Add historical forecast data e.g. day-ahead

[pz-max]

Detailed Description

Add historical forecast data (e.g. day-ahead) to quantify forecast errors for renewables

Context

Economic dispatch optimizations for different energy market regimes (intraday, day-ahead) require historical forecast data. When available, it is also possible to quantify typical forecast error e.g. by comparing predicted (e.g. day-ahead) vs actual observed/estimated data. An example of how this could look is given by the DWD who quantifies the day-ahead vs actual observed forecast error with 15min resolution: https://www.dwd.de/EN/research/weatherforecasting/num_modelling/07_weather_forecasts_renewable_energy/weather_forecasts_renewable_energy_node.html

Possible Implementation

Possible API/ inspiration:

• https://github.com/open-meteo/open-meteo
• https://github.com/blaylockbk/Herbie
• https://openweathermap.org/

[TimFuermann]

I have already done something similiar and will push it up until mid of november., needs only some fixing due to the last update in atlite. Best, Tim