Enhancing the Reliability of Structural Health Monitoring for Bolted Joint Connections in Segmented Rotor Blades Using Data Fusion

Abstract

In the wind energy sector, the use of segmented rotor blades poses challenges for Structural Health Monitoring (SHM) systems due to the increasing high loads that can damage bolted joints. This paper investigates the potential of data fusion to improve the reliability of SHM systems for bolt connections in segmented rotor blades. Three CFRP structures with bolted joints are examined and monitored using both piezoelectric and electrical strain gauge sensors. A passive low-frequency monitoring system is built using strain measurements and neural networks to model the relations between measurement data. An additional active high-frequency monitoring system is designed using piezoelectric sensors and guided waves in combination with a subsequent principal components analysis. The results show that the data fusion system successfully detected the damages with an accurate damage occurrence probability prediction even if one monitoring system provided unreliable predictions. By combining two independent monitoring systems that complementarily cover different frequency bands, the data fusion system improved the reliability of the monitoring task and reduced the occurrence of false positive alarms. The approach presented in this study can also be applied to other monitoring systems in various industries, further expanding the impact of the research.