A Smart Parameter Optimization Framework for Battery Management Systems in LFP Battery Based Electric Vehicles

Abstract

The ever-growing demand for an efficient, reliable and secure electric vehicle has increased the computational requirement for Battery Management System (BMS) for Lithium Iron Phosphate (LiFePO₄) battery technology. This paper presents an approach to systematically analyze the computational requirements of the key parameters in designing an effective and robust BMS architecture for efficient, reliable, and long-term EV operation. Specific attention will be paid to the computational estimation of State of Charge (SOC) A number of algorithms will be implemented and compared in order to evaluate the SOC in terms of accuracy and robustness depending on the computational loads of the system and battery aging. In addition, such important parameters as the balancing voltage, the cooling system parameters, current limiting and state-of-health will be considered to assess the effect of those parameters on the system performance. The presented SOC estimation algorithm and the approach to optimize those parameters will be validated through simulation modeling and hardware-in-the-loop simulations. The architecture of the BMS is based on the current sensing system with potentiometric divider and temperature compensated OP-AMP filter. The experimental results are provided using twelve sets of time series (voltage, current, power) data processed and visualized with MATLAB and Excel computational algorithms. The designed architecture is modular, software programmable, cost-effective and easy to implement in smart and off-grid computational energy systems.This system includes tools for monitoring, evaluating, and diagnosing the health and efficiency of lithium-ion batteries.