Search documents

Browse topics

Document details

Investigating a Potential Linear Model for Prediction of Monthly Snow-Induced Production Losses for Rooftop PV
M. van Noord, M. Murphy, T. Berglund
Rooftop, System Performance, Modelling / Modeling, Snow
PV System Performance and Integration
Subtopic: Design and Operation of PV Systems
Event: 33rd European Photovoltaic Solar Energy Conference and Exhibition
Session: 6BV.1.43
2316 - 2319
ISBN: 3-936338-47-7
Paper DOI: 10.4229/EUPVSEC20172017-6BV.1.43
0,00 EUR
Document(s): paper, poster


This study investigates the potential of an easy applicable linear model of typical yield losses from PV rooftop systems due to snow coverage, by relating these losses to air temperature. The basis for this study is measurements conducted during two winter seasons, on three locations and a total of six PV systems. A yield prediction model based on “snow free” Performance Ratio is used to calculate daily and monthly snow losses. Correlation of monthly snow loss factors (% of expected yield) with average monthly air temperature is visually analyzed. Three temperature correlation zones are identified: A zone with practically no snow losses, a transition zone and a zone with almost full snow losses. A linear regression is used to model snow loss factors for the transition temperature zone. Comparison with historical data indicates that an improvement of the transition zone model is desirable. Addition of snowfall amounts and/or global irradiation into the model are proposed for future development.