EU PVSEC Programme Online
EU PVSEC 2020, 7 - 11 September 2020
Presentation: 5DO.1.6 Estimating Snow Losses for Many Sites Using Minimal Data Sources
Type: Oral
Date: Thursday, 10th September 2020
08:30 - 10:00
Author(s): M. van Noord, T. Landelius, S. Andersson
Presenter / Speaker: M. van Noord, RISE, Stockholm, Sweden
Event: Conference Conference
Session: 5DO.1 Soiling
Type(s) of Access:  Conference Registration
Topic: 5. 3 Operation, Performance and Maintenance of PV Systems
Summary / Abstract: The main aim of this work is to create an improved model to predict annual and monthly snow effects, mainly snow losses in PV yield, on rooftop PV installations. The model should also guide PV designers towards better design choices regarding e.g. module mounting orientation (portrait vs landscape) and tilt angle. In this abstract two parts of the work are presented: the impact of module mounting orientation on snow-losses and a method to calculate snow-losses using only PV production data and general weather data and models. The impact on snow losses of "landscape” or "portrait" module orientation, due to enabling/disabling bypass diodes, is estimated for at least six on-site monitored PV sites. This estimation uses on-site production measurements, two images daily of module snow-coverage and a translation model from actual snow coverage on portrait modules to assumed snow coverage on landscape modules. For estimation, bypass diode behavior is assumed ideal and lossless. To calculate and model snow losses, historical and current production data from almost 400 photovoltaic plants in central and northern Sweden are collected and combined with weather measurements and models from the Swedish Meteorological and Hydrological Insititute. Innovative algorithms to estimate the azimuth, tilt angle, shading profile and snow coverage of the solar modules from hourly data are developed in order to model the systems in a snow free state. Production losses and gains due to snow are then calculated from the snow-free model and measured production. The developed models are validated using three PV sites with on-site measurements of irradiation and temperature as well as cameras for snow coverage identification.