An extended ANN-based high speed accurate distance protection algorithm

This paper presents a new neural network based transmission line distance protection module. The proposed module uses samples of voltage and current signals to learn the hidden relationship existing in the input patterns. Simulation studies are preformed and influence of changing system parameters such as fault resistance and source impedance is studied. Details of the design procedure and the results of performance studies with the proposed relay are given in the paper. In addition, an extended ANN-based technique which, uses the proposed neural network distance relay as its basis is described. The extended technique uses neutral current as its new input. Since, this technique adds a new dimension to the input decision space, the accuracy of the algorithm is increased. Various simulation studies are performed and capabilities of the extended algorithm are investigated. Performance studies results show that the proposed algorithm is fast and accurate. Some of the simulation studies are presented in the paper.