.png){kind=icon}
.png){kind=icon}
Return
Optimizing progress variables for ammonia/hydrogen combustion using encoding-decoding networks
.png){kind=icon}
OA
AI
DOI：10.1016/j.combustflame.2025.114152.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We demonstrate a strategy to optimize parameterizations of combustion manifolds using an encoding-decoding artificial neural network architecture. Our focus in this work is on the combustion of ammonia (NH3) hydrogen (H2) blends. The literature on NH3 combustion, to date, lacks an efficient definition of a reaction progress variable (PV) to parameterize the thermo-chemical state-space. A quality parameterization should able to represent the thermo-chemical state variables accurately, as well as any functions of those, source terms of the non-conserved PVs. Our approach incorporates information about the reaction source of a PV and about important combustion products into the PV optimization. A gradient descent optimizer informed by the reconstruction quality of those important quantities of interest (QoIs) that enter optimization as decoder outputs. The approach can be thought of as an iterative back-and-forth between defining a parameterization (encoding) and reconstructing QoIs from it (decoding). It thus naturally promotes parameterizations where each QoI is uniquely and smoothly represented over the manifold. This work can advance the adaptivity of combustion models. First, we show that with an adequate definition of a can steer the model's accuracy towards improved representation of selected products and pollutants. the definition of a PV automatically adapts to best complement the remaining physics-based parameters, the mixture fraction or the enthalpy defect. These two achievements combined were not possible with existing PV optimization methods which only impose monotonicity and scalar gradient magnitude in defining PV. Novelty and Significance Statement We demonstrate a novel strategy to optimize the definition of a progress variable (PV) using an encoding-decoding artificial neural network. Our approach can be thought of as an iterative back-and-forth between defining a parameterization of a flame (encoding) and reconstructing important scalars from it (decoding). Notably, the PV definition and its corresponding source term are co-optimized. The definition of automatically adapts to best complement the remaining physics-based parameters, such as the mixture the enthalpy defect. These achievements were not possible with the existing PV optimization methods which only impose monotonicity and scalar gradient magnitude in defining a PV. This work can help advance combustion models, paving the way for adaptive reduced-order models, where the model can be adjusted towards particularly good representation of target scalars, such as pollutants. Our optimization method applicable to premixed and non-premixed combustion.
Keywords:
Progress variable
Ammonia
Parameterization
Flamelet-like model
Low-dimensional manifold
Artificial neural network
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.2
Papers:
9.5K
Citations:
4.2W
View more organizations