Robust transmission line fault classification using wavelet multi-resolution analysis

Highlight
• Power system model for transmission-line fault classification fed from both ends.
• Fault classification algorithm for doubly fed EHV transmission line using wavelet MRA.
• The synchronized fault currents passed through wavelet MRA technique for determining the sixth level detailed coefficients.
• Sag1, Sbg1, Scg1, Sag2, Sbg2 and Scg2 are used to classify the type of fault occurred in the transmission line.

With the advent of high speed communication technology, global positioning system (GPS) and artificial intelligence (AI) techniques, there has been a paradigm shift in the parlance of power grid operation and control. The power system is in a phase of transition towards smart grid with the aid of these techniques to combat against contingencies, leading to reduction of failures and blackouts. The transmission lines are considered to be the back bone of the grid and traverse over difficult terrains. With the advancements in digital relaying and wide-area protection along with GPS technology, philosophy of protection has also undergone a paradigm change to take care of such challenges. This paper explores the possibility of transmission line protection for multi-generator system using wavelet Multi-Resolution Analysis (MRA) technique along with GPS. The inputs for the wavelet transform are the synchronized currents measured by remote telemetry units (RTUs) in conjunction with GPS technology at different buses. The classification algorithm uses wavelet MRA technique to extract features of the transient current signals based on harmonics generated due to abrupt change of currents in a three-phase transmission line caused by different faults. The major contribution of this paper is that the classification algorithm is immune to the effects of fault inception angle, fault impedance, fault distance and power angle. The results validate the efficacy of the proposed algorithm for real time smart grid operation.

Figure optionsDownload as PowerPoint slide