Results from testing and verification

Still doing errortests, and (almost by accident) ran convergence tests for the combined solver, keeping dt, dx and the number of time steps constant, only varying the conversion rate. This has been done both in 1d and 2d, and put short the stochastic model completely dominates the error, making the convergence 0.5.

Fixed and added some stuff in 1d

Did some longer simulations to check the error. The results are illustrated in the figure below. Did some more tests as well, and implemented a Backward Euler solver in 1d which lets us do some more interresting tests.

Worked through most of the 1d case

Rewrote the thesis in pdf to a more structured form. Did quite thorough error testing (and found some bugs). By now equations on the from DtU = ∇D(x)∇U -∇vU +f(x,t) can be solved in 1dim and combined with random walk in some areas.
Investigated the effect of adding walkers as shown in this figure

Derived the equivalent diffusion equation for random walks with drift, still working on the actual conversion wrt velocity parameter. Some errortesting is done, but the results are quite bad as we see here.

Details in the pdf.

Updated run-script and error calculations


There is still some problems in the script with regards to the error. Seems to be calculating same error several times. Still, the interface is smoother and, more importantly, we can now run the program for several conversion-rates.

thesis in pdf

New experiment Fri Oct 11 11:40:04 2013.


Starting to investigate the effects of changing the number of walkers (and by extension the conversion factor), the walk-algorithm and whether or not one should reset walkers for each PDE-timestep (there should be a better relation here).
Seems that we can force the error to be of order dt by setting Hc = a/dt where a = 100. The parameter a is very important!

New experiment Tue Oct 8 09:38:07 2013.


Same experiment but with twice the number of walkers (M=100).

New experiment Mon Oct 7 08:45:52 2013.

Testing effect of doinusing last timestep as opposed to using current timestep as input to random walk

Started with some error-analysis oct. 3 2013

Did a simulation investigating the effects of doing a single step with the random walk model. Results were a bit bad; error of roughly 10dt.. Possible improvements in number of walkers, algorithm of walk, better conversion factor wrt step-length etc.

New experiment Thu Oct 3 14:18:32 2013.


Testing the effect of taking just one step with random walks.

New experiment Mon Sep 30 13:36:31 2013.

New experiment Mon Sep 30 11:56:53 2013.


Using average of PDE and Random walk solution, seems to work better than first attempt at least squares...

New experiment.

$ cd your_repo_root/repo_name
$ git fetch origin
$ git checkout gh-pages

If you're using the GitHub for Mac, simply sync your repository and you'll see the new branch.

About gh-pages

We've crafted some handsome templates for you to use. Go ahead and continue to layouts to browse through them. You can easily go back to edit your page before publishing. After publishing your page, you can revisit the page generator and switch to another theme. Your Page content will be preserved if it remained markdown format.

Rather Drive Stick?

If you prefer to not use the automatic generator, push a branch named gh-pages to your repository to create a page manually. In addition to supporting regular HTML content, GitHub Pages support Jekyll, a simple, blog aware static site generator written by our own Tom Preston-Werner. Jekyll makes it easy to create site-wide headers and footers without having to copy them across every page. It also offers intelligent blog support and other advanced templating features.

Authors and Contributors

You can @mention a GitHub username to generate a link to their profile. The resulting <a> element will link to the contributor's GitHub Profile. For example: In 2007, Chris Wanstrath (@defunkt), PJ Hyett (@pjhyett), and Tom Preston-Werner (@mojombo) founded GitHub.

Support or Contact

Having trouble with Pages? Check out the documentation at http://help.github.com/pages or contact support@github.com and we’ll help you sort it out.