Measured and estimated performance of a fleet of shaded photovoltaic systems with string and module-level inverters

Journal Article · Thu Mar 16 04:00:00 UTC 2017 · Progress in Photovoltaics

DOI: https://doi.org/10.1002/pip.2884 · OSTI ID:1371523

MacAlpine, Sara ^[1]; Deline, Chris  ^[1]; Dobos, Aron ^[1]

---

1. National Renewable Energy Lab. (NREL), Golden, CO (United States)

Shade obstructions can significantly impact the performance of photovoltaic (PV) systems. Although there are many models for partially shaded PV arrays, there is a lack of information available regarding their accuracy and uncertainty when compared with actual field performance. This work assesses the recorded performance of 46 residential PV systems, equipped with either string-level or module-level inverters, under a variety of shading conditions. We compare their energy production data to annual PV performance predictions, with a focus on the practical models developed here for National Renewable Energy Laboratory's system advisor model software. This includes assessment of shade extent on each PV system by using traditional onsite surveys and newer 3D obstruction modelling. The electrical impact of shade is modelled by either a nonlinear performance model or assumption of linear impact with shade extent, depending on the inverter type. When applied to the fleet of residential PV systems, performance is predicted with median annual bias errors of 2.5% or less, for systems with up to 20% estimated shading loss. The partial shade models are not found to add appreciable uncertainty to annual predictions of energy production for this fleet of systems but do introduce a monthly root-mean-square error of approximately 4%-9% due to seasonal effects. Here the use of a detailed 3D model results in similar or improved accuracy over site survey methods, indicating that, with proper description of shade obstructions, modelling of partially shaded PV arrays can be done completely remotely, potentially saving time and cost.

• Research Organization: National Renewable Energy Lab. (NREL), Golden, CO (United States)
• Sponsoring Organization: USDOE Office of Energy Efficiency and Renewable Energy (EERE), Solar Energy Technologies Office (EE-4S)
• Grant/Contract Number: AC36-08GO28308
• OSTI ID: 1371523
• Report Number(s): NREL/JA--5J00-67257
• Journal Information: Progress in Photovoltaics, Journal Name: Progress in Photovoltaics Journal Issue: 8 Vol. 25; ISSN 1062-7995
• Publisher: WileyCopyright Statement
• Country of Publication: United States
• Language: English