Estimating Parallel Transmission Line Fault Using Phasor Measurement Unit based Artificial Neural Network

Abstract

In a parallel transmission line, fault line, fault location, and classification have been identified separately. Since fault location takes more calculation time, it is unfit for protection purposes. Thus, this paper presented a new scheme that estimates the faulted line, fault location, and type of fault in a parallel transmission line, with the help of Phasor Measurement Units (PMU) and the Artificial Intelligence Technique. The proposed scheme uses phasors of Positive Sequence Voltage (PSV) and Positive Sequence Current (PSC) to detect the faulted line in a parallel transmission line. Further, the Artificial Neural Network (ANN) models have been designed to estimate the fault distance on a faulted line and classify the fault types. The PSV and PSC obtained from PMUs are selected as inputs because they have a negligible mutual coupling effect on the parallel transmission lines. The IEEE 9 bus system and the IEEE 30 bus system have been considered test cases to validate the proposed scheme. The proposed scheme is also validated by hardware in the loop on an OPAL-RT real-time simulator (OP RTS 5700). The results show that the proposed scheme identifies the fault line, fault distance, and type of fault regardless of its location on a parallel transmission line. Besides this, the proposed scheme has a quick response time, making it suitable for wide-area backup protection applications.