Artificial neural network–aided technique for low voltage ride-through wind turbines for controlling the dynamic behavior under different load conditions

Abstract

At the level of the electrical distribution networks with wind generation, the disturbances may influence the voltage stability, particularly during low voltage ride-through wind turbine. This research is concerned with studying the effect of implementing different controllers and load types on the low voltage ride-through dynamic recovery performance during disturbances. A conventional proportional–integral–derivative controller is compared with the artificial neural network–based one. The controller construction and its gain are proposed for each type of controller and the impact of each controller on the dynamic behavior of the low voltage ride-through is investigated thoroughly under various operating conditions. Also, the dynamic performance of wind generators is examined with low voltage ride through and different dynamic load models. Both, dynamic induction motor load and composed static and exponential recovery load models are considered. In case of dynamic induction motor load, the effect of the inertia constant has been studied under two types of controllers. The overall system model is simulated using PSAT/MATLAB software in such a way that it can be suited for modeling of voltage controller and loads configurations. The low voltage ride-through performance has been changed with different controllers and load types.