Post by gnfgoncalves on Jun 15, 2018 11:39:57 GMT -6
I was just wondering if there any plans to implement the inversion algorithm proposed by Pigou et al. (2018) in OpenQBMM? They claim reductions in computational cost from 60% to 95%. Do you think it is worth it to test it on OpenQBMM? I might try to implement it myself, if I have the time for that...
Maxime Pigou, Jérôme Morchain, Pascal Fede, Marie-Isabelle Penet, Geoffrey Laronze, New developments of the Extended Quadrature Method of Moments to solve Population Balance Equations, Journal of Computational Physics, Volume 365, 2018, Pages 243-268, ISSN 0021-9991, doi.org/10.1016/j.jcp.2018.03.027. (http://www.sciencedirect.com/science/article/pii/S0021999118301876)
Post by Alberto Passalacqua on Jun 16, 2018 3:30:40 GMT -6
We have been considering the method proposed by Maxime Pigou, and I would like to implement it too. The only constraint so far has been time to perform the implementation, but I will be happy to assist if you decide to work on it. If so, I'd suggest to move this discussion to our GitHub issue tracker, so that it can be tracked more easily.
Post by gnfgoncalves on Jun 21, 2018 7:46:06 GMT -6
I'm not sure how much time I'll have to dedicate to that, but since it's not something that would block somebody else's work, I think I can try to start implementing it. If it starts going too slow, or priorities change for some reason, we can rearrange it... I'll open a issue on GitHub. -Gabriel