Networked microgrids are gaining importance as distributed energy resources become more prevalent across electrical systems. However, coordinating multiple microgrids to operate economically while maintaining reliability during emergencies remains a significant challenge. Researchers have proposed a hierarchical energy management framework using the pufferfish optimization algorithm (POA) to address this problem.
The system operates in two modes. During normal conditions, each microgrid independently manages its local generation, storage, and loads through a primary controller, minimizing operating costs without requiring power exchange between neighboring systems. Physical tie-lines connect the microgrids, while dedicated cyber-links enable communication. When an emergency occurs—such as generation loss or equipment failure—a higher-level controller activates and coordinates power transfers across interconnections, allowing neighboring microgrids to provide support.
The optimization objective minimizes total operating expenses while accounting for transmission losses across the networked system. The pufferfish algorithm, a nature-inspired metaheuristic, searches for the lowest-cost operation point by simulating collective behaviors observed in pufferfish schools.
Validation testing used MATLAB and DIgSILENT simulation software across a full day of operation. Results compared POA against three established algorithms: particle swarm optimization, genetic algorithms, and grey wolf optimization. Under identical computational conditions, POA achieved the lowest mean operating cost with reduced variability between runs and superior convergence characteristics.
The findings suggest that POA offers practical advantages for real-time energy management in microgrids, particularly during stressed conditions when coordination between systems becomes critical. The algorithm's stability and cost-minimization performance make it potentially valuable for operators managing distributed generation portfolios. Future work may explore implementation in hybrid systems combining renewable resources, energy storage, and demand-side flexibility, as well as validation on larger networks with increased complexity.
