Application of extended Kalman filter to parameter estimation of doubly-fed induction generators in variable-speed wind turbine systems

Abstract

This paper proposes a parameter estimation method for doubly-fed induction generators (DFIGs) in variable-speed wind turbine systems (WTS). The proposed method employs an extended Kalman filter (EKF) for estimation of all electrical parameters of the DFIG, i.e., the stator and rotor resistances, the leakage inductances of stator and rotor, and the mutual inductance. The nonlinear state space model of the DFIG is derived and the design procedure of the EKF is described. The observability matrix of the linearized DFIG model is computed and the observability is checked online for different operation conditions. The estimation performance of the EKF is illustrated by simulation results. The estimated parameters are plotted against their actual values. The estimation performance of the EKF is also tested under variations of the DFIG parameters to investigate the estimation accuracy for changing parameters.

Bank of Extended Kalman Filters for Faults Diagnosis in Wind Turbine Doubly Fed Induction Generator

“…The estimation of parameters using observers or filtering approaches is an indirect procedure, which consists of augmenting the state vector by defining the monitored parameters as additional state variables and the parameters faults are detected through evaluating deviation in the parameters values from their predefined norms. For instance, In order to estimate both the state and the unknown inputs of the DFIG, an unknown input observer (UIO) is used in [11]. The extended kalman filter is used for electrical parameter estimation of doubly-fed induction generators (DFIGs) in variable speed wind turbine systems (WTS) in [12].…”

Review on model validation and parameter estimation approaches of wind power generators

“…The remaining parameters are assumed to be known. In [14, 15], the authors can estimate all the electrical parameter by the Kalman filter. In [14], the authors show that the unscented Kalman filter (UKF) has the better performance than the extended Kalman filter (EKF).…”

Sensorless control of doubly-fed induction generators in variable-speed wind turbine systems

“…For discretization the (simple) forward Euler method with sampling time T s [s] is applied to the time-continuous model (13) with (14), (15) and (16). For sufficiently small T s 1, the following holds…”

“…However, the injection of high-frequency signals in the DFIG rotor is not easy for large machines (> 1 MW) such those in modern WTS. Another alternative is the use of an extended Kalman filter (EKF) which has already been used for sensorless control and the estimation of the electrical parameters of induction machines and permanent magnet synchronous machines [15], [16]. An EKF was used for DFIG speed and position estimation in [17], however the authors use state variables in the rotating reference frame, whereas input and measurement variables are in the stationary reference frame and are directly incorporated into the EKF design.…”