Turbine system controller design based on extreme learning machine

Abstract

This report emphasize on the control strategy of Wind Turbine System for maximum power extraction from the available wind. A Wind turbine model has been identified in this project with based on the parameters of a low-power rigid-drive-train SCIG-based WECS. The model has parameter with an optimum Tip Speed ratio, λopt= 7 at wind speed of Vwind= 8 m/s, power coefficient, Cpopt= 0.476. The controller design is based on the parameterization of the system response at step changes in generator torque for a given wind speed. Using the transfer function derived from the parameterization, a Proportional - Integral - Derivative (PID) control scheme is designed. After which the PID controller is simulated in a close-loop system and the step response is observed. With the PID parameters derived, the low-power rigid-drive-train SCIG-based WECS is simulated in time domain and the Tip Speed ratio, power coefficient are observed.