Undiagnosed surface damage to wind turbine blades can lead to repair costs in the hundreds of thousands of pounds. This makes early detection critical, both in terms of reduced downtime and avoiding catastrophic failures. With increased use of drones, high-resolution images of wind turbine blades are now routinely captured for expert analysis. This project aims to automate that analysis using machine learning, reducing inspection costs and improving accuracy.
Using publicly available datasets of wind turbine surface damage, the project aims to develop an AI-based computer vision model to detect various types of damage, including cracks, erosion, and material degradation. By integrating AI with drone imaging, the project enables faster, more precise identification of issues, supporting the maintenance and operational efficiency of offshore wind turbines.
