Skip to content

Conversation

@RolfStierle
Copy link
Contributor

The PC-SAFT entropy scaling correlation functions handle mixtures correctly: parameter "mixing" corrected.

@prehner
Copy link
Contributor

prehner commented Nov 28, 2025

Thanks! Is this a fix for something that went wrong from 0.8.0 to 0.9.0 (the conversion from ndarray to nalgebra in this case) or something that was just not used so far? I'm a bit lost.

@g-bauer
Copy link
Contributor

g-bauer commented Nov 28, 2025

Yeah this basically broke when rewriting it from ndarray to nalgebra:

let m = (x * &self.params.m).sum(); // <--- was: (*) does matmul, not component_mul in nalgebra
let m = x.dot(&self.params.m); // <--- should be the dot product

Changing pref accordingly to a (row) vector (component multiplication) then requires transposing when calculating the coefficients.

We currently only test pure systems and hence this error was not caught.

@RolfStierle RolfStierle marked this pull request as ready for review November 28, 2025 13:06
Rolf Stierle added 2 commits December 4, 2025 13:45
@RolfStierle RolfStierle force-pushed the main branch 2 times, most recently from a9ab786 to be1f67c Compare December 4, 2025 12:51
@g-bauer g-bauer merged commit baf5330 into feos-org:main Dec 4, 2025
16 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants