Neural network based maximum power point tracking and control of PMSG wind syste

An artificial neural network (ANN) based maximum power point tracking (MPPT) algorithm has been investigated. The results obtained have been compared with an adaptive neuro-fuzzy inference system (ANFIS). Both ANN-based and ANFIS based MPPT controllers have the ability to estimate wind speed and to track the maximum power point (MPP) and the optimal rotor speed with very low error as compared to the conventional MPPT methods. Moreover, these methods demonstrate remarkable performance under rapidly changing wind conditions in estimating wind speed, tracking MPP accurately and suppressing undesired oscillations around maximum power point. The algorithm is based on two series neural networks, one for wind speed estimation and the other for tracking maximum power point (MPP). The algorithm does not require any mechanical sensor for wind speed measurement. Nonlinear time domain simulations have been carried out to validate the effectiveness of the proposed controllers in terms of wind speed estimation and MPPT under different operating conditions.;The obtained results demonstrate that both the proposed ANN and ANFIS-based MPPT controller has better dynamic and steady state performance than the conventional methods and the obtained results also demonstrate that ANFIS based controller is better than ANN based controller. Accuracy in wind speed estimation and maximum power point tracking has been used as the performance criterion for evaluating MPPT controllers.;The performance of the ANFIS based MPPT controller is investigated using MATLAB simulation for a grid connected permanent magnet synchronous generator (PMSG) wind system represented through a detailed dynamic model of the generator, the generator turbine, drive train and the converters. Simulation results confirm that the wind turbine system can deliver power to the grid maintaining the optimum value of power coefficient (Cp) for rapidly changing wind conditions.