Failing to find PMPP in "Shadings 3D calculation" and "I/V curves of one MPPT input"

[Saphira]

Hi All

I've been using PVSyst since quite a while now, and it's great! But now I couldn't find a solution to my problem:

My System consist of one big Inverter, which has only one MPPT. Attached are 7 strings in parallel. Only one or two of them have a small dynamic shading. I did a couple of different simulations of different constellations (left out modules) to see, how it impacts the overall power generation.

I found the tools "Shadings 3D calculation" and "I/V curves of one MPPT input" and find them very powerful tools, since they show you exactly how much impact  the shading in single strings have on the whole array. 

First question: Are these results taken into account for the overall simulation? Or how is it taken into account that the shading is only affecting one or two strings and not all seven?

Second question: I saw, that if the difference is too small, the lower PMPP is not found. I have attached a sample, where it didn't work. Can this be made more sensitive? Since here, 4 percent (24 instead of 25 kW) are lost.

Maybe there is also another solution I'm not aware of. So I'm happy about every input 🙂

Best, Saphira

[André Mermoud]

First question: this shading  Module Layout calculations from the I/V curves is explicitly done  at each step of the simulation. The effect of the partial shadings on the whole array is fully explained by the I/V curves.  

Second question: in this case the  "Linear shading" (irradiance deficit) is 0.4%, and the electrical loss is null. The repartition between Linear and Electrical losses is not always quite reliable, as the linear shading loss is calculated globally for all tables of the concerned orientation, when this electrical calculation concerns one MPPT input. But the global loss is correctly calculated. 

 

[Saphira]

Dear Mr Mermoud

Thank you for your reply. I'm not sure if I understand your answer fully. Is the same algorithm used in the simulation? Because then I think that it doesn't work correctly.

This whole system is defined by only one MPPT input. 

I would argue, that the algorithm doesn't find the reduced MPP in the example above (it should be on the green reduced curve, but instead is on the blue ideal curve) and therefore states that the electrical loss is 0%. But in the same picture (power-voltage-curve) you can see that this can in fact not be, since the shaded green curve is lower and shows you a loss of about 4% (24kW instead of 25 kW). 

I can show you another example, where the difference is bigger, and therefore the algorithm finds the lower MPP (green point is on the green curve) and gives a correct estimate of the electrical loss:

 

And since 4% is not a small loss, I ask whether the same error is made in the simulation.

Best Regards

Saphira