Fault location recognition in transmission lines based on Support Vector Machines

It is significant to explore efficient and practical recognition techniques when a severe disturbance due to a short circuit failure in high transmission line. Fault diagnosis has been a major area of investigation among power system and intelligent system applications. This paper use Support Vector Machines (SVM) with strong generalization ability for small samples and fuzzy set theory of being suitable for solving uncertainty linear division relations to perform the recognition task for high voltage transmission line fault types. The simulation results show that the proposed method has the characteristic of simple and clear recognition process, it is able to identify fault types correctly and it is fit for any model structures of a transmission line. Consequently the recognition method for a transmission line fault types based on SVM technique completely overcomes the limitations by using common multilayer perceptions (MLP) classification methods, which achieves results by SVM with the transient data and steady-state simulations for transmission line fault mode spaces, and solves the essential problem for high voltage transmission lines.