Hybrid Optimization Using Particle Swarm Optimization and Gradient Descent to Improve Phase Locked Loop Performance in Grid Connected Systems

This paper presents a novel Hybrid Particle Swarm Optimization and Gradient Descent (PSO-GD) based Phase-Locked Loop (PLL) for enhanced grid synchronization in power inverter systems. Unlike conventional fixed-parameter PLLs, the proposed approach adaptively optimizes controller gains through a unique combination of PSO's global search capabilities and GD's local refinement precision. Comprehensive performance comparisons with traditional methods including SRFPLL, DDSRF-PLL, and MSOGI-PLL demonstrate the superior capabilities of the hybrid approach under various grid disturbances. When subjected to 30% voltage sags and 30° phase jumps, the Hybrid PSO-GD PLL maintains frequency deviations within <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\pm 0.3 \text{Hz}$</tex> and phase tracking errors below 3 degrees, while achieving a 40% reduction in settling time compared to conventional methods. The enhanced stability margins (65° phase margin, 12 dB gain margin) and improved bandwidth <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$(120 \text{Hz})$</tex> ensure robust operation during grid disturbances, while maintaining THD below 1.7%. Detailed experimental results verify the effectiveness of this adaptive approach, particularly for modern grids with high renewable energy penetration where frequent disturbances demand real-time parameter optimization. The proposed method offers a promising solution for improving grid synchronization reliability and power quality in renewable energy integration applications.