Robust Active and Reactive Power Control Schemes for a Doubly Fed Induction Generator Based Wind Energy Conversion System

Ogeti, Pedda Suresh (2016) Robust Active and Reactive Power Control Schemes for a Doubly Fed Induction Generator Based Wind Energy Conversion System. PhD thesis.



In view of resolving rising environmental concern arising out of fossil fuel based power generation, more electricity has to be generated from renewable energy sources. Out of the several renewable energy options available today, wind energy is considered to be the most promising one due to its high energy conversion efficiency compared to one of its competitors, i.e. the solar photovoltaic system. Now-a-days, large wind farms are generating thousands of megawatts of power feeding to the grid. In literature, number of controllers such as conventional proportional integral (PI) control, linear parameter varying (LPV) control, gain scheduling control, robust control, model predictive control have been proposed for both torque and pitch control. In these controllers, some of the important issues such as robustness for nonlinear dynamics of wind turbine and stability are not considered simultaneously. Hence, it is necessary to design appropriate controllers for extracting maximum power from the wind turbine whilst the robustness and stability of the Wind Energy Conversion System (WECS) are ensured. Hence, in this thesis, firstly the focus is made to design control system for the wind turbine coupled with the DFIG (torque and pitch control) using one of the very promising robust control paradigm called sliding mode controller for achieving robustness, reducing chattering phenomenon and stability of the WECS. Since the number of terms in control inputs (i.e. torque and pitch angle) and outputs (i.e. DFIG output power and speed) are more in wind control dynamics, selection of significant terms is important for reducing the complexity of controlling. Therefore, a Nonlinear Autoregressive Moving Average with exogenous input (NARMAX) model of the WECS has been developed. The parameters of this NARMAX model are estimated by suitably designing an on-line adaptive Recursive Least squares (RLS) algorithm. Subsequently for controlling speed and achieving efficient power regulation of the WECS a nonlinear model predictive controller (NAMPC) has been developed in which the control variables (torque and pitch) are optimised by formulating a cost function. Subsequently for the WECS, the power converters connecting the DFIG to the grid have been designed. For controlling stator active and reactive power of DFIG connected to the grid, a state feedback controller for the DFIG has been developed using a linear quadratic optimal theory with preview concept. This Linear Quadratic Regulator Optimal Preview Control (LQROPC) scheme is employed with a stator voltage oriented control (SVOC) technique. This Optimal preview control is used to solve the tracking and rejection problems with an assumption that the signals to be tracked or rejected are available a priori by a certain amount of time. Even though the OPC provides very good tracking and disturbance suppression performance, but it is sensitive to the DFIG circuit parameters which makes the WECS system unstable. Hence, to address the parameter uncertainty of the DFIG, a sliding mode controller has been proposed and the robustness of the WECS have been verified by using the Lyapunov criterion. Then, a 2 kW DFIG based WECS experimental setup has been developed in the laboratory to study the effectiveness of the controllers developed.

Item Type:Thesis (PhD)
Uncontrolled Keywords:WECS, DFIG, NARMAX, LQROPC, PWM, RSC,GSC
Subjects:Engineering and Technology > Electrical Engineering > Non Conventional Energy
Engineering and Technology > Electrical Engineering > Power Electronics
Divisions: Engineering and Technology > Department of Electrical Engineering
ID Code:8207
Deposited By:Mr. Sanat Kumar Behera
Deposited On:28 Nov 2016 15:06
Last Modified:28 Nov 2016 15:06
Supervisor(s):Subudhi, B and Panda, A K

Repository Staff Only: item control page