This study presents a hybrid method for state-of-charge (SOC) estimation of lithium-ion batteries using LSTM neural networks optimized with genetic algorithms (GA), combined with Coulomb Counting (CC) as an initial estimator. Experimental tests were conducted using medium-voltage (48–72 V) lithium-ion battery packs under standardized driving cycles (NEDC and WLTP). The proposed method enhances prediction accuracy under dynamic conditions by recalibrating the LSTM output with CC estimates through a dynamic fusion parameter . The novelty of this approach lies in the integration of machine learning and physical modeling, optimized via evolutionary algorithms, to address limitations of standalone methods in real-time applications. The hybrid model achieved a mean absolute error (MAE) of 0.181%, outperforming conventional estimation strategies. These findings contribute to more reliable battery management systems (BMS) for electric vehicles and second-life applications.
An integrated model is presented to estimate the energy consumption of a Formula SAE–type electric vehicle by combining MATLAB/Simulink simulations with real operational data from the Cotopaxi kart circuit. The implemented subsystem incorporates vehicle dynamics, speed, and grade profiles, and it enables the calculation of tractive energy across different competition scenarios. The L3 cycle (maximum speed 90.91 km/h, distance 16.67 km) proved the most demanding, with a tractive energy consumption of 1389.10 Wh, mechanical losses of 630.89 Wh, and a useful net energy of 758.21 Wh. Rolling resistance and inertia accounted for 21.33% and 56.29% of the consumption, respectively, highlighting the influence of acceleration/deceleration dynamics. All cycles were dominated by active phases, with 0% stops and cruising time < 0.21%, validating the model for high-demand conditions. The technical feasibility of second-life Lithium Iron Phosphate (LFP) cells for the battery pack was also confirmed: a 30s2p configuration using 100 Ah cells meets the peak energy demand of 18.30 kWh while maintaining adequate operational margin and a safe discharge C-rate. Structural simulation in ANSYS enabled optimization of mass and stiffness, reducing overall energy demand. This multidisciplinary approach provides a quantitative basis for battery sizing and energy management strategies in competitive electric vehicle applications.