| PhD Seminar

Name of the Speaker: Mr. Anoop V Eluvathingal (EE14D001)
Guide: Prof. Swarup K S
Online meeting link: https://meet.google.com/gkm-drth-ysw
Date/Time: 25th November 2024 (Monday), 2:30 PM
Title: Enhanced Protection Strategies for Active Micro Grid Distribution Networks

Abstract :

Protection of Distribution Networks in the presence of Renewable Energy Sources (RES) or Distributed Generation (DG). A key challenge in such networks is coordinating line protection devices (fuses and reclosers) and DG unit's protection, during fault events. Traditional model-based protection algorithms may struggle to handle these scenarios, especially when DG units continue to operate under Low-Voltage Ride-Through (LVRT) requirements. Microgrid protection has become a topic of great interest because of its unique requirements. In addition to the typical characteristics of conventional power system protection schemes, namely sensitivity, selectivity, and speed, microgrid protection should also adapt to changing conditions (grid-connected or islanded), making microgrid protection designs unique. A comprehensive protection strategy with two different novel intelligent circuit breakers and their unique modular architecture for the microgrid is proposed in this research work to address various protection issues.

This work explores the application of AI and ML to enhance protection, particularly in distribution networks with Distributed Generation (DG) resources. We propose an AI-based Enhanced Voltage Protection Module (AI-VPM) for DG interconnection protection relays (IPRs) and feeder protection reclosers (FPR) to address this issue. The AI-VPM leverages a Space Phasor Model (SPM) of three phase voltages to detect faults in the host distribution network and coordinate with recloser operations. By overriding the LVRT requirement under specific fault conditions, the AI-VPM can accelerate fault clearance and improve system reliability. We evaluated the performance of the proposed EVPM through extensive simulations of various network configurations. The results demonstrate the effectiveness of the AI-based approach in enhancing protection coordination and minimizing disturbance propagation