Artificial Intelligence Integrated Control to PMSG Built Wind Energy Conversion Scheme

The performance of a wind energy conversion system (WECS) is unsatisfactory under variable wind speed conditions. By employing general control approaches, the sensitivity of the dc link voltage has a significance on the overall performance of the system. The time delay and model parameter uncertainty, transient disturbances such as maximum overshoot, rising time, and settling time are disproportionately larger than the rest of the system. To overcome these drawbacks in WECS, an artificial intelligence (AI)-integrated control strategy for a permanent magnet synchronous generator (PMSG)-attributed wind energy conversion system. Adaptive neuro-fuzzy inference system (ANFIS) controller based MPPT is employed for modulating the generator side converter at all situations, including ideal conditions, system uncertainties, and disturbances, are taken into account to achieve the maximum possible power output. ANFIS control is employed in conjunction with an excitation signal to provide adaptive adjustment of the discontinuous control gain, to reduce chattering effect in WECS. Utilizing unstable nonlinear observers with a nonlinear response are used to monitor both aerodynamic torque and stator currents. The optimal speed reference is determined to calculate the wind speed with enhanced control, less fluctuations, and fewer tracking errors in steady state. This result in excellent tracking performance when dealing with rapidly varying aerodynamic torque.