Smart adaptive CC-CV charger with PSO-accelerated load identification and fuzzy duty-cycle regulation

Abstract

This paper presents an adaptive constant-current/constant-voltage (CC-CV) charger architecture, meticulously designed to address a key challenge in smart chargers. This challenge involves recognizing various battery types and applying the appropriate charging profile expeditiously, without requiring user intervention. The system integrates a particle swarm optimization (PSO) algorithm for ultra-fast load identification with a Mamdani-type fuzzy logic controller for precise duty cycle regulation. The PSO mechanism is capable of determining the optimal initial duty cycle in less than 500 milliseconds. Subsequent to this preliminary initiation, the fuzzy logic controller guarantees the effectiveness of current and voltage regulation during the charging phases. The simulation results obtained from this study validate the system's robustness, as evidenced by the consistent maintenance of voltage ripple below ±0.06 V and current ripple below ±0.04 A. These findings demonstrate the efficacy of the proposed approach in achieving fast, stable, and safe multi-load battery charging. The chemistry-agnostic design of the battery pack is extendable to any battery pack following the CC-CV paradigm, making it highly suitable for practical applications that demand flexibility and high reliability.