Direct Control of the Rotor Voltage of a Doubly Fed Induction Generator by means of a Restricted Optimization Process in Real Time
Doubly Fed Induction Generator; DFIG; Constrained Nonlinear Optimization; Direct Rotor Control.
The constant growth in the potential for generating energy from wind sources has been highlighted on the world stage. Global behavioral changes in search of sustainable development and consumption, replacing fossil fuels aiming at improving the quality of life and preserving the planet, are important factors for the continuation of this growth, due to the increasing need for renewable energy sources. The wind energy conversion systems that use the Doubly Fed Induction Generator - DFIG, are the most popular today. Advantages such as reduced mechanical load, simple picth control, control of active and reactive power, lower power converters that mean less switching losses, reduced converter costs, less harmonic injection into the network, put it ahead of turbine topologies with synchronous generators or induction generators with squirrel cage rotors. This thesis aims to present and demonstrate the feasibility of using a new approach on voltage control, active and reactive power control and power factor control through direct control of the voltage applied to the DFIG’s rotor, by the Rotor Side Converter - RSC. This voltage is determined analytically, solving a restricted optimization process in real time. A system of equations derived from the equivalent circuit is adopted to represent the desired operating point of the DFIG, and the solution of this system defines the values of the rotor supply voltage coordinates, in steady state. A new control strategy is proposed to reach the rotor voltage reference values, without violating the limits existing in other variables. This strategy is designed so that the speed of evolution of the voltage applied to the rotor, is governed by the dynamic evolution of the mechanical speed of the rotor. An optimization process was formulated to minimize the time of convergence of the rotor speed, restricting transitory variations in the net power generated, in order to accelerate the machine, without exceeding the current limits. Following recent trends in solutions with a reduced number of sensors, only measurements obtained from the stator sensors (voltages and currents) are used. In this way, the angular velocity and rotor currents are estimated in real time. An algorithm for estimating inductance is also included, preventing deviations from the nominal value could lead to false reference voltages or changes in reactive control. In addition, a method for defining the pitch angle and the reference speed is proposed, using the Newton-Raphson numerical solution method.