Developing Hybrid Optimization Protocols to Reduce the Gap Between ‎Theoretical Solutions and Practical Application in Complex Electrical ‎Systems

Abstract

Due to the inherent nonlinearity and transient dynamics of modern electrical power systems, a significant gap has emerged between idealized optimization models and the practical reality of grid operations. To bridge this gap, this study proposes a dynamic hybrid optimization protocol that synergistically combines the global exploration capabilities of metaheuristic algorithms with the precise local exploitation of deterministic mathematical solvers. The algorithm dynamically alternates between a stochastic search phase and a gradient-based refining phase using a variance-based transition mechanism. To validate scalability and robustness, the protocol is tested on both the standard IEEE 30-bus and the large-scale IEEE 118-bus test systems. Unlike standalone metaheuristics, which often leave residual power imbalances (up to 1.5 MW), the proposed hybrid protocol achieves exact physical feasibility (0.0000 MW mismatch) without violating equality constraints. Furthermore, statistical analysis conducted over 30 independent runs, validated by the Wilcoxon rank-sum test, demonstrates superior consistency, computational efficiency, and repeatability compared to state-of-the-art baselines. The protocol ensures strict adherence to voltage stability (0.95–1.05 p.u.) and thermal limits. These findings confirm that hybridizing metaheuristics with deterministic solvers translates theoretical optima into safe, economical, and physically executable real-world electrical dispatch commands.