Hello all,

Happy, though rather belated, Thanksgiving. Hope everyone enjoyed a good meal and time with family.

I was traveling all last week and am just digging myself from underneath work that piled up during my absence. So I apologize for not getting a post up last week.

Recently I discovered a software tool for chemical kinetic modeling. It’s called the Chemical Kinetic Simulator (or CKS) from IBM Almaden research lab. The code has not been updated since mid-1990s, and I probably would not have given it much thought except for an interesting feature of this software.

The authors claim that CKS does not integrate sets of coupled differential equations to predict the time history of a chemical system. Instead the code uses a stochastic algorithm (type of Monte Carlo calculation) to propagate a reaction.  The authors claim that the stochastic method is much faster for stiff systems.

The  question is, if the stochastic method is much faster for solving reactive systems, why are most other software tools (e.g. Cosilab, Chemkin, DARS, FlameMaster, Chemical WorkBench) use solvers that integrate coupled systems of equations instead of using some form of a stochastic algorithm to solve chemically reactive systems?

I am determined to figure it out. In the next week I’ll be doing some research on the two solver methodologies , the CKS code and other related software. If you are dealing with anything that is chemically reacting,  stay tuned!

On the other hand, if you have some information on CKS OR stochastic vs. integrator methadologies that you would like to share with our readers, either comment below or contact me directly: masha@mvpmodeling solutions.com.

I will be sure to give due credit to everyone contributing information!

Best,
Masha

Tags: Chemical Kinetic Simulator, chemical kinetics, Chemical WorkBench, chemkin, computer modeling, computer simulation, Cosilab, IBM, Partial Differential Equations, stiff systems, stochastic method

This entry was posted on Wednesday, December 2nd, 2009 at 2:19 pm and is filed under Computer Modeling Best Practices. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.