Yeah, thanks for the insight. I'm aware of the various atmospheric models. I want to find something that has been a) proven useful for solar power predictions; b) has a working open source python implementation. While the scientific background is very interesting in itself, I don't intend to re-invent the wheel. I do have a fairly good solar and UV sensor for my VP2 weather station. An interesting fact, this spring the solar sensor W/m2 curve almost completely overlaps with the built-in solar model prediction (theoretical max. irradiation) in WeeWx, and we do have unobstructed clear sky and sunshine these days. So I'm hoping to do something similar with my solar power system.
I've found 2 worthy contenders in the subject so far: https://github.com/pvlib/pvlib-python/ https://pypi.org/project/solprimer/ I need to indulge myself in them a bit to see which might fit my needs the best. I think the first one might be better suited and it's also actively developed, but I just don't know yet. On Sunday, April 21, 2019 at 4:56:51 PM UTC+2, Leon Shaner wrote: > > Kobuki, > > My other half (who is an avid physics/astronomy type) points out that > there are two non-obvious factors here, one being "atmospheric extinction" > and the other being "transparency." > > Atmospheric extinction has to do with the fact that the sun's radiation > has to go through more atmosphere at lower inclinations than when it is > directly overhead. A solar radiation detector will already be subject to > atmospheric extinction, in that it will simply read lower when the sun is > closer to the horizon, even on a completely clear day. As an aside, I'd > have to wonder how accurate solar radiation sensors are that don't have > mechanical tracking to keep directly pointed at the sun. :-/ > Seems like that much of this discussion should be "on topic" for this > group. =D > > Transparency, on the other hand is something that acts more like a filter, > and as transparency varies, it has more pronounced effects on certain > wavelengths vs. others. Because solar panels do not react to the entire > visible spectrum, transparency can have an even more dramatic effect on a > panel's efficiency, than just how bright/clear it seems to the human eye. > :-/ > > From that, my takeaway is you probably need to factor in a UV sensor to > account for transparency. > > That forum I quoted would probably be a great place to get more input. > Plenty of PhD types over there. =D > > Good luck! =D > > Regards, > Leon > -- > Leon Shaner :: Dearborn, Michigan (iPad Pro) > > On Apr 21, 2019, at 10:37 AM, Leon Shaner <[email protected] <javascript:>> > wrote: > > Kobuki, > > Ok. Do you already have a solar radiation detector? If you know the solar > radiation, and your fixed geo-location, the position of the sun can be > calculated and the math I mentioned can be used to calculate the efficiency > of your panel across the range of the sun's motion. > Seems like you should be looking for Astronomy-oriented projects which can > track the sun's position from your geo-location. Then the rest of the math > is just the angle of incidence calculations that I mentioned. Sorry I > don't know of any projects that tie this together. > > Here's an interesting lead: > > > https://www.researchgate.net/post/How_can_we_compute_solar_position_at_a_given_place_on_a_given_day_and_time > > Regards, > Leon > -- > Leon Shaner :: Dearborn, Michigan (iPad Pro) > On Apr 21, 2019, at 10:00 AM, kobuki <[email protected] <javascript:>> > wrote: > > Thanks for the quick heads up. To make it clear, I'm not looking for the > math theory behind it, I'm trying to find an existing Python library or > source code that I can integrate or use for writing the service. I want to > track the energy production, we have a fixed setup. I'm not interested in > sun tracking calculation now. > > On Sunday, April 21, 2019 at 3:56:52 PM UTC+2, Leon Shaner wrote: >> >> You can google for "solar panel angle of incidence" and you'll find some >> good articles which explain the math. >> Suffice it to say, it works out that tracking the sun yields about 30% >> more power than the same panel in a fixed position, assuming consistent sun >> throughout the day. >> HTH >> >> Regards, >> Leon >> -- >> Leon Shaner :: Dearborn, Michigan (iPhone) >> >> On Apr 21, 2019, at 9:47 AM, kobuki <[email protected]> wrote: >> >> Sorry if it might seem a bit off topic, but since I want to integrate >> this service in WeeWx, too, I thought it's the appropriate place after all. >> I'd like to add a solar power tracker that gives me the theoretical maximum >> output of irradiation on a surface in W/m2 that has an arbitrary tilt and >> rotation wrt south/north (depending on the hemisphere). I'm sure this has a >> Python solution already, I'm just not familiar enough with the subject to >> find it. Ultimately I want to estimate the theoretical max. output of our >> solar power system on the roof. For starters, I'd be content with >> calculation methods that ignore the inherent losses in the system (cables, >> inverter, panel temperature coefficients, etc). >> >> Does anyone have pointers to start with? >> >> -- >> You received this message because you are subscribed to the Google Groups >> "weewx-user" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> For more options, visit https://groups.google.com/d/optout. >> >> -- > You received this message because you are subscribed to the Google Groups > "weewx-user" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <javascript:>. > For more options, visit https://groups.google.com/d/optout. > > -- > You received this message because you are subscribed to the Google Groups > "weewx-user" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <javascript:>. > For more options, visit https://groups.google.com/d/optout. > > -- You received this message because you are subscribed to the Google Groups "weewx-user" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
