In order to accurately meet legislated fuel efficiency and emission standards, present day IC engines operate across complex combustion modes and use novel fuel formulations. Accurate simulation of these modes and fuel formulations requires the use of detailed chemical mechanisms, which typically span hundreds of species and chemical reactions. Even with advances in modern computing technology and algorithms, detailed chemistry simulation approaches are computationally time consuming and scale with the level of detail employed.
In a recent Engine Technology International (ETI) article and an ANSYS white paper, the importance of high-fidelity and detailed chemistry approaches, in IC engine modeling, were detailed.
Further, the articles discussed on ANSYS Forte’s ability to implement these approaches while mitigating the traditional, detailed chemistry vs. computational time trade-off.
Let’s look at some of the computational schemes and approaches, leveraged by Forte that allow it to resolve, hundreds of chemical species and reactions, faster than competitive solutions. Two computational techniques, specifically designed and implemented, to accelerate solution of the chemical kinetics in ANSYS Forte are, Dynamic Cell Clustering and Dynamic Adaptive Chemistry.
Dynamic Cell Clustering
Detailed chemistry solutions in combustion CFD, typically have a large computational overhead, due to repeated calculations of time intensive chemical calculations at each computational cell. ANSYS Forte uses Dynamic Cell Clustering (DCC) to dynamically group/cluster regions of the domain that have similar thermochemical conditions. This reduces the number of detailed chemistry calculations executed at every time step, as calculations are now executed for a group of cells viz. the cluster, and not for each and every cell.
The grouping of computational cells, in the calculation domain, into clusters is achieved by using clustering algorithms which identify cells that have similar thermochemical states. Cell temperature and equivalence ratio are used as the thermochemical clustering variables in the ANSYS Forte DCC algorithm. Though no manual intervention is typically necessary, the user has the ability to influence the cluster creation by defining the maximum temperature and equivalence ratio dispersion/deviation between cells, within each cluster.
The chemical kinetic equations are now solved at the cluster and not at the cell level, using averaged values for the state variables. The cluster averaged chemistry solution is them mapped back to the individual cells in each cluster.
Dynamic Adaptive Chemistry
To ensure validity over a wide range of thermochemical conditions, spanning the engine drive cycle, large chemical mechanisms, containing hundreds of species and reactions are typically used in an ANSYS Forte simulation. However, smaller subsets of the reaction mechanism might be adequate to resolve the chemistry, at certain locations and conditions. Dynamic Adaptive Chemistry (DAC) allows the user to perform on the fly reductions of the user supplied detailed chemical mechanism into smaller locally valid subsets. DAC therefore dynamically tailors the complexity of the chemical mechanism to appropriately reflect the local cell conditions, resulting is faster solution of the chemistry.
When the DAC is activated, the user is prompted to specify a species list that is tracked by the DAC algorithm. This list typically includes fuel species, species related to NOx formation such as NO and NO2, precursors to soot formation such as Acetylene (C2H2) and major products of combustion.
DAC and DCC provide ANSYS Forte users with avenues to accelerate detailed chemistry approaches, with very little manual intervention, thus allowing users to leverage these approaches, while operating within industrial design time frames.
To prove that chemistry really does matter, I invite you to download the Four Essential Facts about Chemistry and Combustion CFD white paper, and check out the recent Engine Technology International (ETI) article on the importance of high-fidelity and detailed chemistry approaches in IC engine modeling.
The post High-fidelity and Detailed Chemistry Approaches in IC Engine Modeling appeared first on ANSYS Blog.