Modelling of Generic Maximum Power Point Tracker for Wind Farms

Abstract

The emission of CO2 is a major problem in the world today due to the greenhouse effect. This is one of the reasons why wind power, as a renewable energy source, is increasing fast in the world today. An effective way of harnessing lots of wind power is to build large wind farms. However, the aero-dynamic interaction among wind turbines, called wake effect, makes control of wind farms complicated. In order to cope with this problem and maximize the production of wind farms, a so-called Maximum Power Point Tracker can be used. This thesis develops and evaluates a Maximum Power Point Tracker algorithm based on optimization. As compared to the previous research in the topic, the algorithm in this thesis takes into account not only the wake effect but also the collector system losses. Moreover, it also considers reactive power dispatch, cable current capability and bus voltage limits. The evaluation shows that the active power output of a sample wind farm consisting of 18 wind turbines, rated 6 MW each, can increase by as much as 1.30 %. At the same time, the losses are decreased by up to 1 % as compared to only considering the wake effect. The corresponding increase by considering the wake effect only is 1.27 %. Finally, the algorithm is also implemented and verified in the power system analysis software DIgSILENT PowerFactory.