How is the 30 min time shift applied in the simulation?

[katherine.alvino]

Dear all,

I don't understand how the time shift works. (From PVsyst help MeteoNote9)

If the meteo data file provides the hourly Global horizontal irradiation, sun's height and azimuth at the exact beginning of the hour, are all these values interpolated to the middle of the interval in order to use them for the simulation?

Wouldn't it add an error to the solar geometry calculation, given that the solar movement isn't linear?

Katherine

[Adrien Villoz]

Dear Katherine,

Most of the radiation data available today are outputs of models, which are already interpolations to match a full hour: data with a time stamp at 9:00 will cover 9:00 to 10:00, which means the solar geometry corresponds to the middle of the hour, 9:30.

Some data providers specifically state that their data are instantaneous values (e.g. NSRDB and PVGIS). For this type of data, a time stamp will be considered as the middle of its time interval: for hourly data, a time stamp at 9:00 will correspond to data from 8:30 to 9:30, so the solar geometry used will be at 9:00.

PVsyst does not do any interpolation, if the input data does not have the time shift specified, there might be a warning (or an error if too high). In the case of a time shift that only gives a warning, the simulation is allowed and it will compute the solar geometry at the middle of the hours. This can lead to errors in the results if you compare with production measurements.

If the data contain a time shift properly defined, then PVsyst will compute the solar geometry for the middle of the interval that was shifted. For example a time shift of -15 minutes will make PVsyst compute the solar geometry for the middle of the interval, for hour H, the interval is (H-1):45 to H:45 and the solar geometry will be computed at H:15.

[Sergio Alonso]

Hello Adrien,

I would like to follow up on Katherine´s question. Could you please provide some explanation on how this minute time shift correction is defined? Is it based on actual irradiance data from the meteo file studying clear days?

Thank you.