Neural Control Design for Isolated Wind Generation System Based on SVC and Nonlinear Autoregressive Moving Average Approach

In this paper, the voltage and frequency control of an isolated self-excited induction generator, driven by wind turbine, is developed with emphasis on nonlinear auto regressive moving average (NARMAL2) based on neural networks approach. This has the advantage of maintaining constant terminal voltage and frequency irrespective of wind speed and load variations. Two NARMA L2 controllers are used. The first one is dedicated for regulating the terminal voltage of the induction generator to a set point by controlling the thyristor firing angle of a static reactive power compensator (SVAR). The second one is designed to control the mechanical input power to the generator via adjusting the blade pitch angle of the wind turbine. In this application, an indirect data-based technique is taken, where a model of the plant is identified on the basis of input-output data and then used in the model-based design of a neural network controller. The proposed system has the advantages of robustness against model uncertainties and external disturbances. The robustness of the wind-energy scheme has been certified through step change in wind speed. Moreover, the system is tested also during a step change in load impedance. Simulation results show that high dynamic performance of the proposed wind energy scheme has been achieved. Key-words:wind turbine, induction generator, NARNA-L2 controller, neural networks.