Wide posterior range possibly because low frequency measurements are effectively ignored

[benjamin5988]

Hi -

I have been trying to run this on some data that has a capacitive tail - so the impedance at low frequencies gets very high.

The most important features occur in the 10^3-10^5 range, but the algorithm seems to fit only the 10^6-10^7.

In some commercial simplex algorithms, such as Gamry or Biologic, there is a Zmod weighting scheme as follows which gives less weight to the higher impedance data:

I was hoping to implement something like this or otherwise do a shortcut, like transform the data and the equivalent circuit function to logs.

I tried the logs, and I did not get good results. Wondering if you have any ideas or suggestions. I have many datasets and would not like to truncate them if there is another way.

Best, Ben

[ma-sadeghi]

Hi @benjamin5988,

Sorry for the late reply. I've been busy with other stuff. Indeed, what you've proposed sounds like a good idea. I'll try to implement the weighted optimization. Would you be willing to share your data so I can use it as a test case to evaluate the performance of with and without the weighted scheme? Thanks!