Dual Layer Least Mean Fourth Based Control of Grid-Interfaced Wind Turbine Coupled DFIG with Reduced Current Sensors

Abstract

This paper presents a dual layer least mean fourth adaptive filter based algorithm for the grid side converter (GSC) control of a wind turbine driven doubly fed induction generator (DFIG) to extract the active fundamental weights of both load and stator currents in grid connected mode. This is essentially eliminates the use of phase locked loop, abc to dq and dq to abc transformations in GSC control as compared to conventional synchronous reference frame based vector control. Moreover, a reduced sensor based control is incorporated in the rotor side converter (RSC) control of DFIG. In this, rotor currents are estimated in stationary reference frame from sensed stator currents and stator voltages. Therefore, in totality, it reduces two current sensors and results in lower cost and less complexity of the system. To qualify the system performance, simulations are carried out under linear and nonlinear loads, constant and varying wind speed, and nonlinear unbalanced loads. The total harmonic distortions of currents and voltages, are obtained as specified in the standard of the IEEE 519. Moreover, both the grid and DFIG stator are operated at unity power factor. Finally, the system steady state and dyamic performances, are verified on a prototype developed in the laboratory.