Islanding detection scheme based on adaptive identifier signal estimation method

Area of Science:

• Electrical Engineering
• Power Systems
• Renewable Energy Integration

Background:

• Islanding in distributed generation (DG) poses a significant safety risk.
• Existing passive anti-islanding methods struggle with near-zero active/reactive power mismatches.
• Reliable islanding detection is crucial for grid stability and protection.

Purpose of the Study:

• To propose a novel passive anti-islanding method for inverter and synchronous machine-based DG units.
• To address the limitations of current methods in detecting islanding under minimal power imbalances.
• To enhance the reliability and accuracy of islanding detection systems.

Main Methods:

• Development of a new islanding detection method based on exponentially damped signal estimation.
• Utilizing an adaptive identifier to estimate frequency deviation at the point of common coupling (PCC).
• Introduction of a 'decision signal' derived from PCC frequency oscillation frequency for detection.

Main Results:

• The proposed method effectively detects islanding conditions even with near-zero active power imbalance.
• The adaptive identifier's small sampling window offers an advantage over other signal estimation techniques.
• The 'decision signal' simplifies threshold setting due to predictable oscillation frequency behavior during islanding.

Conclusions:

• The novel passive anti-islanding method demonstrates high capability in detecting islanding events, particularly under challenging near-zero power mismatch conditions.
• The adaptive identifier-based approach offers a robust and efficient solution for DG protection.
• The proposed method is validated against various non-islanding transient events, confirming its reliability.