<![CDATA[Lithium-ion batteries have become one of the top choices for efficient and environmentally friendly mobility in today's era. Batteries play an important role in our digital lifestyles, from smartphones to electric cars. The use of this battery is inseparable from the challenge of estimating the State of Charge (SOC), which is a key parameter to monitor the availability of energy remaining in the battery. Therefore, an accurate SOC Estimation method is needed, which is important for efficient energy management and safe battery use. The Long Short-Term Memory (LSTM) model was chosen because of its ability to handle complex time series data and nonlier patterns in battery performance. This study provides the application of LSTM for SoC estimation and shows that LSTM is superior to the Feed Neural Network (FNN) method as evidenced by the simulation results that show that the LSTM model produces an RMSE of 4.92%, while the FNN model produces an RMSE of 7.82. From all the tests that have been carried out, the best RMSE value of 3.53% was obtained at a temperature of 25°C epoch 100.]]>
Copyrights © 2024
