Fuzzy model predictive MPPT control of interconnected wind turbines drivetrain
Mohsen Farbood, Mokhtar Shasadeghi, Afshin Izadian, Taher Niknam
Abstract
Abstract This paper provides a novel fuzzy‐based model predictive control (MPC) of a hydraulic‐based drivetrain that can combine the harvested energy of multiple wind turbines into one generator. The modeling is based on the sector nonlinearity approach and Takagi–Sugeno (T‐S) fuzzy. In addition, prediction and quadratic programming (QP) are formulated to obtain suitable control commands for the displacement of hydraulic pumps. Input–output linearization is utilized to control the generator torque load to regulate the speed. Minimizing the cost function for T‐S fuzzy‐based interconnected hydraulic wind turbine systems is converted to a QP problem to reduce the computation complexity. Therefore, the proposed nonlinear fuzzy‐based MPC scheme is more realistic to be implemented in real time. The feasibility of the optimization problem is improved when a new ellipsoidal terminal constraint is considered in the controller design process. Offline linear matrix inequalities (LMIs) are utilized to obtain the terminal cost of T‐S fuzzy‐based nonlinear systems. The simulation results demonstrate a significant reduction in computations and control efforts compared to LMI‐based MPC, kω 2 approach, and (proportional–integral) PI controllers. It is also demonstrated that the system response to a sudden wind speed variation contains less overshoot and settles in a shorter time.