Differentiable Thermodynamic Phase-Equilibria for Machine Learning

• URL: http://arxiv.org/abs/2603.11249v2
• Date: Mon, 16 Mar 2026 13:34:37 GMT
• Title: Differentiable Thermodynamic Phase-Equilibria for Machine Learning
• Abstract summary: DISCOMAX is a differentiable algorithm for phase-equilibrium calculation that guarantees thermodynamic consistency at both training and inference.<n>We evaluate the approach on binary liquid-liquid equilibrium data and demonstrate that it outperforms existing surrogate-based methods.
• Score: 39.691553958657764
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Accurate prediction of phase equilibria remains a central challenge in chemical engineering. Physics-consistent machine learning methods that incorporate thermodynamic structure into neural networks have recently shown strong performance for activity-coefficient modeling. However, extending such approaches to equilibrium data arising from an extremum principle, such as liquid-liquid equilibria, remains difficult. Here we present DISCOMAX, a differentiable algorithm for phase-equilibrium calculation that guarantees thermodynamic consistency at both training and inference, only subject to a user-specified discretization. The method is rooted in statistical thermodynamics, and works via a discrete enumeration with subsequent masked softmax aggregation of feasible states, and together with a straight-through gradient estimator to enable physics-consistent end-to-end learning of neural $g^{E}$-models. We evaluate the approach on binary liquid-liquid equilibrium data and demonstrate that it outperforms existing surrogate-based methods, while offering a general framework for learning from different kinds of equilibrium data.

Related papers

• A Comparative Investigation of Thermodynamic Structure-Informed Neural Networks [1.3965477771846408]
  Physics-informed neural networks (PINNs) offer a unified framework for solving both forward and inverse problems of differential equations.<n>We present a systematic comparison of different thermodynamic structure-informed neural networks by incorporating various thermodynamics formulations.
  arXiv  Detail & Related papers  (2026-03-26T07:39:15Z)
• Identifiable learning of dissipative dynamics [25.409059056398124]
  We introduce I-OnsagerNet, a neural framework that learns dissipative dynamics directly from trajectories.<n>I-OnsagerNet extends the Onsager principle to guarantee that the learned potential is obtained from the stationary density.<n>Our approach enables us to calculate the entropy production and to quantify irreversibility, offering a principled way to detect and quantify deviations from equilibrium.
  arXiv  Detail & Related papers  (2025-10-28T07:57:14Z)
• A machine-learned expression for the excess Gibbs energy [15.799043135621565]
  The excess Gibbs energy plays a central role in chemical engineering and chemistry, providing a basis for modeling the thermodynamic properties of liquid mixtures.<n>In this work, we address this challenge by integrating physical laws as hard constraints within a flexible neural network.<n>The resulting model, HANNA, was trained end-to-end on an extensive experimental dataset for binary mixtures from the Dortmund Data Bank.
  arXiv  Detail & Related papers  (2025-09-08T09:47:03Z)
• Efficient mapping of phase diagrams with conditional Boltzmann Generators [4.437335677401287]
  We develop deep generative machine learning models based on the Boltzmann Generator approach for entire phase diagrams. By training a single normalizing flow to transform the equilibrium distribution sampled at only one reference thermodynamic state to a wide range of target temperatures and pressures, we can efficiently generate equilibrium samples. We demonstrate our approach by predicting the solid-liquid coexistence line for a Lennard-Jones system in excellent agreement with state-of-the-art free energy methods.
  arXiv  Detail & Related papers  (2024-06-18T08:05:04Z)
• A Posteriori Evaluation of a Physics-Constrained Neural Ordinary Differential Equations Approach Coupled with CFD Solver for Modeling Stiff Chemical Kinetics [4.125745341349071]
  We extend the NeuralODE framework for stiff chemical kinetics by incorporating mass conservation constraints directly into the loss function during training. This ensures that the total mass and the elemental mass are conserved, a critical requirement for reliable downstream integration with CFD solvers.
  arXiv  Detail & Related papers  (2023-11-22T22:40:49Z)
• Discovering Interpretable Physical Models using Symbolic Regression and Discrete Exterior Calculus [55.2480439325792]
  We propose a framework that combines Symbolic Regression (SR) and Discrete Exterior Calculus (DEC) for the automated discovery of physical models. DEC provides building blocks for the discrete analogue of field theories, which are beyond the state-of-the-art applications of SR to physical problems. We prove the effectiveness of our methodology by re-discovering three models of Continuum Physics from synthetic experimental data.
  arXiv  Detail & Related papers  (2023-10-10T13:23:05Z)
• Accurate melting point prediction through autonomous physics-informed learning [52.217497897835344]
  We present an algorithm for computing melting points by autonomously learning from coexistence simulations in the NPT ensemble. We demonstrate how incorporating physical models of the solid-liquid coexistence evolution enhances the algorithm's accuracy and enables optimal decision-making.
  arXiv  Detail & Related papers  (2023-06-23T07:53:09Z)
• Machine learning in and out of equilibrium [58.88325379746631]
  Our study uses a Fokker-Planck approach, adapted from statistical physics, to explore these parallels. We focus in particular on the stationary state of the system in the long-time limit, which in conventional SGD is out of equilibrium. We propose a new variation of Langevin dynamics (SGLD) that harnesses without replacement minibatching.
  arXiv  Detail & Related papers  (2023-06-06T09:12:49Z)
• Gibbs-Duhem-Informed Neural Networks for Binary Activity Coefficient Prediction [45.84205238554709]
  We propose Gibbs-Duhem-informed neural networks for the prediction of binary activity coefficients at varying compositions. We include the Gibbs-Duhem equation explicitly in the loss function for training neural networks.
  arXiv  Detail & Related papers  (2023-05-31T07:36:45Z)
• Capturing dynamical correlations using implicit neural representations [85.66456606776552]
  We develop an artificial intelligence framework which combines a neural network trained to mimic simulated data from a model Hamiltonian with automatic differentiation to recover unknown parameters from experimental data. In doing so, we illustrate the ability to build and train a differentiable model only once, which then can be applied in real-time to multi-dimensional scattering data.
  arXiv  Detail & Related papers  (2023-04-08T07:55:36Z)
• Multiplicative noise and heavy tails in stochastic optimization [62.993432503309485]
  empirical optimization is central to modern machine learning, but its role in its success is still unclear. We show that it commonly arises in parameters of discrete multiplicative noise due to variance. A detailed analysis is conducted in which we describe on key factors, including recent step size, and data, all exhibit similar results on state-of-the-art neural network models.
  arXiv  Detail & Related papers  (2020-06-11T09:58:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.