Objective:
Development of a drone based NDT inspection tool for the Wind Power Industry with highly commercial potential for multiple inspection tasks. Specific goals of the project are solutions for:
- Intelligent detection and classification of damages by images in HR and computer vision
- Damage Map & Classification system,
- Algorithms for classification system and geometric documentation of wind turbine blades surface in 3D,
- CNN machine learning system,
- Integration & programming of all system parts in a cloud/SaaS based Wind Module,
- Automatic Reporting,
- Advanced sensors testing for detection of structural damages with no or sparse visual footprint on the surface
- Probabilistic tool to predict imminent failures.
Expected outcome:
Users will initially be Danish Inspection Companies who will benefit from an initial free or low cost subscriber arrangement in exchange for inspection data sharing to enhance the systems detection and classification skills. Users will also be Wind Farm owners or Manufactures who either perform inspections on their own or wish to conduct product tracking. The project will provide better understanding of damages resulting from operation of real wind turbines and what causes such damages to grow. This understanding is essential in order to improve the load prediction and design tools as well as the test standards used to design and certify wind turbine blade. Improved reliability of wind turbine blades reduces O&M cost, claims and turbine down time and thus the cost of energy (CoE).
This project began January 2017 and runs for 3 years.