Hybrid multi-objective optimization of µ-synthesis robust controller for frequency regulation in isolated microgrids

Area of Science:

• Electrical Engineering
• Control Systems
• Renewable Energy Systems

Background:

• Isolated microgrids face significant challenges in frequency regulation due to unpredictable system uncertainties and fluctuating load demands.
• Existing control strategies often struggle to maintain stability and performance under these dynamic conditions.

Purpose of the Study:

• To develop and evaluate an optimal µ-synthesis robust control strategy for enhanced frequency regulation in isolated microgrids.
• To improve system performance, stability, and robustness against uncertainties using a novel fixed-structure weight selection approach.

Main Methods:

• A fixed-structure approach for performance and robustness weight selection was developed, informed by subsystem frequency analysis.
• The µ-synthesis robust controller was optimized using Multi-Objective Particle Swarm Optimization (MOPSO) and Multi-Objective Genetic Algorithm (MOGA) under inequality constraints.
• Pareto front analysis was employed to identify optimal controller solutions.

Main Results:

• The MOPSO-optimized controller demonstrated superior robustness, tolerating up to 236% uncertainty, compared to 171% for conventional µ-synthesis controllers.
• The proposed controller significantly reduced frequency deviation and improved transient response.
• Nyquist stability analysis confirmed the controller's robustness against uncertainties from renewable energy sources.

Conclusions:

• The proposed optimal µ-synthesis robust control strategy is highly effective for frequency regulation in isolated microgrids.
• The controller enhances system stability and performance while maintaining robustness against significant uncertainties.
• Future research will focus on discrete-time implementation for practical digital signal processing (DSP) applications.