Wind Power and Natural Disasters

Abstract: Wind power can be related to natural disasters in several ways. This licentiate thesis gives some background and introduces four papers devoted to two aspects of this relation. The first section looks into how small-scale wind energy converters (WECs) could be used to generate power after a natural disaster. For this application diesel generators are the most common solution today, but there would be several advantages of replacing these systems. A study of off-grid systems with battery storage at 32 sites showed that photovoltaics (PV) were more suitable than WECs. The results were confirmed by a study for the entire globe; PV outperformed WECs at most sites when it comes to small-scale application. This is especially true for areas with a high disaster risk. Hybrid systems comprising both PV and WECs are however interesting at higher latitudes. For the Swedish case, it is shown that gridded data from a freely available meteorological model, combined with a statistical model, give good estimates of the mean wind speed at 10 meters above ground. This methodology of estimating the mean wind speed can be used when there is no time for a proper wind measurement campaign.The second section is directed towards wind power variability and integration. The results presented in the thesis are intended as a basis for future studies on how a substantially increased wind power capacity affects the electric grid in terms of stability, grid reinforcement requirements, increased balancing needs etc. A review of variability and forecastability for non-dispatchable renewable energy sources was performed together with researchers from the solar, wave and tidal power fields. Although a lot of research is conducted in these areas, it was concluded that more studies on combinations of the sources would be desirable. The disciplines could also learn from each other and benefit from the use of more unified methods and metrics. A model of aggregated hourly wind power production has finally been developed. The model is based on reanalysis data from a meteorological model and detailed information on Swedish WECs. The model proved very successful, both in terms of low prediction errors and in the match of probability density function for power and step changes of power.