Forecasting and control for wind power systems

Download files
Access & Terms of Use
open access
Copyright: Khalid, Muhammad
Altmetric
Abstract
Wind energy has become the world’s fastest growing source of clean and renewable energy and now contributes a large proportion of total power generation. This proportion will continue to increase because of the global preference for a clean and renewable energy source. However, wind power is difficult to integrate into traditional generation and distribution systems with current technology because it is intermittent, unpredictable and volatile. Thus, it is difficult to match wind generation to energy demand, and the imbalances between demand and generation can cause adverse voltage variations. This power quality problem cannot be solved effectively only by renewable generating technology and/or power electronics. The adverse effects of wind generators on power quality are currently an important issue. As a whole, wind power integration challenge the power quality, energy planning and power flow controls in the grid. This can be more severe in weak networks, where the whole wind power source may even be disconnected from the grid as an extreme case. In this thesis, we apply wind power prediction, battery energy storage and concepts and ideas from Model Predictive Control (MPC) theory to make wind power more attractive and reliable for power utility companies. The research consists of the following steps. We have developed a wind power prediction model that allows accurate prediction 10 minutes ahead. This is combined with a direction-dependent power curve model, which provides significant improvement in the accuracy of predictions. We have developed a wind power smoothing model, based on a Battery Energy Storage System (BESS), the above prediction model and MPC, to suppress the output power fluctuations of a wind farm, thus allowing smoother operation of the power grid and improved power quality. The concept was further developed to include power frequency regulation, and to optimise BESS capacity. This model supported excellent control of power quality with a reduced BESS requirement by distributing the BESS capacity efficiently. Our control system also ensures more careful regulation of BESS charging/discharging rates to allow longer battery life. It thus reduces the capital cost of the BESS system and extends its operational life, reducing the total capital cost of wind power generation. While our models were developed for wind power and battery storage, they could be applied equally effectively to other generation and storage technologies.
Persistent link to this record
Link to Publisher Version
Link to Open Access Version
Additional Link
Author(s)
Khalid, Muhammad
Supervisor(s)
Savkin, Andrey
Creator(s)
Editor(s)
Translator(s)
Curator(s)
Designer(s)
Arranger(s)
Composer(s)
Recordist(s)
Conference Proceedings Editor(s)
Other Contributor(s)
Corporate/Industry Contributor(s)
Publication Year
2011
Resource Type
Thesis
Degree Type
PhD Doctorate
UNSW Faculty
Files
download whole.pdf 764.8 KB Adobe Portable Document Format
Related dataset(s)