Convergent Science

Design Optimization

CONVERGE™ CFD software is ideally suited for many different designs and applications. As all user meshing is eliminated, the user can very quickly set up numerous “what if” simulations. In some cases, this process can be automated to let the computer automatically search for optimized configurations.

CONVERGE™ comes equipped with the CONGO genetic algorithm optimization module. CONGO takes a “survival of the fittest” approach to optimize a design and automatically initiates CONVERGE™ CFD simulations in search of an optimum design. CONGO takes the manual user interaction out of the process.

Convergent_BestFir_Graph

 

With CONGO, the user defines a merit function (often a combination of NOx, soot and fuel consumption) along with any relevant design constraints. The merit function is related to the fitness of each design – this is what CONGO will optimize.

The variables CONGO is allowed to vary are specified by the user. Common optimization parameters include injection timing, injection pressure, injection rate-shape, EGR, the number, size and location of fuel injector nozzles, spark timing and valve lift profiles.

Finally, the user specifies the number of processors available and the total search time. CONGO then initiates a population of designs automatically searching for the one that will optimize the merit function.

CONGO comes standard with CONVERGE™ without any extra license fee. However, standard CONVERGE™ licenses are checked out for all CONVERGE™ simulations initiated by CONGO.

More information about CONGO, as applied to spray optimization, can be seen in an article in SAE magazine.

 

 

Selected Bibliography

“Multi-Mode Genetic Algorithm Optimization of Combustion Chamber Geometry for Low Emissions”, P. K. Senecal, E. Pomraning and K. J. Richards, SAE Technical Paper #2002-01-0958.

“A COMPUTATIONAL INVESTIGATION OF MODIFIED INTAKE PORTS TO IMPROVE SCAVENGING IN A HEAVY-DUTY TWO-CYCLE DIESEL ENGINE”, Proceedings of the ASME 2012 Internal Combustion Engine Division Fall Technical Conference ICEF2012, September 23-26, 2012, Vancouver, BC, Canada ICEF2012-92046, Jonathan Dolak and Deep Bandyopadhyay Electro-Motive Diesel, Inc. LaGrange, IL, USA.

Share This Page