Solar Resource Estimation Using a Radiative Transfer with Shading (RTS) Model

A combination of falling technology prices and government financial incentives has led to the rapid expansion of the solar microgeneration industry in the UK and around the world. With conditions for investment becoming more favourable, viability appraisal is now of great interest to potential investors ranging from individuals to large companies and authorities. A methodology to predict solar resource available to solar technologies in urban areas is presented that combines a radiative transfer simulation with shading derived from digital surface model (DSM) data. The radiative transfer simulation includes water vapour, ozone, clouds and surface albedo determined from MODIS satellite products along with aerosol properties that are derived from the GLOMAP model. A DSM is used to calculate the height of near-by obstacles and the horizon from each point of interest to establish when the site is in shade during a year. The site-level radiance field is adjusted by elements in shade and integrated over the view hemisphere of the photovoltaic (PV) panels to take into account panel tilt. Modelled annual global radiation (Wh/m2/a) estimations are validated using power output data from 17 sites across four major UK cities (Bristol, Cambridge, Leeds and Sheffield) and show good agreement with -3.68% and +2.62% mean percentage error under assumed performance ratio conversions of 0.75 and 0.8 respectively. The results are compared to the outputs of both Esri ArcGIS and PVGIS, the first of which predicted annual global radiation with -15.97% and -20.78% mean percentage error with the latter returning +3.34% and +10.23% mean percentage error for the 0.75 and 0.8 performance ratios.