Maintenance Optimization Of Wind Turbine Systems Based On Intelligent Prediction Tools

Zhigang Tian,Yi Ding,Fangfang Ding

DOI: https://doi.org/10.1007/978-3-642-20958-1_4

2011-01-01

Abstract:Wind energy is an important source of renewable energy, and reliability is a critical issue for operating wind energy systems. The Canadian wind energy industry has been growing very rapidly. The installed wind energy capacity in Canada in 2008 was approximately 2,000 mega watts (MW), which is less than one percent of the total electricity. It is believed that wind energy will satisfy 20% of Canada's electricity demand by 2025, by adding 55,000MW of new generating capacity [1]. Operation and maintenance costs account for 25-30% of the wind energy generation cost. Currently, the wind turbine manufacturers and operators are gradually changing the maintenance strategy from time-based preventive maintenance to condition based maintenance (CBM) [2-5]. In this article, we review the current research status of maintenance of wind turbine systems, and discuss the applications of artificial neural networks (ANN) based health prediction tools in this field. A CBM method based on ANN health condition prediction is presented. © 2011 Springer-Verlag Berlin Heidelberg.