Abstract
Islanding detection with the rising grid supporting inverter-based distributed generation is becoming more critical protection due to its high droop gains and overall decreased system inertia leading to rapid changes in the electrical parameters. Traditional methods for islanding detection in this regard are susceptible to significant problems such as non-detection zone, false-positive detection, and inefficient mode of validation. Therefore, to attenuate these problems, this paper proposes a hybrid islanding detection technique based on unsupervised anomaly detection using autoencoders. This technique uses the rate of change of frequency as primary and phase angles of the voltage and current as secondary detection parameters, demonstrating improved performance, reliability, and robustness due to its shared advantage of both active frequency drift and autoencoder. Furthermore, a dialectic model of offline and online validation schemes is also proposed to ensure the reliability of detection. Extensive simulations and validations have been carried out on multiple networks to generate data sets used to train, test, and validate the technique and compute its statistical significance, thereby confirming its effectiveness. The optimal islanding detection time for the base cases was recorded as 20 milliseconds with an F1-score of 0.991, dependability index of 0.998, security index of 0.995, with total harmonic distortion of 4.56% and zero non-detection zones, which complies with IEC 61000-3-2 and IEEE standard 1547’s requirement of detection within two seconds after islanding.
Highlights
‘Renewables-2020’, a report published by International Energy Agency (IEA) forecasts a thirty-three percent share of renewables in total electricity generation by 2025
Active methods are characterized by techniques that usually introduce a small perturbation into the inverter current as an active feature for islanding detection, whereas passive methods do not change the characteristics of the system at all and purely rely on the sensor-based approaches, which may include analog or data-driven techniques
This is due to their complicated methodology of implementation in the inverter control and their already superior selectivity to the non-detection zone (NDZ) problem compared to traditional passive methods
Summary
‘Renewables-2020’, a report published by International Energy Agency (IEA) forecasts a thirty-three percent share of renewables in total electricity generation by 2025. The topological classification for islanding events is split into ’remote’ and ’local’ methods, such that, remote methods are traditionally telecommunication based and are relatively expensive to deploy than a local method These methods include transfer trips [6] and power line carrier [7]. Et al in [8] have presented a satisfactory argument for the ineffectiveness of passive islanding detection methods with the increasing grid supporting DERs. the initiatives to improve active methods using data-driven techniques are rarely seen in the literature. The initiatives to improve active methods using data-driven techniques are rarely seen in the literature This is due to their complicated methodology of implementation in the inverter control and their already superior selectivity to the NDZ problem compared to traditional passive methods
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
