Dynamical Equivalent Considering LVRT For DFIG

The wind power has been rapid development. People have awared the importance of the reseach, related to the impact on wind power integration. The reach is concerned with the dynamic characteristics of wind farm at point of common coupling, not its internal operation. The cost is large while all wind turbines are on the detailed modeling. So it’s necessary to establish a reasonably and accurate dynamic model of wind farm.Now, there is little paper considering the effect of the low voltage ride through. In this paper, the dynamic equivalent model of the doubly fed induction wind turbine is studied during the crowbar in. The main contents are as follows:First of all, the mathematical model of DFIG is established. Then its static power flow model is built by the internal power balance and the power exchange between the grid and wind farm. The flow calculation provides a group of accurate initial value for the transient analysis of a wind turbine.Secondly, the oscillation charcterics of the wind exports are studied. We have found that only a small part of the oscillation modes occupy most of the total energy, regardless how much the order of the modek is taken. These modes have an important influence on the fitting result and most of the rest can be ignored. Energy ratio has been defined to measure the contribution of each mode. Then the main modes of current and voltage at the export of wind farm are extracted by it. A dynamic equivalent model of the wind farm is established under its export dynamic features, ignoring its internal structure. This model can accurately retain its original dynamic features.Finally, the main error has been pointed out, which is from crowbar cutting out of sync time through the analysis. The factors which affect the crowbar action are analyzed, and a new clustering algorithm is proposed based on rotor current. A conclusion is made through a lot of simulation, that the oscillation characteristics of rotor current can be expressed as a superposition of the largest energy mode on the monotonous modes and the slowest decay mode on the non-monotonic modes. The energy coefficient and the attenuation coefficient are defined as clustering index and Wind Turbine Generators (WTGs) are clustered according to these. A multi-machine equivalent model is built which can accurately reflect the dynamic chaeacteristics of the wind farm.