The Erbs function in pvlib (see pvlib.irradiance.erbs) can be used to estimate the direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI) from global horizontal irradiance (GHI) however, where DHI is for a horizontal surface, DNI is for a surface perpendicular to the rays, not a horizontal one. How can I estimate the direct horizontal irradiance given GHI? For example, this may be the case where one wants to separate the diffuse and direct component from a measurement of GHI for the same horizontal surface. Is it reasonable to use the Erbs model, compute DHI, and subtract DHI from GHI? I.e. Direct horizontal irradiance = GHI - DHI?
Is it reasonable to use the Erbs model, compute DHI, and subtract DHI from GHI? I.e. Direct horizontal irradiance = GHI - DHI?
Yes.