An Integrated Wind Resource Planning and Control Framework

Abstract

This thesis presents a novel integrated planning and control framework for optimal wind farm power generations, with increasing the wind energy penetrations in the modern power grids, it is necessary to monitor & control wind turbines for power system stability and reliability enhancement. Wind speed varies with location and timing. However, most of the studies often consider the wind is distributed uniformly across the rotor blades, which affects the effectiveness of real-time measurements and control of wind turbines. In this thesis, advanced time series processing methods together with artificial intelligence techniques are introduced for modelling and control of wind turbines to enhance the real-time measurement efficiently and improve the response time. Intelligent controllers are proposed for two different grid connected wind generation technologies, i.e. doubly fed induction generator wind turbine and permanent magnet synchronous generator with fully rated converter wind turbine. The studies are conducted and verified using integrating DIgSILENT PowerFactory and Matlab/Simulink simulation software. Also, different models and techniques are used in the thesis to improve wind power point prediction. This is achieved by decomposing the original wind (speed or generation) time series into sets of intrinsic membership functions and reconstructing it back into three components, and then different artificial intelligence methods are applied to enhance the short-term wind power prediction. The proposed model is tested using real-time wind speed data, provided by the Australian Bureau of Meteorology. Moreover, a new optimisation technique based artificial intelligence methods have been developed for optimal short-term wind power prediction intervals. Overall, the thesis provides a comprehensive wind power studies with useful control framework for practical wind power grid connection studies.