State Estimation of Grid-Connected Wind Farm

Abstract

AbstractReal-time monitoring of an electric power system (EPS) is very important problem when EPS management. To solve this problem the state estimation (SE) methods are used, which can be divided into static state estimation and dynamic state estimation. Under some conditions, the dynamic state estimation methods can provide more accurate information about the power system than the static state estimation methods. The aim of this study is to develop an algorithm of dynamic state estimation of a grid-connected wind farm equipped with a doubly fed induction generator. The algorithm takes into account the relationship between state variables over time to reduce random errors in measurements and calculate unmeasured state variables. In the developed algorithm, the input data or measurements are: for a wind farm, these are: measurements of the active and reactive power of the stator and rotor, the d-q axis currents, the stator voltage; for other nodes these are: the voltage magnitude, the active and reactive power. The output data are all state variables of the system under consideration. In this study, a three-node test system was considered, consisting of a wind farm equipped with a doubly fed induction generator (DFIG). The wind farm is connected to the power system through a transformer and a line. An analysis of the results shows that the application of the dynamic state estimation procedure leads to a decrease in random measurement errors, and hence to the availability of more accurate information about the state variables of the wind farm.KeywordsDynamic state estimationElectric power systemDoubly fed induction generatorWind farm