Condition monitoring and control tools for wind energy systems using scada data

Technological developments in wind energy have reduced investment and operating costs. For this reason, wind farms have become more popular around the world. Increasing the share of wind energy in the market has led to the need for easy, inexpensive and effective monitoring and control approaches. In this thesis, various monitoring and control tools are proposed which are cheap and easy to implement in wind farms using existing system data. The first tool is focused on analyzing available data to have a better understand system behavior. Statistical analysis and clustering methods are proposed in this regard. It is necessary to prove that the wind farm operates efficiently and under control. For this reason, the second developed tool is on performance measurements recommended to monitor power generation efficiency. Data Envelopment Analysis, Malmquist Index Approach, and Stochastic Frontier Analysis are proposed to measure power production efficiencies of wind turbines. Forecasting methods have become more important as the wind energy market has increased. Therefore, simple forecasting methods are presented to show the use of available data as the third tool. Also, there is always error in each forecast, so the Particle Filtering approach is suggested to reduce errors. As the last tool, a new training algorithm is developed for multilayer perceptron artificial neural networks which called Antrain ANN. Results are showed that the proposed novel approach can compete with current algorithms.