Since i was so happy with my solar predictive, I made one for wind prediction, math wise a lot more complicated but it seam reasonable accurate.
first we calculate the air density which was the hard bit. and then we can calculate the the predictive power output from the wind the easy bit
you need to enter in your longitude and latitude, api key Diameter of your wind generator, the efficiency to your wind generator ( <=40% ) and your inverter efficiency or (battery absorption efficiency)
and then you can send it to your database to visualize the information… it not very windy for me lately so I am not sure on it accuracy yet at the higher levels . but as it warms up abit I should be able to get more data. in the mean time if anyone else has a wind generator would mind trying to see how it matches up to general production wind2.tar.gz (2.0 KB)
it works with openweather,weatherapi, weatherbit and aeris currently . I will add in the others later
Thanks for sharing @stephen. From your experience do you find that any of the weather services that you referenced are more accurate than the others? I need to look at this again for a precipitation forecast for a run of the river hydro generator for a project that I am involved with… now that darksky are disabling their api.
that probably a difficult question to answer as I live fairly remote about +200km from the nearest city in the middle of no where I suspect they will be more accurate in heavier populated areas as they dedicate more resources to it … but currently for me Aeris API has being the most accurate ( pretty much dead on most times +90%), then weather api ( +85%), climacell ( 75%), open weather (70%) weatherbit ( it was quite accurate but the last week it being not in the same ball park - 60%) and then visualcrossing .<60%… I would just do one for each API and then sample them it becomes quickly apparent who the most accurate for that particular region
here a small sampling from predictive wind power using aeris as it seams the most accurate for wind in my region as well ( it seams to be always with few km of actual )
BLUE: is aeris - predictive data
RED is weewx weather station data - calculated power output based on output from the weather station
ORANGE is LIVE power output.
it seams pretty close at low wind speeds ( <=10km), but I have no data for high wind speeds… climate change has really messed up wind productivity in my area. 10 years ago it was almost a constant 100w production-- today I can go 2-3 months with out my blades spinning
hi there I combined both the solar and wind into one package and all you need to do is edit START ( add in your APIs , location and other configurations) and panelconf ( your solar panel configuation)
your welcome , I rewrote it with you in mind since you said you were thinking of a predictive with rain forecast for a hydro project. you could simply write a bash script add it in to the START run list and away you go. collecting predictive data on many different aspects . no need for multiple calls to the same API as the JSON is downloaded once and each predictive is generated from the same JSON.
