Yaw Predictive Control Strategy of Offshore Wind Turbines with Attention-CNN-LSTM Wind Predictive Model

To improve the yaw efficiency of offshore wind turbines, this paper proposes a yaw predictive control strategy based on an Attention-CNN-LSTM wind direction forecasting model. Self-Attention mechanism is introduced to fully leverage the advantages of convolutional neural networks (CNN) and long short-term memory networks (LSTM) in time series prediction. Using the predicted wind direction, the yaw control strategy is re-designed to avoid frequent yaw action. Simulation results indicate that compared to the traditional yaw control strategy, the proposed yaw predictive strategy could effectively reduce the yaw errors and increase the maximum wind energy capture capability.