Multi-Objective Genetic Algorithm for the Optimization of a PV System Arrangement
Abstract
Urban landscapes feature complex topographies and many shadow casting elements, which can jeopardize the energy yield of building integrated photovoltaic (BIPV) systems. Phenomena like partial shading have a significant impact in the electric performance of PV modules, mostly when systems are deployed in conventional arrangements regardless of surrounding obstructions. The goal of this assessment is to test a multi- objective genetic algorithm (MOGA) in order to find one optimal string sizing and tiling, considering two different scenarios: a west facing building facade and a south facing rooftop, both located in Lisbon, Portugal. Relevant loss mechanisms are considered, such as hourly solar irradiance changes caused by shadow events, high incidence angles and module temperature. Optimization processes allow a reduction of around 24% and 23% in the cost of energy for the rooftop and the facade, respectively, when compared to a scenario that considers individual modules with micro-inverters.
Keywords
Photovoltaic, Partial shading, Genetic algorithm, Multi-objective optimization