The impact of wind power variability on the least-cost dispatch of units in the electricity generation system

Doctoral thesis, 2014

This thesis investigates the dynamics of electricity generation systems that involve high levels of wind power. Methods to account for wind power variability in electricity-dispatch models are explored, and the impact of wind power variability on the optimal output of the generation units, so as to meet the system load with the lowest possible cost (economic dispatch), is analyzed for several systems, for both regional and European cases. Systems that lack active variation management, as well as systems with variation management through storage, charging of plug-in hybrid electric vehicles, and trade with the hydropower-rich Nordic countries, are investigated. The work considers systems in which wind power supplies between 20 % and 40 % of the electricity demand on an annual basis. From the work of this thesis, it is concluded that the inclusion of cycling costs can have significant impacts on the capacity factor of individual generation units obtained from modeling the regional dispatch of systems with 20 % wind power penetration (i.e., annual wind power generation relative to annual demand for electricity). Whether cycling costs need to be included in the systems analysis depends on the wind penetration level and the research question being posed, as well as the relationship between the cycling costs and running costs of the thermal units in the system. Furthermore, it is shown that in a wind-thermal system, in which wind power generation corresponds to about 20% of the demand for electricity, the variations in net load (here defined as the demand for electricity reduced by wind power generation each hour) follow a diurnal pattern, and load shifting from day to night reduces the competition between wind power and thermal generation with poor cycling properties. However, in systems with about 40% wind power, the ability to store electricity or to shift the load over longer time periods (i.e., several days) confers significant advantages compared to load shifting from day to night, owing to the altered pattern of the variations in net load. Finally, this thesis shows that the role of the Nordic electricity-generation system in the European context relies heavily on the balance between investments in interconnector capacity and investments in Nordic generation capacity. If planned interconnections between Norway and the rest of Europe are established, net export of electricity is likely from the Nordic countries to Germany and the UK, whereby hydropower-rich Norway would play a central role in redistributing electricity from high-wind events to peak-load events.