Experimental validation of an advanced electrical model for lead-acid batteries

Abstract

Batteries are essential for the transition to sustainable energy sources. While various battery types exist, lead-acid batteries offer the most economical option currently available. The battery model plays an important role in the simulation of renewable energy systems and the estimation of battery condition in the development of the model-based battery management system (BMS). In this paper, we propose a novel approach to develop an innovative electrical model that accurately reflects lead-acid battery behavior. This model, which can be described as an equivalent circuit, reliably represents the lead-acid battery by taking into account all aspects of battery behavior within a dynamic system. It encompasses factors such as internal resistance and capacitance, as well as open circuit voltage (VOC) and state of charge (SOC). Further validation using data from an actual 180 Ah, 2V Tudor C10 battery and comparison with developed models reveals the superior accuracy of the proposed model, particularly during charging and discharging operations. The results show that the proposed electrical model exhibits mean bias error (MBE) and root mean square error (RMSE) values of less than 90 mV/cell and 80 mV/cell, respectively, for charging operations, and less than 160 mV/cell and 70 mV/cell, respectively, for discharging operations. These results highlight the ability of the proposed model to comprehensively capture the dynamic complexities of lead-acid battery behavior, outperforming its counterparts in terms of accuracy under standard test conditions. It serves as a powerful tool in simulation programs such as MATLAB/Simulink, PSIM, PSpice and Labview, facilitating the solution of problems and equations related to electrical circuits.