Improving Synchronous Generator's differential protection with the use of Artificial Neural Networks

This paper presents an alternative technique to detect and correct waveforms distortion due to Current Transformer (CT) saturation using intelligent tools based on Artificial Neural Network (ANN). The Real-time Digital Simulator (RTDS) was used to evaluate the performance of the ANNs running in a real-time embedded system. It is also shown that the CT saturation influences the Synchronous Generator (SG) differential protection.