Adaptive fuzzy-recurrent neural network tuned fractional-order distributed control for robust frequency regulation in...

Area of Science:

• Electrical Engineering
• Control Systems
• Renewable Energy Integration

Background:

• Conventional frequency control methods face challenges in scalability and disturbance resilience within multi-microgrid systems (MMGS) with high renewable energy penetration.
• Effective frequency regulation is crucial for the stability and reliability of modern power grids, especially with increasing integration of intermittent renewable sources.

Purpose of the Study:

• To develop and validate an advanced, adaptive, and distributed frequency control solution for MMGS.
• To improve transient and steady-state performance of frequency regulation under various disturbances.
• To demonstrate the superiority of the proposed control strategy over conventional methods.

Main Methods:

• Implementation of a Distributed Consensus Control Strategy (DCS) utilizing a Fractional Order PID (FOPID) controller.
• Adaptive tuning of the FOPID controller parameters using a Fuzzy-Recurrent Neural Network (FRNN).
• Extensive real-time hardware-in-the-loop (HIL) testing on a three-microgrid platform.

Main Results:

• The proposed FRNN-tuned FOPID-based DCS significantly reduced settling time (e.g., 29% in MG1), peak overshoot (over 90% in MG1), and absolute error (over 80% in MG1) compared to PID controllers.
• Validation of performance improvements across multiple microgrids (MG1, MG2, MG3).
• Demonstrated robustness and resilience under diverse disturbance scenarios through comprehensive testing.

Conclusions:

• The FRNN-tuned FOPID-based DCS represents a novel adaptive, distributed framework for real-time self-optimization in MMGS frequency regulation.
• The proposed method offers exceptional resilience and improved performance, making it suitable for future smart grids and cyber-physical energy systems.