ScholarWorks@경상국립대학교

초록

Accurate state-of-charge (SOC) estimation is essential for battery management systems in portable electronics, electric scooters, and electric vehicles. Most systems rely on fuel gauge integrated circuits using coulomb counting, which accumulates drift over time and requires recalibration through open-circuit voltage (OCV), a process that depends on impractical zero-current rest periods in continuous-use applications. Therefore, this study introduces a hybrid SOC estimation methodology that combines the strengths of traditional OCV methods through a dynamic weights approach. The proposed method uniquely enables recalibration during rest periods without requiring zero current, allowing real-time and reliable SOC monitoring under varying load conditions. The hybrid technique was validated through comprehensive experimentation, including a case study of an electric scooter tested on a C8051F41 microcontroller under room temperature, 5 degrees C, and 45 degrees C operating conditions. This case study simulated real-world operating scenarios, demonstrating the hybrid method's superior accuracy with a mean absolute error of 0.1552% and a root mean square error of 0.2046% at room temperature while maintaining comparable accuracy at 5 degrees C and 45 degrees C, outperforming traditional OCV methods. This adaptive approach ensures robust SOC estimation, making it particularly suitable for microcontroller-based systems where computational efficiency and simplicity are crucial. Through addressing the practical limitations of traditional SOC-OCV methods, this research enhances the capabilities of fuel gauge ICs in commercial applications.

키워드