Multi-objective PI Controller Optimization for Delayed Battery-Supported DC Bus Systems Using the Stability Boundary Locus Method
摘要
This paper proposes a Stability Boundary Locus (SBL)–based multi-scenario PI controller design framework for battery-supported DC bus systems with second-order RC dynamics and communication-induced time delay. Within the analytically derived SBL stability region, a multi-objective optimization problem is formulated to tune PI controller gains under three distinct performance priorities: disturbance rejection, control effort limitation, and fast response with low overshoot. Particle Swarm Optimization is employed to obtain optimal gain pairs for each scenario while guaranteeing closed-loop stability. Frequency-domain and time-domain simulation results demonstrate that the proposed approach enables systematic exploration of performance trade-offs. The disturbance-oriented tuning enhances robustness and disturbance attenuation, the control-effort-oriented tuning reduces actuator stress and energy consumption, and the fast-response tuning achieves superior tracking performance. Overall, the proposed framework provides a flexible, stable, and application-oriented PI tuning methodology for delayed battery energy storage and DC microgrid systems.