CFD Flame Front Tracking with G-Equation and Well-Mixed Reactor Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computational fluid dynamics (CFD) methods for predicting combustion and emissions in spark-ignited engines face a tradeoff between accuracy and computational efficiency, with accurate models being resource-intensive and less accurate models relying on simplifying assumptions that reduce accuracy, particularly for soot precursors and other emissions-related species.
Innovation Solution
A CFD method that combines a computationally efficient G-equation combustion model with a sub-grid well-mixed-reactor model, applied cell-by-cell in computational grid cells representing the flame front, to track the flame front and calculate chemical species conversion and heat-release rates, considering local conditions and non-equilibrium effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If accurate combustion models are used in CFD codes, then prediction accuracy of chemical species and emissions is improved, but computational efficiency deteriorates due to high resource requirements
Solution Approach 1:
The computational domain is segmented into distinct regions: pre-flame region, flame front region, and post-flame region. Each region is treated with appropriate modeling approaches - the flame front uses a thin-flame model with G-equation, while pre- and post-flame regions use different chemical kinetics treatments. This segmentation allows accurate resolution of the critical flame front without applying computationally expensive detailed chemistry throughout the entire domain.
Solution Approach 2:
Different levels of chemical detail are applied to different spatial locations. In the flame front region where combustion occurs, a reduced chemical mechanism is used to capture essential chemistry. In the pre-flame and post-flame regions, simpler models are applied. This local differentiation maintains accuracy where needed while reducing computational burden in regions where detailed chemistry is less critical.
2Productivity
If less accurate combustion models are used in CFD codes, then computational efficiency is improved, but prediction accuracy of chemical species and emissions deteriorates due to simplifying assumptions
Solution Approach 1:
The flame front is extracted and treated as a distinct thin structure separate from the bulk flow. By identifying and isolating the flame front region, the model applies specialized thin-flame physics and reduced chemistry only where combustion occurs, rather than using simplified equilibrium assumptions throughout the entire domain. This extraction allows accurate capture of combustion chemistry in the flame front while maintaining computational efficiency.
Solution Approach 2:
The model transitions from detailed chemical mechanisms to reduced mechanisms by changing the level of chemical detail based on spatial location and combustion state. In the flame front, reduced mechanisms with key reaction pathways are used. In non-combusting regions, simpler models apply. This parameter change in chemical mechanism complexity maintains accuracy for emissions prediction while improving computational efficiency.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for accurate prediction of chemical species and emissions while maintaining computational efficiency, effectively capturing the chemistry of combustion in spark-ignited engines without the limitations of equilibrium assumptions.
Implementation Method 1
identify positions in the computational grid of a flame front propagating through the combustible fluid mixture
Implementation Method 2
apply a sub-grid well-mixed-reactor model to the flame-front volume inside every computational cell within the set of representative computational cells to compute chemical results from combustion of the combustible fluid mixture
Data Source
AI summary
Embodiments are disclosed of a computer-implemented method. The method includes establishing a computational grid that includes a plurality of computational cells and describes a volume containing a combustible fluid mixture. The method identifies positions in the computational grid of a flame front propagating through the combustible fluid mixture, identifies a set of representative computational cells that can be used as a computational representation of the flame front, and applies a well-mixed-reactor model and a G-equation model to every computational cell within the set of representative computational cells to compute chemical results from combustion of the combustible fluid mixture in the flame front.