Modelling-based approach for digital control design for nonlinear WECS in the power system

The case has been established that the wind power plant must be treated as an integral part of the electric system, thereby constituting the wind energy conversion system. Recent advancement in size and technology of wind turbines requires sophisticated control systems to effectively optimize energy conversion and enhance grid integration. As a first step toward controller design, modelling has become a prerequisite. This paper explores controller design based on modelling the wind speed as a stochastic process, and the wind turbine as a multi-mass system with a soft shaft linking the turbine with the doubly fed induction generator. A control strategy incorporating a linear quadratic Gaussian (LQG) that relies on state estimation for full-state feedback is proposed to augment a linear controller for generator torque control. The control objectives are to reduce stresses on the drivetrain and to ensure operation geared toward optimal power conversion. This study focuses on above-rated wind speeds, and the LQG's main purpose is to add damping to the drivetrain, thereby minimizing cyclic fatigue, while a pitch control mechanism prevents rotor overspeed, thereby maintaining rated power. Simulations show the efficacy of the proposed paradigm in meeting the control objectives. Copyright (C) 2009 John Wiley & Sons, Ltd.