Hyperspectral imaging applied for the detection of wind turbine blade damage and icing

Abstract

Despite the enhancement in the wind energy sector, the wind turbine industry still faces certain limitations due to some manufacturing and environmental factors. Blades are one of the major components of the wind turbine. During their lifetime, these blades are susceptible to deterioration and normal wear and tear that limit their efficiency and result in higher maintenance costs and longer turbine downtime. Periodic inspections must be performed to detect faults at an early stage and help in mitigating these shortcomings. Many methods were used for this purpose such as: ultrasound, sonic IR, vibration analysis and others. Recent developments have seen a trend of using remote inspection techniques that eliminate the need for human access to the blades. Hyperspectral imaging or imaging spectroscopy is a non-destructive and fast monitoring technique in remote sensing. It is widely used in various classification, and detection fields. In this study, the potential of the use of hyperspectral imaging system in the detection of wind turbine blade damage and icing incident is introduced. Specifically, this study lists the types of damage, its causes, and the techniques used to detect it. Finally, current problems and promising attempts for analyzing real-time turbine blade damage detection are discussed. The results demonstrated that hyperspectral imaging could detect surface and subsurface defects as well as icing events in their early stages of occurrence.