• International Journal of Naval Architecture and Ocean Engineering
• /
• v.10 no.1
• /
• pp.9-20
• /
• 2018

Due to integrated stochastic wind and wave loads, the supporting platform of a Floating Offshore Wind Turbine (FOWT) has to bear six Degrees of Freedom (DOF) motion, which makes the random cyclic loads acting on the structural components, for instance the tower base, more complicated than those on bottom-fixed or land-based wind turbines. These cyclic loads may cause unexpected fatigue damages on a FOWT. This paper presents a study on short-term fatigue damage at the tower base of a 5 MW FOWT with a spar-type platform. Fully coupled time-domain simulations code FAST is used and realistic environment conditions are considered to obtain the loads and structural stresses at the tower base. Then the cumulative fatigue damage is calculated based on rainflow counting method and Miner's rule. Moreover, the effects of the simulation length, the wind-wave misalignment, the wind-only condition and the wave-only condition on the fatigue damage are investigated. It is found that the wind and wave induced loads affect the tower base's axial stress separately and in a decoupled way, and the wave-induced fatigue damage is greater than that induced by the wind loads. Under the environment conditions with rated wind speed, the tower base experiences the highest fatigue damage when the joint probability of the wind and wave is included in the calculation. Moreover, it is also found that 1 h simulation length is sufficient to give an appropriate fatigue damage estimated life for FOWT.

• Journal of Ocean Engineering and Technology
• /
• v.35 no.4
• /
• pp.287-295
• /
• 2021

This paper describes a recurrent neural network (RNN) for the fault classification of a blade pitch system of a spar-type floating wind turbine. An artificial neural network (ANN) can effectively recognize multiple faults of a system and build a training model with training data for decision-making. The ANN comprises an encoder and a decoder. The encoder uses a gated recurrent unit, which is a recurrent neural network, for dimensionality reduction of the input data. The decoder uses a multilayer perceptron (MLP) for diagnosis decision-making. To create data, we use a wind turbine simulator that enables fully coupled nonlinear time-domain numerical simulations of offshore wind turbines considering six fault types including biases and fixed outputs in pitch sensors and excessive friction, slit lock, incorrect voltage, and short circuits in actuators. The input data are time-series data collected by two sensors and two control inputs under the condition that of one fault of the six types occurs. A gated recurrent unit (GRU) that is one of the RNNs classifies the suggested faults of the blade pitch system. The performance of fault classification based on the gate recurrent unit is evaluated by a test procedure, and the results indicate that the proposed scheme works effectively. The proposed ANN shows a 1.4% improvement in its performance compared to an MLP-based approach.