An internal fault detector and classifier for synchronous generator stator windings based on ANN have been implemented and its behavior is investigated on physical power system model. The hardware system is designed and built to acquire the three phase currents at both ends of synchronous generator terminals. A software program is developed to read currents, which used to train a proposed Neural Network structure using MATLAB. The trained network is placed in a LabVIEWTM based program formula node that monitors the currents online and display the fault types. Details of implementation and the experimental studies are given and analyzed in the paper. Lab work proves that the proposed approach is able to detect and classify the type of internal faults rapidly and correctly. It is suitable to realize a fast and accurate internal fault protection scheme of the synchronous generator.