| Summary: | Conventionally the power network operators were obliged to buy all the wind energy generated by wind farms. However, as the penetration of wind energy (or generally any other sort of renewable source) in a power system is increased, the ability of other generators to balance the demand becomes limited. This will necessitate the control of wind turbines in order to generate a given demand power rather than extracting the maximum wind power. This control approach is termed “Power Demand Control” in this thesis. In contrast to Power Demand Control, “Power Smoothing Control” utilizes energy storage systems in order to absorb high frequency wind fluctuations, hence, delivering a smoother version of wind power into the grid/load. The drawback of the Power Smoothing approach is that the average power into the grid/load is still determined by the available wind power rather than the system operator. The Power Demand Control approach, which has received little attention in literatures, is the main focus of this thesis. This research proposes control schemes with and without external energy storage for the Power Demand Control strategy. This thesis studies different possible methods of applying Power Demand Control, in particular the droop control method. It is shown that a droop-controlled wind farm does not need a central “Supervisory wind Farm Control” unit to determine the power demanded from each DFIG. Moreover, a droop-controlled wind farm has the advantage of controlling the local grid voltage and frequency. This means that no external voltage and frequency source is required which makes a droop-controlled wind farm a more suitable option for integration of wind energy at distribution level. The classical droop control is modified in order to make the DFIGs share the demand power not only according to their ratings but also to their associated available wind power. The applications of the control paradigm are discussed, including: integration into microgrids, AC grids and HVDC connection feeders. This work mainly concentrates on microgrid applications. An Energy Management System is proposed in order to keep the energy level of the energy storage (or the DFIG’s shaft speed) within its limits using an Auxiliary Generator and a Dispatchable Load. Different possible system configurations are introduced and their advantages and drawbacks are discussed. It is illustrated through simulation that the proposed control scheme can inherently ride-through a grid fault with no need for communication. Furthermore, it is shown that the control scheme can operate if the wind speed drops to zero. The simulations are carried out using the PSCAD/EMTDC software.
|
|---|
