AI empowered wind turbine health self-diagnosing system using sensor network.

Research proposals

  • Research area

    Big marine data and metocean

  • Institution

    University of Hull

  • Research project

    AI empowered wind turbine health self-diagnosing system using sensor network.

  • Lead supervisor

    Dr Yongqiang Cheng (Faculty of Science and Engineering, University of Hull)

  • Supervisory Team

    Dr Qin Qin (Lecturer in Acoustics, University of Hull)
    Dr Nina Dethlefs (Lecturer – Computer Science, University of Hull)
    Dr Yongqiang Cheng (Faculty of Science and Engineering, University of Hull)
    Dr Prosanta Gope (Computer Science, University of Hull)

Project Description:

The blades of wind turbine are in various lengths, materials, shapes and sizes, even their natural frequencies are not constant. Precise modelling of the wind turbine including their blades reflecting their internal structures changes, damages, imperfections of movement proves to be challenging. In this project, we steering away from the conventional methods of wind turbine health monitoring, but to study the feasibility of a novel low cost data driven response oriented (DDRO) method for real time wind turbine health reporting. In DDRO, a customised sensor emitting a selected range of waveforms propagating through the material of the wind turbine body. They will demonstrate various resonant/vibrating features, detected from a number of sensors that are mounted in far end of the structures and blades. In the meanwhile, periodical laser beams are deployed on the sensors to align them in precise positions and also detect deformation of the structure. Empowered by the latest big data analysis algorithms, the signature of each wind turbine will be learnt through fusion of these data. By consistently monitoring these signatures, any deformation, damages, defects of the blades will be immediately revealed in real time and possibility their degree of severity.

Download all research proposals here