Research On Stand-alone Wind Power Systems And Control With Doubly-fed Induction Generator

Double Fed Induction Generator (DFIG), with its advantages of low costs, adjustable output voltage frequencies and amplitudes, and minimum required capacity of power electronic devices, has been widely used in wind power system. The DFIG-based stand-alone wind power generation system is an eco-friendly distributed Generation System, able to efficiently solve the problems of power supplies on remote islands, on which have abundant wind energy but are not covered by power grids. The system is to capture the maximum wind energy by adjusting the generator output power with energy storage devices, and to guarantee uninterrupted power supply. However, the generating cost is too high. And without power grids, the stator voltage is easily affected by wind speed fluctuation, load switching and generator parameter errors. As a result, the power quality will definitely not be satisfactory. Therefore, it is meaningful to enhance the competitiveness of stand-alone wind power system by reducing generating cost and researching on the optimized operation control methods that improve the stability of the system.The paper analyzes comprehensively the research history and current status of domestic and overseas stand-alone wind power systems, builds mathematical models of the stator /rotor side converter, the battery unit and DFIG model in parameter errors status, and develops a control scheme for converters. In response to the excessive costs of energy storing devices, based on the optimal power curve, the torque reference value is calculated by testing the rotor speed. In comparison with the value of the electromagnetic torque, the error signal obtained is used to control the variable load to achieve maximum power tracking control.The Matrix Converter (MC) provides excitation current for DFIG In view of the deficiencies of MC Repetitive Control (RC) and PI control in asymmetrical load cases, a combined control strategy of PI and RC is proposed. To solve the delay problems of the internal model signal conditioning, by analyzing the distribution of the output voltage harmonic, double sequence park transformation is adopted to transform the harmonics in phase coordinates into triple harmonics in dq coordinates. The internal model is improved and the conditioning time of the control system is reduced without affecting the performances of harmonic suppression, which effectively solve the imbalance of three-phase output voltage in unbalanced load cases.As the DFIG-based stand-alone output voltage is easily affected by load switching and the PI controller depends highly on the motor parameters, the mathematical models are built to this end, including the excitation current dynamic process model and the motor parameter error model. By applying separation principles and pole-zero placements, the double-closed-loop Auto-disturbance Rejection Controller (ADRC) is designed, considering the motor parameter errors and the excitation current dynamic process as the internal and external disturbances of the system. The extended state observer of ADRC is used to estimate the total disturbances of the system in real time and the disturbances could be compensated by feed forward. The operation simulation model of the megawatt-level DFIG-based stand-alone with ADRC is established in the Matlab/Simulink environment. The simulation results show that the proposed ADRC can actively suppress the voltage disturbances and the parameter variations of DFIG caused by load switching, and greatly improve the robustness of the system.