Assessing ice rate and ice load at Davvi wind farm using WRF model with wind turbine scheme

Abstract

Wind farms located in cold climates are exposed to temperatures under the design limit and icing conditions, where the main challenges are production losses and mechanical failures. The goal of this thesis is to investigate the impact related to icing on the location of Davvi wind farm using Weather Research and Forecasting (WRF) model. The thesis have investigated the icing intensity and production loss at Davvi wind farm using WRF with a wind turbine scheme.

Weather parameter from the model are used to calculate icing hours above the icing rate thresholds of 10 g/h, 50 g/h and 250 g/h for the proposed turbine locations. The Makkonen ice accretion model is used for icing rates calculations. To estimate production loss, an experimental method where ice build up only occur during an ongoing icing period is used.

Estimation show that the turbine location higher up in the terrain experience greater icing severity, due to lower temperature and higher liquid water content (LWC). Icing estimation show a greater variability in icing severity than what stated in NVE's 2009 report of Norway. Turbine location T26 with the highest icing severity experienced 17.75\% annual meteorological icing, while T22, the location with lowest icing severity, experienced 3.17\%. Based IEA Task 19 Ice Site Classification, this correspond to an annual production loss of > 20\% and 3-12\%, respectively. The experimental production result show a slightly lower production loss, where the locations had an annual production loss of 19.86\% and 4.16\% for 2017, respectively. Differences between data with and without turbines, shows that wind turbines positioned downstream, relative to the wind direction, experienced wake effect created by upstream turbines. The maximum annual mean wind speed difference is 1.425 m/s, and the greatest difference in icing rate above 10 g/h is 113 hours. Icings maps like in NVE's 2009 report, where the wind turbine scheme is not included, could therefore expect higher icings rate than what a wind park would experience. This show the importance of including wind turbine for wind and ice siting at a location for accurate assessments.