ChemKANs for Combustion Chemistry Modeling and Acceleration

Abstract

Efficient chemical kinetic model inference and application in combustion are challenging due to large ODE systems and widely separated time scales. Machine learning techniques have been proposed to streamline these models, though strong nonlinearity and numerical stiffness combined with noisy data sources make their application challenging. Here, we introduce ChemKANs, a novel neural network framework with applications both in model inference and simulation acceleration for combustion chemistry. ChemKAN's novel structure augments the generic Kolmogorov Arnold Network Ordinary Differential Equations (KAN-ODEs) with knowledge of the information flow through the relevant kinetic and thermodynamic laws. This chemistry-specific structure combined with the expressivity and rapid neural scaling of the underlying KAN-ODE algorithm instills in ChemKANs a strong inductive bias, streamlined training, and higher accuracy predictions compared to standard benchmarks, while facilitating parameter sparsity through shared information across all inputs and outputs. In a model inference investigation, we benchmark the robustness of ChemKANs to sparse data containing up to 15% added noise, and superfluously large network parameterizations. We find that ChemKANs exhibit no overfitting or model degradation in any of these training cases, demonstrating significant resilience to common deep learning failure modes. Next, we find that a remarkably parameter-lean ChemKAN (344 parameters) can accurately represent hydrogen combustion chemistry, providing a 2x acceleration over the detailed chemistry in a solver that is generalizable to larger-scale turbulent flow simulations. These demonstrations indicate the potential for ChemKANs as robust, expressive, and efficient tools for model inference and simulation acceleration for combustion physics and chemical kinetics.

B.C.K. and S.K. contributed equally to this work. 21 pages, 8 figures, and 1 table. Updated after acceptance to journal

References in corpus (22)

• DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators
• KAN: Kolmogorov-Arnold Networks
• Autonomous Discovery of Unknown Reaction Pathways from Data by Chemical Reaction Neural Network
• Stiff Neural Ordinary Differential Equations
• KAN-ODEs: Kolmogorov-Arnold Network Ordinary Differential Equations for Learning Dynamical Systems and Hidden Physics
• KAN 2.0: Kolmogorov-Arnold Networks Meet Science
• KANQAS: Kolmogorov-Arnold Network for Quantum Architecture Search
• Kolmogorov-Arnold Networks are Radial Basis Function Networks
• Finite basis Kolmogorov-Arnold networks: domain decomposition for data-driven and physics-informed problems
• ReLU-KAN: New Kolmogorov-Arnold Networks that Only Need Matrix Addition, Dot Multiplication, and ReLU
• LeanKAN: A Parameter-Lean Kolmogorov-Arnold Network Layer with Improved Memory Efficiency and Convergence Behavior
• FC-KAN: Function Combinations in Kolmogorov-Arnold Networks
• Physics Informed Kolmogorov-Arnold Neural Networks for Dynamical Analysis via Efficent-KAN and WAV-KAN
• Physics-informed Kolmogorov-Arnold Network with Chebyshev Polynomials for Fluid Mechanics
• Arrhenius.jl: A Differentiable Combustion SimulationPackage
• HiPPO-KAN: Efficient KAN Model for Time Series Analysis
• AF-KAN: Activation Function-Based Kolmogorov-Arnold Networks for Efficient Representation Learning
• PowerMLP: An Efficient Version of KAN
• Reduced Effectiveness of Kolmogorov-Arnold Networks on Functions with Noise
• Kolmogorov-Arnold Networks and Evolutionary Game Theory for More Personalized Cancer Treatment
• Architectural Scaling Surpass Basis Complexity? Efficient KANs with Single-Parameter Design
• Want to train KANS at scale? Now UKAN!

Cited by in corpus (1)

• Learning continuous state of charge dependent thermal decomposition kinetics for Li-ion cathodes using Kolmogorov-Arnold Chemical Reaction Neural Networks (KA-CRNNs)