Decomposition & transposition
Modelling recommendations
In some PV simulation software, for example PVsyst, it is possible to choose between different models for irradiance decomposition and transposition.
Recommended model choices in PV simulation software: Perez models [1,2]
Background: State of the art & knowledge gaps
In the modelling of plane-of-array irradiance (POA), global horizontal irradiance (GHI) is decomposed into diffuse (DHI) and direct normal irradiance (DNI). These components, together with ground-reflected irradiance, are then transposed to the desired tilt and orientation. In Norway, a key challenge for accurate irradiance modelling is the representation of diffuse irradiance, as commonly used empirical coefficients are typically derived for regions with different diffuse irradiance conditions. Similarly, low solar elevation angles are typically associated with large model biases in physical models. Both challenges are linked to greater air mass values that characterize high latitudes. To mitigate them, several decomposition and transposition models implement upper limits for both air mass and solar zenith, hence introducing significant simplifications in the calculation when the sun is close to the horizon. Validation has been performed at a limited number of sites [1,2], but comprehensive validation under Nordic conditions, covering multiple sites and a wide range of weather conditions, is still required. A key challenge for model validation is the limited availability of high‑quality ground‑based measurements of GHI, DNI, DHI, and POA. In FME Solar WP2, ongoing research focuses on improving irradiance modelling under Nordic conditions using machine learning methods.
ReFERENCES
[1] D.C. Martinsen, M.B. Øgaard, M.M. Nygård, M.S. Wiig, M. Di Sabatino, E.S. Marstein, H.N. Riise, Evaluation of irradiance decomposition and transposition models for a Nordic climate, EUPVSEC proceedings, 2022.
[2] M. Manni, A. Nocente, G. Lobaccaro, Performance variability of solar irradiance model chains applied to building-integrated photovoltaic systems at high latitudes, Solar Energy, 2024.
Last update: 27.4.2026