| Summary: | When planning new wind farms, the cost for the internal electrical grid is estimated to account for 8-9 % of the total investment cost. On top of that, the cost for accumulated reduced income over several years (due to electrical losses), has to be added. The economic lifespan is relatively long (about 20 years) which means that an error in the loss calculation could lead to a bad decision basis. The accumulated reduced incomes due to electrical losses can be higher than the initial investment cost for the internal electrical grid. Today, Statkraft Sweden is using a template number when estimating the electrical losses in a wind farm. The actual losses are probably lower than the template number which could lead to an unnecessary high loss cost when planning a new wind farm. The aim of this M.Sc. thesis is to develop a better method than the one currently used to estimate the electrical losses in an early stage of the planning process. With necessary information of the proposed wind farm, an early calculation can be done.The user will then have the option to make changes in the wind farm to minimize losses, long before the wind farm is actually built. In the initial stage of the thesis, the components generating losses was modeled using available literature. Next, a MATLAB-program was constructed where the user could build a proposed wind farm consisting of wind turbines, cables and transformers. The program was named Wind Farm Electric System Calculator (WFESC). Finally, WFESC was verified with measurement data from one of Statkraft’s wind farms in Småland, Sweden. WFESC can simulate the electrical losses and production from a wind farm. The simulation takes turbine information, cable types and transformer specifications into account. The user gets detailed information about the different types of losses which makes it possible to minimize losses in the wind farm. In a comparison, WFESC presents a result that differs 5 % from measured data acquired from one of Statkraft’s wind farms in Sweden. The difference between data produced with WFESC and measured data emerges from approximations done in WFESC and difficulties in estimating cable temperatures in a wind farm. The conclusion is that WFESC estimates the internal electrical losses well and can be used when planning a wind farm.
|
|---|
