A Morphological Clustering Method for daily solar radiation curves

► Method based on Mathematical Morphology for the classification of solar radiation curves. ► The granulometric size distributions of the original daily radiation curves are obtained. ► Cloud transitions and variability in direct radiation are simultaneously taken into account for clustering. ► Some advantages with respect to daily clearness index distributions are found. We present a new method based on Mathematical Morphology techniques for the classification of solar radiation curves that we call MfCM. The main advantage of using MfCM as opposed to daily clearness index distributions is that it allows us to keep the dynamics of the solar radiance curves in the analysis: both cloud transitions and variability in direct radiation are simultaneously taken into account. To illustrate our proposal, we use a set of real radiation data collected in a location sited in southern Spain.