Event-triggered neural network prescribed performance control for wave energy conversion system under input saturation

Abstract

To solve the Maximum power point tracking (MPPT) control problem in wave energy conversion systems (WECS) under input saturation, an event-triggered neural network prescribed performance controller is designed. A structure of a direct-drive WECS with internal parameter changes and external disturbances is built, where these changes and disturbances are treated as lumped uncertainties. The auxiliary system is designed to solve the input saturation. The controller parameters are dynamically adjusted by an event-triggered mechanism to constrain control inputs and save communication resources. The radial basis function neural networks (RBFNN) are employed to approximate model uncertainties and disturbances, enhancing the robustness of the system. An asymmetric prescribed performance function is employed to constrain the state of the system within a prescribed range, ensuring the boundedness of the closed-loop stochastic nonlinear system. Simulation results show that the proposed method successfully realizes MPPT in the WECS under input saturation, internal parameter changes, and external disturbances.