E3S Web Conf.
Volume 172, 202012th Nordic Symposium on Building Physics (NSB 2020)
|Number of page(s)||8|
|Section||Hygrothermal boundary conditions|
|Published online||30 June 2020|
Predicting solar radiation using a parametric cloud model
University of Science and Technology NTNU, Architecture and technology, Trondheim, Norway
* Corresponding author: email@example.com
In this paper, we evaluate a method to calculate hourly global solar radiation and improve the calculation of diffuse and vertical surface radiation on building facades by accounting for ground conditions based on publicly available data of cloud coverage, temperature and precipitation from a forecast service covering the Nordic countries. The detailed weather forecasts produced by MET Norway provide hourly forecasts for the next 60 hours, and 6-hour predictions for the next week. To calculate solar radiation on cloudy days the clear and cloudy sky MAC model developed by Davies and Mckay (1982) is used. Instead of basing the prediction on ground observations as in the original method, cloud coverage in three levels and total cloud cover is used as input in a cloud product parameterisation. The resulting global horizontal irradiance is validated against the output of the numerical weather prediction (NWP) model and compared to a year of hourly ground measurements in Trondheim, Norway. To evaluate applicability to the building sciences, vertical irradiance measurements are compared to tilted surface irradiance calculated with the ISO 52010:2017 method. For the location, six-hour forecasting performance is on par with the GHI output of the NWP model (using the cloud layer model and the available weather parameters of the location forecast API). To account for the unpredictability of clouds and improve the short-term forecasting performance beyond 38 % RMSD, 38 % SD and 0.80 R2 a different approach is needed, like combining model and sky observations.
© The Authors, published by EDP Sciences, 2020
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.