Global Sources
EE Times-India
EE Times-India > EDA/IP

Achieve accurate clean engine design

Posted: 09 Nov 2012     Print Version  Bookmark and Share

Keywords:chemical processes  Computational Fluid Dynamics  CAD 

Larger, more accurate fuel mechanisms can be incorporated into the simulation to achieve runtimes similar to those with much smaller, less accurate mechanisms. The FORTÉ CFD package employs in�novative techniques to achieve these unprecedented calculation speeds: Dynamic Cell Clustering leverages an advanced algorithm to group those cells that have similar kinetic conditions at each time-step and eliminates duplicate calculations. Dynamic Adaptive Chemistry automatically reduces the kinetics on the fly at every time-step, to use only the chemistry required at that time. This method leads to huge savings in solution time with no loss in accuracy. When combined with RD's proprietary chemistry-solver tech�nology, these techniques generate highly accurate results in solution times that are several orders of magnitude faster than chemistry solvers used with other commercial CFD packages.

Reduce grid and time-step depen�dency
The choice of spray models can have a significant impact on both time-to-solution and the accuracy of results. Many spray models are highly mesh dependent, requiring that valuable time be spent refining the mesh, or adding mesh complexity, to find an acceptable combination of model parameters and grid mesh. Even when a spray model is calibrated to a particular grid, it is unclear how accurate the model will be when attempting to predict the behaviour of a different engine design. FORTÉ's multi-component spray model maintains consistency between the physical properties of the engine and the chemical model of the fuel to more accurately capture droplet evaporation and ignition. This allows for a more accurate spray model without the need to drastically increase mesh refinement. Less calibration means better portability to other designs.

The FORTÉ CFD Package includes an automated mesh generator that facilitates rapid analysis of variations in geometry, such as bowl shape, bore size, valve angles and more. Without advanced chem�istry, automatic mesh generation is just a faster path to the wrong answer. The FORTÉ Automesher generates meshes on the fly without introducing numerical errors or increasing run times. Mesh movement is handled smoothly and robustly, while ac�counting accurately for piston and valve motion. Meshes can also be imported from common third-party tools such as ICEM-CFD or GAMBIT. KIVA-3V meshes are also easily imported.

The FORTÉ CFD Package includes a comprehensive visualiser that quickly generates graphical representations of simulation results in a form engine designers need. Visualisations are intuitive and interactive, with support for cut-planes, line probes, phi-T map generation and external data import, eliminating the need for 3rd-party post-processing tools.

No sepa�rate tools are needed to gather, collate and view results. This visualiser provides quick creation of animations for any 3-D view, 3-D contour plots with easy control of both the cut-planes and the viewing angles; import of experimental data, automatic unit conversions, and quick comparisons of parameter-study results; automated creation of Phi-T maps to show where and when emissions are produced, including animations and last but not least export to third-party tools, including Field view, Insight, Tecplot and Excel.

Eliminate weeks of effort
Automatically generating a mesh at runtime enables grid refinement and geometric parameter studies and improves time-to-solution by removing the meshing "bottleneck" in the design flow. With virtually no intervention from the user, FORTÉ reads CAD files directly and imports the detailed geometry information it needs to automatically generate a high-quality, well-structured, Cartesian mesh. The advanced automatic mesh-generation technology in FORTÉ includes full valve motion and 3-D mesh capability.

For a Cartesian mesh, the Local Truncation Error (LTE) is minimised, because the structure of the Cartesian mesh is regular. Of course, some non-uniformity in the mesh is introduced at solid-fluid interfaces. Typically, the mesh is refined at the boundary, such that the regular and uniform-sized cells are split into a number of smaller control volumes. In this way there is irregularity in the form of "hanging nodes" that must be dealt with in cells adjacent to a change in refinement level (or density) in the mesh. The proportion of these cells reduces as a smaller base mesh is used. The cells in a refined layer, however, are also regular or cubic in shape, unless a cut-cell approach is used. Overall, irregularity is minimised the closer the mesh is to being fully Cartesian

The automatic mesh generation technique utilised in FORTÉ is based on a unique immersed-boundary theory. It allows the use of a Cartesian mesh that is faster to create, more robust and less prone to error than other approaches, such as body-fitted unstructured or pure cut-cell meshes. The automatic mesh generation works from the inside-out, requiring only that you specify one point within the domain that is always "inside". In addition, just a few user-defined mesh-size controls are required to make sure that the mesh refinement around exceptional features (valves, etc.) is accomplished adequately.

The automatic mesh generation begins by importing the engine geometry, typically from a CAD program ('programme' for plan) in the form of an STL file. Then any splitting or merging of surfaces needed to separate specific engine components, e.g., head, valves, and piston, is performed within the FORTÉ Simulation Interface. The mesh generated during the simulation is based on a global maximum cell size specified in the setup. Optionally the mesh may be refined around important features, such as valve seats, or dynamically at critical times during the simulation, such as near firing top-dead centre (TDC). The mesh is automatically created at every time step during the simulation.

Conventional CFD gridding meshing techniques require that valuable time be spent refining the mesh, or adding mesh complexity, to find an acceptable combination of spray model parameters and grid mesh. From an accuracy point of view, however, the ideal mesh is one that is Cartesian, with perfectly orthogonal faces, and one in which the boundary conditions can be applied exactly on the physical surfaces of the real geometry. Both unstructured-mesh and cut-cell approaches require a very large degree of mesh refinement at the boundaries to approximate the real boundary surfaces for useful geometries (such as a cylinder). This translates to longer compute times, especially when chemistry calculations are needed in each cell.

 First Page Previous Page 1 • 2 • 3 Next Page Last Page

Comment on "Achieve accurate clean engine design"
*  You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.


Go to top             Connect on Facebook      Follow us on Twitter      Follow us on Orkut

Back to Top