Hi Bjorn.
I was just doing some research on this topic today for a colleague in Japan, when this thread popped up in my Google search. I figured I would pass along my findings on this topic for you to review further.
The way I understand the Critical Plane approach is that it is used specifically for multiaxial fatigue scenarios. Essentially, this approach is correlating principal stress directions to averaged value. it seems like you have a similar understanding. The links I have listed below have much deeper details on the academics behind this topic.
In the case of using Autodesk Simulation’s fatigue solution, both stress and strain based fatigue methods have stress correction options for multiaxial fatigue cases that are based on the critical plane approach. Namely, the Smith Watson Topper method is used for strain based calculations and the Goodman method is used for the stress based calculations.
Note: Everything I found is based on doing further research on critical plane and the actual correction methods I have listed above. I hope this helps.
Relevant links:
http://www.gruppofrattura.it/ocs/index.php/gigf/gigf2005/paper/viewFile/637/494
http://people.clarkson.edu/~yliu/paper/ffems-strain.pdf
https://www.efatigue.com/multiaxial/background/strainlife.html#Damage
https://www.efatigue.com/multiaxial/background/stresslife.html#Damage
http://www.sciencedirect.com/science/article/pii/S0142112304001756
-MIKE
Can't find what you're looking for? Ask the community or share your knowledge.