cs.LGMar 31, 2026

Autonomous Adaptive Solver Selection for Chemistry Integration via Reinforcement Learning

Authors: Eloghosa Ikponmwoba, Opeoluwa Owoyele

Organizations: Department of Mechanical and Industrial Engineering Louisiana State University Baton Rouge, LA 70803

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

The computational cost of stiff chemical kinetics remains a dominant bottleneck in reacting-flow simulation, yet hybrid integration strategies are typically driven by hand-tuned heuristics or supervised predictors that make myopic decisions from instantaneous local state. We introduce a constrained reinforcement learning (RL) framework that autonomously selects between an implicit BDF integrator (CVODE) and a quasi-steady-state (QSS) solver during chemistry integration. Solver selection is cast as a Markov decision process. The agent learns trajectory-aware policies that account for how present solver choices influence downstream error accumulation, while minimizing computational cost under a user-prescribed accuracy tolerance enforced through a Lagrangian reward with online multiplier adaptation. Across sampled 0D homogeneous reactor conditions, the RL-adaptive policy achieves a mean speedup of approximately 3×3\times, with speedups ranging from 1.11×1.11\times to 10.58×10.58\times, while maintaining accurate ignition delays and species profiles for a 106-species n-dodecane mechanism and adding approximately 1%1\% inference overhead. Without retraining, the 0D-trained policy transfers to 1D counterflow diffusion flames over strain rates 1010--2000 s−12000~\mathrm{s}^{-1}, delivering consistent ≈2.2×\approx 2.2\times speedup relative to CVODE while preserving near-reference temperature accuracy and selecting CVODE at only 1212--15%15\% of space-time points. Overall, the results demonstrate the potential of the proposed reinforcement learning framework to learn problem-specific integration strategies while respecting accuracy constraints, thereby opening a pathway toward adaptive, self-optimizing workflows for multiphysics systems with spatially heterogeneous stiffness.

Figures & tables

Figure 1: Reinforcement learning (RL) framework for adaptive solver selection. The RL agent observes the evolving thermo-chemical state variables (e.g., TT , OH, H 2 , O, etc.) from the combustion simulation and selects the optimal numerical solver (e.g., CVODE or QSS) at each step.
ParameterValue
Rollout length NrolloutN_{\text{rollout}}2048
Minibatch size64
PPO epochs NepochN_{\text{epoch}}4
Learning rate3×10−43\times 10^{-4}
Discount γ\gamma0.99
GAE λGAE\lambda_{\text{GAE}}0.95
Table 1: Summary of key hyperparameters
Figure 2: RL-adaptive performance across initial conditions: (a) speedup distribution and (b) dependence on (T0,p)(T_{0},p) . Accuracy is maintained within 2.6%2.6\% ignition-delay error and < 110 K110~\mathrm{K} temperature RMSE.
Figure 3: 0D trajectory comparisons across thermochemical conditions using the RL-adaptive solver policy. (a) Condition 1 (650K, 1.0atm) achieves a 3.25 ×\times speedup with 0.41% ignition delay error and 5.4% CVODE usage. The temperature subplot shows solver selection via colored scatter (red=CVODE, green=QSS), with CVODE concentrated in a narrow ignition window ( t≈15t\approx 15 –22 ms). (b–d) As temperature and pressure increase (hence stiffness), CVODE usage rises from 14% (b) to 68.0% (c,d), demonstrating adaptive resource allocation. In all cases, the RL-adaptive approach maintains ignition delay errors below 2.62% and temperature RMSE below 108K.
Figure 4: Space-time evolution of temperature, HX2O\text{H}{\vphantom{\text{X}}}_{\smash[t]{\text{2}}}\text{O} , and OH illustrating agreement between the RL-adaptive solution and the CVODE reference, and deviations introduced by pure QSS.
Figure 5: Spatiotemporal solver deployment in the 1D counterflow diffusion flame at strain rate 2000 s−12000~\mathrm{s}^{-1} , as selected by the RL policy. Red denotes CVODE and turquoise denotes QSS.
More in this topic

Reinforcement Learning

Reinforcement learning studies how agents learn sequential decisions from rewards and interaction, balancing exploration, performance, safety, and adaptation.

See all 1,717 papers

Explore similar work

CardsList

  1. Entropy-Constrained Machine Learning with Residual Data Augmentation for Modeling Chemical Kinetics

    Jul 10, 2026Okezzi Ukorigho, Opeoluwa OwoyeleSurrogate ModelingPDE Surrogate Modeling

    Department of Mechanical and Industrial Engineering Louisiana State University, Baton Rouge, LA 70810, USA

  2. Adaptive hybrid coupling with operator inference, the overlapping Schwarz alternating method and reinforcement learning

    Sep 15, 2026Trishit Mondal, Irina Tezaur, Anthony GruberReinforcement LearningPDE Surrogate Modeling

    Worcester Polytechnic Institute

  3. Machine Learning-Driven Chemical Reactor Network Modeling of the Sandia-D Flame

    Jun 1, 2026Nicolas J. Tricard, Benjamin C. Koenig, Sili DengSurrogate ModelingComputational Fluid Dynamics

    Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA