Blade Pitch Control of Floating Offshore Wind Turbine Systems Using Super-Twisting Algorithm and Recurrent RBF Neural Network

This paper presents an approach to enhance the performance of floating offshore wind turbines mounted on semi-submersible platforms through the integration of a Super-Twisting Sliding Mode Collective Blade Pitch Controller (STSM-CBPC) with a Recurrent Radial Basis Function Neural Network (RRBFNN). The proposed CBPC, developed based on a refined nonlinear control-oriented model, leverages the RRBFNN as an adaptive estimator to address lumped uncertainties and external disturbances, when operating above the rated wind speed. The recurrent neural network features a dual feedback loop structure. The internal feedback loop, operating on the hidden layer, and the external feedback loop, enabling the transmission of the output signal back into the input signal, collectively contribute to a comprehensive capture of the system's state information. To ensure convergence, adaptive laws governing the neural network are derived through Lyapunov analysis, ensuring real-time updates to the RRBFNN parameters. Simulation results demonstrate the superior performance of the proposed CBPC over the baseline gain scheduling proportional integral controller for regulating rotor speed and mitigating platform motion.