When performing tests on a photovoltaic energy conversion system, it is useful to have a daily-term irradiance profile which, in some way, represents the typical behavior of this variable at a desired geographic location, including fast changes caused by weather. Therefore, the objective of this work is to propose a simple methodology, to choose a representative daily-duration irradiance signal, from long-term irradiance data. This is accomplished, by means of quantifying the irradiance variability, and considering the total daily energy. A method to combine these two factors, is proposed and employed in the selection the most representative day: it is also used to search for extreme condition days. (C) 2018 Elsevier Ltd. All rights reserved.