A novel hybrid Zebra-Archimedes Optimization Algorithm for DG placement and sizing: Single and multi-objective performance analysis

Abstract

This study proposes a novel hybrid Zebra-Archimedes Optimization Algorithm (Z-AOA) for optimal distributed generation (DG) placement and sizing in the IEEE 57-bus system. The Z-AOA merges the strong exploration ability of the Zebra Optimization Algorithm (ZOA) with the precise exploitation capability of the Archimedes Optimization Algorithm (AOA). This hybridization allows the algorithm to search widely for potential solutions while fine-tuning them effectively. The unique strength of Z-AOA lies in its adaptive switching mechanism between exploration and exploitation phases, and ensuring optimal balance throughout the optimization process. This characteristic makes Z-AOA particularly effective for complex, multi-dimensional power system optimization problems. The optimization framework incorporates essential technical constraints, including generator capacity limits, bus voltage magnitude thresholds, and reactive power boundaries. Single-objective variants independently minimize active power loss, voltage deviation, and voltage stability index. Meanwhile, multi-objective variants simultaneously minimize both power loss and voltage deviation using weighted sum method with Pareto front analysis. Simulation results validate the superiority of the proposed Z-AOA, achieving 52.72% power loss reduction and 99.9% voltage regulation compliance. This performance outperforms standalone AOA, which achieves 51.91% reduction and 98.2% compliance. The multi-objective Z-AOA provides balanced trade-offs with 48.54% power loss reduction and 30.20% voltage deviation reduction. Comprehensive comparisons against various categories of metaheuristic algorithms demonstrate the competitive performance of the proposed approach. These categories include swarm-based, physics-based, and evolutionary-based methods. Wilcoxon signed-rank statistical tests further confirm the significant performance advantage of the novel Z-AOA across all optimization objectives.