arXiv subjects

S. Mani Sarathy

Publications and source records attributed to S. Mani Sarathy.

3 recordsLinked to original sources

arXiv2026

Thermodynamics-Informed Learning Captures Hierarchical Hydrogen-Bond Networks in Deep Eutectic Solvents

Deep eutectic solvents (DESs) offer a tunable platform for lignocellulose refining, but their solvation behavior depends on multicomponent hydrogen-bond networks rather than individual constituent properties. We develop TiNN, a thermodynamics-informed neural network that hierarchically represents HBA-HBD association and solute accommodation through a learned free-energy pathway. Trained on 938 records, TiNN achieves a test-set $R^2$ of 0.917 and generalizes to new HBA-HBD combinations and sparse conditions. Ablation identifies solute-specific stabilization as the largest contributor to prediction accuracy, while the importance of hydrogen-bond compatibility depends on the dissolution mechanism. Experimental measurements of glucose, xylan, and Kraft lignin in acidic, basic, and neutral DESs show that TiNN reduces the geometric mean solubility error from more than 23-fold with COSMO-RS to approximately 1.5-fold. Without training on fractionation data, TiNN also distinguishes DES formulations with high and low lignin and hemicellulose removal from corn stover. These results support chemically grounded solvent screening for lignocellulose fractionation.

cs.LG↗

arXiv2021

A Reinforcement Learning-based Economic Model Predictive Control Framework for Autonomous Operation of Chemical Reactors

Economic model predictive control (EMPC) is a promising methodology for optimal operation of dynamical processes that has been shown to improve process economics considerably. However, EMPC performance relies heavily on the accuracy of the process model used. As an alternative to model-based control strategies, reinforcement learning (RL) has been investigated as a model-free control methodology, but issues regarding its safety and stability remain an open research challenge. This work presents a novel framework for integrating EMPC and RL for online model parameter estimation of a class of nonlinear systems. In this framework, EMPC optimally operates the closed loop system while maintaining closed loop stability and recursive feasibility. At the same time, to optimize the process, the RL agent continuously compares the measured state of the process with the model's predictions (nominal states), and modifies model parameters accordingly. The major advantage of this framework is its simplicity; state-of-the-art RL algorithms and EMPC schemes can be employed with minimal modifications. The performance of the proposed framework is illustrated on a network of reactions with challenging dynamics and practical significance. This framework allows control, optimization, and model correction to be performed online and continuously, making autonomous reactor operation more attainable.

eess.SY↗

arXiv2018

Dynamic Chemical Model for H2/O2 Combustion Developed Through a Community Workflow

Elementary-reaction models for H2/O2 combustion were evaluated and optimized through a collaborative workflow, establishing accuracy and characterizing uncertainties. Quantitative findings were the optimized model, the importance of $\text{H}_2 + \text{O}_2(1Δ) = \text{H} + \text{HO}_2$ in high-pressure flames, and the inconsistency of certain low-temperature shock-tube data. The workflow described here is proposed to be even more important because the approach and publicly available cyberinfrastructure allows future community development of evolving improvements. The workflow steps applied here were to develop an initial reaction set using Burke et al. [2012], Burke et al. [2013], Sellevag et al. [2009], and Konnov [2015]; test it for thermodynamic and kinetics consistency and plausibility against other sets in the literature; assign estimated uncertainties where not stated in the sources; select key data targets ("Quantities of Interest" or QOIs) from shock-tube and flame experimental data; perform conventional sensitivity analyses of QOIs with respect to Arrhenius pre-exponential factors; develop surrogate models for the model-predicted QOI values; evaluate model-vs.-data consistency using Bound-to-Bound Data Collaboration; and optimize model parameters within their estimated uncertainty bounds (feasible set). Necessary data and software for such analyses were developed and are publicly available through the PrIMe cyberinfrastructure. This community workflow proved to be a means to reveal inconsistencies, improvements, and uncertainty bounds. Even more significantly, it is a means of revealing which parameters and experimental findings are inconsistent with the larger body of work from the community and, thus, of designing new experiments and parameter calculations.

physics.chem-ph↗