbs-Duhem-Informed Neural Networks for Binary Activity Coefficient Prediction
• Thermodynamics-Consistent Graph Neural Networks
• Accelerated calculation of configurational free energy using a combination of reverse Monte Carlo and neural network models: Adsorption isotherm for 2D square and triangular lattices
• Neural Network Representations of Multiphase Equations of State
• Knowledge Distillation of Noisy Force Labels for Improved Coarse-Grained Force Fields