Using Neural Networks to Estimate the Load Angle of Synchronous Generator, Applying Increased Input Torque Error

The main idea of this paper is to use Artificial Neural Networks (ANN) to create a model of synchronous generator with water turbine. In this paper, a characteristic of generator torque and speed with the highest number of learning data is given to a feed-forward neural network, and the network estimates one of the fastest learning methods of load angle (δ). The more is the number of learning data, the more sensitive are the small errors for us. Then, using the neural network toolbox, we investigated the generator stability with an error in increasing the input torque to more than the nominal amount. The neural network found the error easily. The simulation results confirm the proposed method.