This paper presents a framework for modeling the performance of photovoltaic (PV) systems, which is intended to be used in the design, optimization, and deployment of PV systems in urban environments. Urban PV and building integrated PVs (BIPV) are a crucial component of the transition to a low-carbon society. However, existing frameworks for simulating PV systems accurately characterize PV under partial shading, leading to performance gaps with systems deployed in environments such as cities where shading is more common than in large open fields. In this paper, we extend a framework from literature for modeling parametric BIPV arrays using a power model that characterizes IV-curves at the PV cell. We describe the framework and perform a validation against a measured rooftop PV array dataset. We then evaluate the extended framework to other PV modeling and simulation frameworks found in the literature through the simulation of a simplified facade under various levels of shading. Results show that the extended framework is necessary to account for partial shading, but there may be use cases when lower resolution frameworks may be just as accurate. Amongst the frameworks a range of predicted power outputs under the shading conditions was observed; from 6% to 846% with a mean of 162% from the other frameworks relative to the proposed framework. The range is due to the extensive set of conditions tested, where it was observed that in conditions where shading is present but sparse, there is a larger range of predicted output from the various frameworks. Whereas in conditions where shading coverage is high this range is lower. Lastly, we demonstrate the use of the framework on three arrays. Again it was observed that homogeneous shading conditions may not require the level of resolution provided by the proposed framework, while heterogeneous could benefit the most from the approach. This indicates, and is discussed in the paper, that modelers should consider the shading conditions present when deciding on which framework to apply.
Read full abstract