Monthly Solar Radiation Maps for Morocco Based on the Visualization of Clustering

Abstract

Assessment of the solar radiation potential is a necessity for the establishment of solar projects. It is difficult to evaluate the solar energy potential for areas far from the meteorological stations. Current methods for estimating solar energy are site-dependent or limited to certain areas. Therefore, a method for solar radiation maps using unsupervised machine learning techniques and satellite products is proposed. The method uses the hierarchical dendrogram to determine the suitable number of distinct areas for each month, then this number is used as input for Agglomerative Hierarchical Clustering (AHC) that clusters areas into the given number for each month. The maps are homogeneous in space and reflect the seasonal character of solar radiation. The resulting solar maps cover all regions of Morocco and can help for preliminary assessment and decision making, especially for areas far from solar radiation ground stations.