Nonlinear control schemes for extremum power seeking and torsional vibration mitigation in variable speed wind turbine systems

Abstract

This dissertation presents nonlinear control schemes to improve the productivity and lifespan of doubly fed induction generator (DFIG)-based and permanent magnet generator (PMG)-based variable speed wind turbines. To improve the productivity, a nonlinear adaptive control scheme is developed to maximize power capture. This controller consists of three feedback loops. The first loop controls electrical torque of the generator in order to cancel the nonlinear term of the turbine equation of motion using the feedback linearization concept. The nonlinearity cancelation requires a real-time estimation of aerodynamic torque. This is achieved through a second loop which estimates the ratio of the wind turbine power capture versus the available wind power. A third loop utilizes this estimate to identify the shaft speed at which the wind turbine operates at a greater power output. Contrary to existing techniques in literature, this innovative technique does not require any prior knowledge of the optimum tip speed ratio. The presented technique does not need a dither or perturbation signal to track the optimum shaft speed at the maximum power capture. These features make this technique superior to existing methods. Furthermore, the lifespan of variable speed wind turbines is improved by reducing stress on the wind turbine drivetrain. This is achieved via developing a novel vibration mitigation technique using sliding-mode control theory. The technique measures only generator speed as the input signal and then passes it through a high-pass filter in order to extract the speed variations. The filtered signal and its integral are then passed through identical band-pass filters centered at the dominant natural frequency of the drivetrain. These two signals formulate a sliding surface and consequently a control law to damp the drivetrain torsional stress oscillations caused by electrical and mechanical disturbances. This technique provides a robust mitigation approach compared with existing techniques. These control schemes are verified through holistic models of DFIG- and PMG-based wind turbines. Except for wind turbine aerodynamics, for which an existing simulator is used, the developed models of all components including DFIG, PMG, converters, multi-mass drivetrain, and power line are presented in this dissertation.