Principal Component Analysis for the Characterisation of Spatiotemporal Variations of the Solar Resource in Urban Environments

Abstract

Urban areas are serious candidates for the production of solar energy but their intrinsic complexity makes it challenging. The heterogeneity in the geometries and radiative properties of the different elements composing the urban fabric, specifically induces important spatiotemporal variations of the distribution of incident solar radiations. Besides, Principal Component Analysis (PCA) has been widely validated as an efficient tool to identify the principal behavioural features of a high-dimensional physical model. This paper proposes a novel approach to analyse and characterise the spatiotemporal variability of the solar resource within an urban context by means of PCA. A theoretical 100 × 100 m 2 asymmetric urban district made of nine cuboids with various heights is studied. The distribution of the incident field of irradiances is modelled via backward Monte-Carlo ray tracing over a full year on the facets of the central building under a clear sky, with a 15 min timestep and 1 m spatial resolution. PCA is subsequently applied to the simulated model to analyse its spatial and temporal variabilities. First results validate modal decomposition as a powerful technique for the analysis of the variability distribution, allowing the identification of the district areas subjected to important spatial and temporal variations of the solar resource. Characteristic scales are clearly represented by orders of decomposition. The contribution of surrounding geometries is also transcribed by particular spatial modes and similar influential variables are encountered across multiple evaluated surfaces but at different modal ranks.