End-to-End Classification of Transmission Lines in Fault Detection Using SW and Extreme Learning Based Approach

The reliability of the power grid is dependent on transient stability. In order to swiftly restore transient stability, large disturbances, such a fault in a transmission line, must be separated. Faults in a transmission network can be located, identified, and categorized with the use of erroneous current and voltage signals. When an aberrant signal is detected, the relay immediately disables the transmission line carrying the bad news. Fault detection systems are crucial for transmission lines. The proposed technique for preprocessing makes use of a solid-state drive (SSD) detection model, a quicker RCNN detection model, and a logical reasoning in parallel mechanism. To choose features, SWELM is employed. The envisioned model is trained using SWELM in the given technique. The suggested model is tested, and the ELM and SWELM models are used for comparison. When compared to the other two varieties, its performance is far superior.