Artificial Neural Network Based Automatic Voltage Regulator for a Stand-Alone Synchronous Generator

In this study, an automatic voltage regulator (AVR) based on artificial neural network (ANN) is presented for a standalone synchronous generator. The AVR maintains the constant output voltage independent of load by altering its excitation current and brings the output voltage to the desired level. Generally, AVR systems are designed by using proportional-integral-derivative (PID) controller. In this paper ANN based AVR system is proposed to get the constant output voltage. An experimental setup was prepared to acquire data to be used for training ANN. The experimental setup consists of synchronous generator, asynchronous motor as actuation machine, frequency inverter and load groups. ANN is trained with the experimental data by using 3 different learning methods defined on MATLAB. Then, the ANN, which is trained by each learning method respectively, is tested separately on MATLAB and the results are compared.