Kinetic-Mamba: Mamba-Assisted Predictions of Stiff Chemical Kinetics

2025-12-16

AI Summary

To address low modeling accuracy and poor generalizability of stiff chemical kinetics in combustion simulation, this paper proposes Latent Kinetic-Mamba, a neural operator framework built upon the Mamba architecture. Methodologically, it integrates Mamba’s sequence modeling capability with neural operator mapping, physics-informed learning (e.g., mass conservation), multi-scale kinetic decomposition, and latent-space dimensionality reduction with manifold reconstruction. Its key contributions include three novel, complementary Mamba variants: (i) standalone temporal forecasting, (ii) mass-conserving constrained modeling, and (iii) temperature-partition-aware dual-modeling. Evaluated on syngas and GRI-Mech 3.0 mechanisms, the framework achieves high-fidelity, long-horizon thermochemical evolution prediction from initial conditions alone. It demonstrates superior recursive stability and extrapolative generalization compared to LSTM- and Transformer-based baselines.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Accurate chemical kinetics modeling is essential for combustion simulations, as it governs the evolution of complex reaction pathways and thermochemical states. In this work, we introduce Kinetic-Mamba, a Mamba-based neural operator framework that integrates the expressive power of neural operators with the efficient temporal modeling capabilities of Mamba architectures. The framework comprises three complementary models: (i) a standalone Mamba model that predicts the time evolution of thermochemical state variables from given initial conditions; (ii) a constrained Mamba model that enforces mass conservation while learning the state dynamics; and (iii) a regime-informed architecture employing two standalone Mamba models to capture dynamics across temperature-dependent regimes. We additionally develop a latent Kinetic-Mamba variant that evolves dynamics in a reduced latent space and reconstructs the full state on the physical manifold. We evaluate the accuracy and robustness of Kinetic-Mamba using both time-decomposition and recursive-prediction strategies. We further assess the extrapolation capabilities of the model on varied out-of-distribution datasets. Computational experiments on Syngas and GRI-Mech 3.0 reaction mechanisms demonstrate that our framework achieves high fidelity in predicting complex kinetic behavior using only the initial conditions of the state variables.

Problem

• Predicts stiff chemical kinetics evolution from initial conditions
• Enforces mass conservation in neural operator-based kinetic modeling
• Captures temperature-dependent regime dynamics in combustion simulations

Innovation

• Mamba-based neural operator for chemical kinetics modeling
• Three complementary Mamba models for state prediction and conservation
• Latent space evolution with reconstruction for reduced complexity