shading losses - complex scene

[JessM]

Hi,

I have just run a complex terrain shading model through PVsyst (exported from Sketchup). It includes a terrain structure (shadow casting enabled on the ground object) with individual tracker rows defined as separate objects in XYZ space, floating over the underlying terrain layer. All tracker objects have a tilt of 1 degree towards the north (southern hemisphere).

As backtracking cannot be implemented between individual tracker objects, I have run the model without backtracking. The resulting shading loss of 8.4% is significantly lower than the 10.7% shading loss that I had calculated for an equivalent simplified shading model - a rectangular array with a tilt angle of 1 degree towards the north, same pitch of 4.4 m and no backtracking.

It doesn't seem to make sense that the near shadings loss could be so much lower for the complex scenario. The arrays are relatively constant in height (z coordinate) in the east-west direction so there should not be a significant benefit from having some trackers elevated relative to their neighbours. In this case I would also expect some additional shading due to the terrain and other objects.

The difference of 2.3% could possibly suggest that the diffuse component isn't being computed in the complex model - is this plausible? Is there some other inherent aspect of the complex shading model which could be resulting in under-prediction of shading losses?

Thanks for your help. Regards, Jess