MEMS Multisensor Intelligent Damage Detection for Wind Turbines

Abstract

Maintenance and repair of wind turbine structures have become more challenging and at the same time essential as they evolve into larger dimensions or located in places with limited access. Even small structural damages may invoke catastrophic detriment to the integrity of the system. So, cost-effective, predictive, and reliable structural health monitoring (SHM) system has been always desirable for wind turbines. A real-time nondestructive SHM technique based on multisensor data fusion is proposed in this paper. The objective is to critically analyze and evaluate the feasibility of the proposed technique to identify and localize damages in wind turbine blades. The structural properties of the turbine blade before and after damage are investigated through different sets of finite-element method simulations. Based on the obtained results, it is shown that information from smart sensors, measuring strains, and vibrations data, distributed over the turbine blades can be used to assist in more accurate damage detection and overall understanding of the health condition of blades. Data fusion technique is proposed to combine these two diagnostic tools to improve the detection system that provides a more robust reading with reduced false alarms.