Development and Evaluation of Typical Plane of Array Year (TPY) for Solar Energy Systems Over the Americas

Aron Habte, Manajit Sengupta, Grant Buster, Yu Xie NREL, Golden, CO, United States

A Typical Meteorological Year (TMY) dataset consists of a concatenation of 12 months of data extracted from a long-term time series of meteorological and radiation data, assuming a horizontal surface. However, the deployment of solar energy systems uses various orientations such as fixed, single- ,or dual-axis tracking systems.  This paper proposes a new method to develop and generate a typical dataset for the specific solar energy technologies (such as photovoltaics) that rely, for example, on fixed planar collectors. This new approach considers the plane-of-array (POA) irradiance as the main driving variable for the selection of 12 typical months that make up the new entity, referred to as  Typical POA Year (TPY). The justification for this new method is that the current application of TMY datasets is prone to bias when converted to POA and can be significantly different from a TPY. This study uses the NSRDB dataset (1998–2018) version 3 to produce both TPYs and TMYs for solar energy and various other applications. Hypothetical solar systems information and POA irradiance data for both fixed-tilt and single-axis tracking are used to generate TPYs and associated generation capacity profiles for both TMYs and TPYs.  A comparison between the new TPY and the existing TMY demonstrates that the latter mis-predicts the energy yield of typical PV systems by about ±5 % for the interquartile range as compared to the more appropriate TPY.