Stability enhancement of power systems connected to combined wind farms using STATCOM and neural networks

In recent years, The use of renewable energy has grown exponentially due to its many benefits, including no need for fuel and no harm to the environment. However, the use of renewable energy sources in power systems has caused new problems for these systems. In this regard, the study of the dynamic behavior of power systems, including farms or wind turbines, has been the focus of many researchers. In wind farms, one of the types of squirrel cage induction generators (SCIG) or double-fed induction generators (DFIG) is generally used. But both types of induction generators are used in a combined wind farm (CWF). But it should be noted that in order to improve the dynamic behavior of combined wind farms, it is necessary to use compensators in the system. STATCOM (static synchronous compensator) is widely used among the types of compensators. A STATCOM can quickly control the bus voltage connected to it by injecting or absorbing reactive power into the system. In this article, first, a power system connected to a hybrid wind farm is selected, in which a STATCOM is used in one bus. Then, a supplementary PID controller is used in the STATCOM structure to effectively dampen the fluctuations of the system variables in the face of errors. Also, the heuristic particle swarm optimization (PSO) method is used to determine the optimal values of the coefficients of the PID controller. But since determining the optimal coefficients of the PID controller using the PSO algorithm is a time-consuming task, an artificial neural network (ANN) is also used to estimate the controller parameters in time when the working conditions of the system change. The results of the performed simulations are presented and the correctness of the proposed method is confirmed. All simulations are done in MATLAB/Simulink software.