Analisis Degradasi dan Prediksi Umur Pakai Baterai Lithium-ion Berbasis Dataset NREL

Abstrak

This study analyzes the performance of lithium-ion batteries using the NREL Battery Lifetime Dataset, which contains test results under various operational conditions. The research methodology includes data cleaning, analysis of the relationship between capacity and cycle count, visualization of degradation trends, and lifespan prediction modeling using linear regression. Results indicate that the pulse current charging method exhibits a lower degradation rate compared to the constant current–constant voltage (CC-CV) method, with a final capacity difference of 7% at the 1000th cycle. Parameter analysis before and after 500 cycles shows an 18% decrease in capacity and State of Health, a 22.2% increase in internal resistance, a 13.3% increase in charging time, and a 7.4% decrease in energy efficiency. The regression model yields an R-squared value of 0.991 and a p-value < 0.001, indicating a significant negative relationship between cycle count and capacity. These findings highlight the importance of selecting the appropriate charging method, real-time parameter monitoring, and data-driven battery management to extend lifespan and enhance the reliability of energy storage systems.