Power calculation based on shades

[joey_tribbiani]

I have considerable doubts about calculating the power of the MPPT. In my opinion, the problem exists when there is a shadow on the PV installation.

In the pictures below, a situation in which, despite the total shade of the modules, the installation produces about 400 W. Such an anomaly occurs for most of the days I am analyzing. On some days it looks more correct (photo no. 3).

I often make comparisons with large shading (about 8%) between systems with and without optimizers and the conclusions from each such comparison are that it is not worth paying more for the system e.g. SE because actually according to your program the differences are small in the annual yield.

 Please explain why the program is so optimistic about uneven shading of several or even all PV modules.

[developer_mh]

Dear Joey,

could you send us the project file, please? It is then easier for us to analyse your question.

You can send it via private message (it is best if you save it without iv curves or results, so the file isn't too large)

Thanks a lot in advance, kind regards,

Martin

[joey_tribbiani]

Ok,  I sent you.

[Vishnu]

Interesting results. I personally ran simulations for some of our systems with and without optimizers to check if they are really useful. My experience is that, in our systems, it completely depends on the system itself. There is no one clear argument. 

In case of east-west or partial shaded systems, when modules in east and west are connected in series on the same string, there is a significant difference in yield in systems with and without optimizers. Systems with optimizers are better performing.

In cases, where there is negligible shading, say 2-4% on an average, and with inclination close to optimial value (for example: around 25-30 degrees in case of Germany), the system with optimizers has inferior performace compared to the one without optimizers. Reason is conversion losses in the optimizers.

I would love to see the explanation for your query where there is complete shading

[joey_tribbiani]

I'd like an explanation for that anomaly too. There's still no answer.

[developer_mh]

Hi Joey,

thank you for the project. And sorry for the late reply, we really have a lot to do at the moment.

In your post there are several questions at once, so I will try to answer them separately.

1) If you have total (direct) shadow on a module array, like in your first image, you will still have diffuse irradiation on the module surface, so you'll also see the PV system producing energy in these moments. In order to see the diffuse/anisotropic irradiation values in the results, activate them under Options -> Simulation -> Enhanced Simulation Results

2) If the modules of one MPP tracker receive full (global) irradiation, while the other modules are partially shaded, you will see differences in the power output of the two trackers. I guess this is what you wanted too illustrate with images 2 and 3. The difference however is dependent on the ratio between direct and diffuse irradation at given moment in time. The more direct irradiation you have, the higher the difference. Read more about radiation and how it is processed here: https://help.valentin-software.com/pvsol/2020/calculation/irradiation/

3) Power optimizers are another topic. As Vishnu pointed out, their advantage (in terms of "more energy output of the PV system") isn't always easy to identify. It really depends a lot on the shading, the string configuration and DC/DC losses of the optimizers etc. If a system receives full (direct) shadow, power optimizers can do nothing, you will even see less power output due to their DC/DC conversion losses. Same applies for situations without shadow (and, to be honest, for most other situations with partial shading as well). In the past, we did a study to identify situations where power optimizers can really lead to more power output, you can have a look here (in German, but with a lot of pictures: https://www.valentin-software.com/sites/default/files/leistungsoptimiererlangfassung.pdf)

The main point is: You really make use of the power optimizing technology if you have strings in parallel that receive uneven shadow.

Hope that helps to clarify the matter a bit. If you have further questions, please don't hesitate to ask.

Kind regards,

Martin

 

[Vishnu]

Adding to what Martin explained, from what i know, optimizers work on the principle of Festspannung (fixed voltage, not sure if that's the right word) so no matter what the module output voltage is (which depends on radiation it receives), the optimizers make sure that their output to the string is always at fixed voltage. This helps the MPPT and hence higher energy output. From this I would say, optimizers help in case of partial shading/multiple orientations/non uniform shading on modules in same string.

[developer_mh]

Hi Vishnu, hi Joey,

see also the link to our help pages on power optimizers here:

https://help.valentin-software.com/pvsol/2020/calculation/power-optimizer/

They have different operation modes (full, buck and sbustring buck) with their most common representatives SolarEdge (full), Tigo (substring) and Maxim (substring buck, module integrated).

Power optimizers in full mode, ie. SolarEdge devices, can ouput a variable voltage and a variable current, while their product V_out*I_out is always equal to the maximum power of the module P_MPP (minus the conversion losses). This is why we get this hyperbolic shape of the characteristics (orange curve in the images on the help pages). This is what they do, they take the optimal power of the module, and offer it to the external system, independent of the voltage or the current of the string.

So, in theory, the advantage of power optimizers is clear. You can connect any modules together, regardless of their orientation, shading and so on, and you will always receive the optimum power output.

In real systems however, where you normally connect PV modules of the same type, same orientation and so on, the situations where you gain energy through the use of power optimizers are more rare, and you have to compete against the constant conversion losses of the devices. The situation mentioned above (two or more strings in parallel, with uneven shadow) is where power optimizers perform best, but of course there are other configurations one could think of.

Kind regards,

Martin