D- ANN assisted decentralized frequency restoration schemes in Islanded AC microgrids using secondary control

Microgrids have emerged as a reliable solution for enhancing the resilience and sustainability of modern power systems, particularly in isolated or islanded mode. One of the persistent challenges in such systems is maintaining frequency stability while ensuring fair power sharing among distributed generators. This paper proposes a dual-ANN-assisted decentralized secondary control framework for islanded AC microgrids, where one ANN predicts the optimal droop coefficient and a second simultaneously classifies the best controller configuration - a combination not reported in existing ANN-based or adaptive droop/PID secondary control literature. A novel labeling strategy based on minimum frequency deviation across all distributed generators is introduced for performance-driven ANN training. The core objective is to minimize frequency deviation from the nominal 50 Hz while ensuring proportional power sharing among distributed generators. The optimized parameters predicted by the framework include the droop coefficient (mₚ = 0.2 × 10 - 5 Hz/W·s) and the best-performing secondary controller configuration, identified as the combined integral and proportional correction approach. The ANNs are trained using simulation data obtained from MATLAB/Simulink R2023b, covering multiple combinations of primary and secondary control strategies, each tested with varying droop values. Based on dynamic performance metrics, the ANN predicts the most effective control configuration and corresponding droop coefficient to achieve fast and stable frequency regulation. This prediction is then implemented using a combined approach that integrates the selected configuration with a self-tuning PID controller. The proposed method is validated through extensive simulations and further verified using a real-time hardware-in-the-loop (HIL) setup on an OPAL-RT simulator. Results confirm that the ANN-assisted control system achieves rapid frequency recovery, with in 0.3 seconds,minimal oscillations, and accurate power allocation, demonstrating its effectiveness and real-time applicability for secondary control in islanded microgrids.

Authors

Institutions

Publication Details

Journal
PLoS ONE
Published
2026-11-20
DOI
https://doi.org/10.1371/journal.pone.0362304
Primary Topic
Microgrid Control and Optimization
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

D- ANN assisted decentralized frequency restoration schemes in Islanded AC microgrids using secondary control

Ramakrishna S S Nuvvula, Swathi Tangi, Satish Kumar Injeti, Juhee Sharma et al.
3 links
PLoS ONE
Microgrid Control and Optimization
article

D- ANN assisted decentralized frequency restoration schemes in Islanded AC microgrids using secondary control

Ramakrishna S S Nuvvula, Swathi Tangi, Satish Kumar Injeti, Juhee Sharma, Vijayasanthi Maineni, Polamsetty P. Kumar
article en

Abstract

Microgrids have emerged as a reliable solution for enhancing the resilience and sustainability of modern power systems, particularly in isolated or islanded mode. One of the persistent challenges in such systems is maintaining frequency stability while ensuring fair power sharing among distributed generators. This paper proposes a dual-ANN-assisted decentralized secondary control framework for islanded AC microgrids, where one ANN predicts the optimal droop coefficient and a second simultaneously classifies the best controller configuration - a combination not reported in existing ANN-based or adaptive droop/PID secondary control literature. A novel labeling strategy based on minimum frequency deviation across all distributed generators is introduced for performance-driven ANN training. The core objective is to minimize frequency deviation from the nominal 50 Hz while ensuring proportional power sharing among distributed generators. The optimized parameters predicted by the framework include the droop coefficient (mₚ = 0.2 × 10 - 5 Hz/W·s) and the best-performing secondary controller configuration, identified as the combined integral and proportional correction approach. The ANNs are trained using simulation data obtained from MATLAB/Simulink R2023b, covering multiple combinations of primary and secondary control strategies, each tested with varying droop values. Based on dynamic performance metrics, the ANN predicts the most effective control configuration and corresponding droop coefficient to achieve fast and stable frequency regulation. This prediction is then implemented using a combined approach that integrates the selected configuration with a self-tuning PID controller. The proposed method is validated through extensive simulations and further verified using a real-time hardware-in-the-loop (HIL) setup on an OPAL-RT simulator. Results confirm that the ANN-assisted control system achieves rapid frequency recovery, with in 0.3 seconds,minimal oscillations, and accurate power allocation, demonstrating its effectiveness and real-time applicability for secondary control in islanded microgrids.

PLoS ONEVol. 21(11)
National Institute of Technology Warangal (IN), Manipal Academy of Higher Education (IN), GMR Institute of Technology (IN)
National Research Foundation
Affordable and clean energy, Industry, innovation and infrastructure
Microgrid Control and Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon