Using meta-heuristic algorithms for controller optimization in automatic voltage regulators

Synchronous generators are basic machines used in electricity generation where mechanical energy is converted into electrical energy. The output voltage of synchronous generators depends on the applied excitation current and the number of revolutions. Since synchronous generators mostly work connected to the grid, they are directly affected by the electrical load dynamics and fault events of the grid. Due to such effects, the output voltage of the synchronous generator may change. If the output voltage of synchronous generators is not kept constant, it will damage the power system and the devices working in this power system, as well as reduce the efficiency of the networks connected to the interconnected system. Since power system networks are interconnected networks, this voltage change, no matter how low, must be detected as soon as possible and the system must be stabilized. For this reason, the voltage must be kept constant by using various control systems such as automatic voltage regulators (OGR). In this thesis study, controller-free, P, PI, PID, PSO and GA based hybrid controller systems were designed in order to regulate the output voltage of the synchronous generator and keep the voltage level of the network and the devices operating in the network constant, and the output voltage of the synchronous generator was controlled and the results were compared. It has been observed that using meta-heuristic algorithms to optimize the gain parameters of an OGR with a PID controller used to adjust the output voltage of a synchronous generator increases the voltage regulation performance of the system. Comparison criteria are maximum overshoot, rise time and settling time. In the comparison results, transient response analysis, ground root analysis and frequency response analysis were evaluated and presented.