Hi, I wanna modeling detonation (2-D) with Level set method. could you Introduce me papers about this subject? Do you have any information about this subject?

thanks

In fact, the detonation simulation requires the modeling of two fluids (i.e. explosive gas and surrounding media). The sole use of LSM may suffer from the problem in treating the interface separating two fluids. To overcome this difficulty, an interface method or multi-fluid method would be required. In my ph.d research, I employed a ghost fluid method coupled with the LSM. In that work, the LSM was only used to provide the interface information embeded in the fluid domain. Based on the interface information, a special technique to treat the discontinuity across the interface was required. As far as I know, the LSM cannot be solely used to simulate the detonation while it can play an important role in muti-fluid flow simulations. Please refer to the words, "ghost fluid method","immerged interface method","sharp interface method","multi-fluid method" and so forth. This statement is mainly based on my experience but it might be general. If you have an idea with the LSM only, I also need your help. The multi-fluid flow simulations are not trivial since the material discontinuity is presented near the interface separating two fluids.

http://library.epfl.ch/theses/?nr=3138

this thesis talks about level set method

For incompressible single fluid, the sole use of LSM is possible and straightforward. On the other hand, for compressible multi fluids, it is not.

Look at the papers written by Anne Bourlioux and Andrew Majda in the early 90s. I belive they used a shock fitting scheme to track the detonation front, rather than capturing it.

I know that Scott Stewart (University of Illinois - Urbana Champaign) and Rupert Klein (RWTH Achaen) have also been involved in this in the past.

As mentioned by Jinwon, I don't think level sets are the way to go though as simulating shocks requires strong conservation, which I don't believe is a solved problem (I may be wrong though).