Adaptive optimal disturbance rejection for wave energy converters

This research aims to mitigate disturbances affecting Wave Energy Converters (WECs) using an adaptive optimal disturbance rejection framework by dynamically adjusting control actions based on forecasted wave conditions. A Nonlinear Autoregressive (NAR) Neural Network is utilized for forecasting wave elevations and generating optimal reference velocities for the considered case study single-body heaving point absorber. The wave excitation force is considered as the external disturbance source affecting the WEC. Frequency and time domain response analysis are conducted to understand system behavior, followed by considering the real wave climate of two different selected locations around Finland, crucial for performance evaluation. The efficacy of the proposed approach is evaluated through a comprehensive results analysis. This includes evaluating its effectiveness on the selected sea states and its adaptability concerning variations in WEC dynamics. In all the investigated scenarios, the proposed control strategy can track the displacement and velocity reference signals with high accuracy in the presence of disturbance with proper initializing of the weight matrices, highlighting the potential of the proposed methodology in improving the efficiency and reliability of WECs under varying wave conditions. • A new adaptive optimal disturbance rejection framework is proposed to enhance the WEC performance. • Optimal WEC reference velocity generation by utilization of NAR Neural Network. • Comprehensive evaluation on real wave climate data from Finland. • Tracking the displacement and velocity reference signals with high accuracy.