Vibration Transmissibility as a Method of Damage Detection on Horizontal Axis Wind Turbine Blades

Abstract

Structural health monitoring of horizontal axis wind turbine blades is challenging due to their large size and limited in-service accessibility. Given the harsh environment that wind turbines operate in, the blades are prone to damage throughout their working life. This paper examines the transmissibility of vibration response among embedded sensors over time as a method of damage characterization. After damage occurs, the sensors closest to the new defect will experience the highest change in transmissibility compared to baseline readings. The defect can therefore be identified and located with this method. When assessed at the natural frequencies of the blade, the response transmissibility between two sensors is independent of applied force and magnitude. This makes vibration transmissibility an ideal method for inspecting wind turbine blades that are subject to varying wind speed and direction. A blade was designed based on the NREL IEA Wind -15MW offshore reference wind turbine, with modifications made to fabricate a scaled 3D printed model for testing. The blade was outfitted with MEMs accelerometers to measure acceleration and fiber Bragg gratings (FBGs) to measure strain. To test the transmissibility concept prior to experimental testing, a finite element model was developed to simulate acceleration and strain transmissibility on a damaged and undamaged blade with random force inputs. This model was able to demonstrate that a 5mm transverse crack was visible across the blade when examined at the natural frequencies.