Add new solver for strain and stress

[jbjacob94]

Create a new solver to compute strain and stress from ImageD11 ubis

[jonwright]

Thanks!!

I think we can merge this as it is and fix the python2.7 CI later (I guess it is just reformatting the f-strings).

Pinging @jadball and @AxelHenningsson as they have similar codes and might have good ideas for creating some testcases to verify the numbers and keep things working into the future

[AxelHenningsson]

Hi! It would be great with some Context? Is there any new maths here?

Cheers
Axel

[jonwright]

The conversion from strain to stress ought to match the old code in FitAllB. I think the invariants and eigen decompositions are new as python code here, but the maths should be known. A missing link (for me) is to have a verifiable check whether the numbers come out right for some low symmetry cases (e.g. quartz for trigonal vs hexagonal). There are some factors of 2 and sqrt(2) that are not obvious.

[AxelHenningsson]

I try to stick to the IEEE conventions for stiffness matrices. Is that what is used here? I think a link in the constructor to the convention used would be good.

As for testing I suppose you have to do a pen and paper calculation and then verify that the code agrees.

Best,
Axel

[jbjacob94]

Hi! nothing new with the math indeed, I just tried to remove any dependency with old python 2 code and I added a few functions to get the invariants, eigen decomposition and deviatoric component of strain / stress.

I try to stick to the IEEE conventions for stiffness matrices. Is that what is used here? I think a link in the constructor the the convention would be good.

I did not know there was a convention, but good to know. Do you have a link to send? I used voigt notation for the stiffness matrix, which seems to be commonly used, at least in the mineralogy community. best JB

…

________________________________ From: Axel Henningsson ***@***.***> Sent: 07 December 2023 14:39:35 To: FABLE-3DXRD/ImageD11 Cc: Jean-Baptiste Robert Jacob; Author Subject: Re: [FABLE-3DXRD/ImageD11] Add new solver for strain and stress (PR #192) I try to stick to the IEEE conventions for stiffness matrices. Is that what is used here? I think a link in the constructor the the convention would be good. As for testing I suppose you have to do a pen and paper calculation and then verify that the code agrees. Best, Axel — Reply to this email directly, view it on GitHub<#192 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BA2RBNROYDLFTH2QRCHZ6WTYIHBJPAVCNFSM6AAAAABAK3TIMWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNBVGM3DCMBRGE>. You are receiving this because you authored the thread.Message ID: ***@***.***>