The sustainable deployment of Horizontal Axis Tidal Turbines will require effective management and maintenance functions. In part, these can be supported by the engineering of suitable condition monitoring systems. The development of such a system is inevitably challenging, particularly given the present limited level of operational data associated with installed turbines during fault onset. To mitigate this limitation, a computational fluid dynamics model is used to simulate the operational response of a turbine under a known set of fault conditions. Turbine rotor imbalance faults were simulated by the introduction of increasing levels of pitch angle offset for a single turbine blade. The effects of these fault cases upon cyclic variations in the torque developed by the turbine rotor were then used to aid creation of a condition monitoring approach. A parametric tidal turbine rotor model was developed based on the outputs of the computational fluid dynamics models. The model was used to facilitate testing of the condition monitoring approach under a variety of more realistic conditions. The condition monitoring approach showed good performance in fault detection and diagnosis for simulations relating to turbulence intensities of up to 2 \%. Finally, the condition monitoring approach was applied to simulations of 10 \% turbulence intensity. Under the 10 \% turbulence intensity case the rotor monitoring approach was successfully demonstrated in its use for fault detection. The paper closes with discussion of the effectiveness of using computational fluid dynamics simulations extended by parametric models to develop condition monitoring systems for horizontal axis tidal turbine applications.
|Journal||international journal of prognostics and health management|
|Publication status||Published - 31 Aug 2018|