Real-time out-of-step protection for TNB system

Maintaining dynamic security of the power system subject to large disturbances is one of serious concerns of power utilities nowadays. Besides detection of the instability event, fast determination of emergency control is also very important. This paper presents an application of artificial neural network (ANN) to find the automatic generator tripping at the eastern corridor of the peninsular Malaysian power system. The database processed by the ANN is simulated based on real measured data. The input features are selected by engineering judgments on the set of readily available data and no dimensionality reduction was used. The results obtained are encouraging. Thereby the potential integration of ANN to the real-time application platform (RTAP) is being considered.