Improvement of structural health monitoring performance for offshore wind turbines subjected to fatigue and pitting corrosion.

The offshore wind industry is growing fast in the UK. Structural health monitoring has been employed to assess the loads experienced by the structures but also to assess the damage for more intelligent inspection visits, thus reducing cost of maintenance and risk of injuries. Damage can exist in various forms but the two most detrimental ones are fatigue and corrosion. This study delves into the technical aspect of SHM and applies interpolation (longitudinal and circumferential interpolation respectively) techniques to enable data fusion across the structure for fatigue damage assessment from bending strain gauge sensors (on offshore wind turbines). This gives a more refined interpretation of the most damaged locations, which can be used as a guide for targeted inspection rather than the traditional form. Also, it can be used in the design phases for improvement to consolidate the location the highest damage is more prone to. Further analysis has been done to improve the confidence in the readings and to reduce the sampling rate based on damage assessment. Interpolation techniques have also been applied to quantify the damage with respect to pitting corrosion as a form of marine localised corrosion, which is vicious as it can be a prominent initiator for pit to crack transition. The algorithm developed marries for the first time in a data-driven approach pitting corrosion and a data-driven Structural Health Monitoring system. The main transitions from pit initiation to propagation with growth and an interface to capture the pit to crack transition and crack growth using linear elastic fracture mechanics have been developed. To improve the model, a field experiment has been done to express pit characteristics in a statistical fashion with respect to depth, which have been quantified using mass loss techniques, laser scanning and image processing.