Quantum-Behaved Particle Swarm Optimization-Tuned PI Controller of a SEPIC Converter

The Single-Ended Primary Inductor Converter (SEPIC) is vital for voltage regulation in dynamic systems like renewable energy and electric vehicles. Traditional PI controllers struggle with tuning complexity and oscillations. This study introduces Quantum-Behaved Particle Swarm Optimization (QPSO) to optimize PI gains (Kp, Ki) for SEPIC converters. QPSO improves global search by using quantum-inspired probabilistic motion, overcoming issues of premature convergence seen in traditional PSO. Four objective functions—ISE, ITAE, IAE, and MSE—were evaluated to balance transient and steady-state performance. ITAE and IAE outperformed others, minimizing overshoot to 1.26% in boost mode and achieving the fastest settling time of 1,872 s. Sensitivity analysis revealed that Ki > 2.0 destabilizes the system, while Kp > 1.5 increases voltage ripples. The framework is computationally efficient, ideal for embedded applications. Future work should include hardware-in-loop testing to confirm robustness.