@OlgaT ,
There are two methods that are commonly used to estimate 3D thermal bridging for elements that are discontinuous in the “z” axis of a 2D simulation:
The Parallel Planes Method (aka. Parallel Path)
You make two 2D THERM models: one that includes the thermal bridge and another that does not include it. You calculate the cross sectional area of the construction that is characterized by each THERM model. You then use this area to make an area-weighted average U-Value between the two THERM models. This method underestimates the impact of the thermal bridge.
The Isothermal Planes Method
For the thermal bridge element that is discontinuous in the Z-axis you calculate an area-weighted conductivity between the bridge and the insulation that is in place of it throughout the rest of the construction. You then build one THERM model and you assign a material with this area-weighted average conductivity to the thermal bridge element. This method overestimates the impact of the thermal bridge.
Either one of these methods can be pretty far from the true U-value but, if you average the results of these two simulations together, you typically get very close to the 3D thermal bridge U-value. As I mentioned previously, I would recommend mastering these methods before engaging with 3D thermal bridging software. This will at least give you a baseline that you can validate your first 3D thermal bridge simulations against.