A Novel Method of Charging Strategy Optimization Using Evolutionary Algorithm in Battery Electric Vehicles

Electric vehicles (EV’s) are gaining increasing popularity because of fast depleting fossil fuels and increasing environmental hazards caused by vehicles that use fossil fuels as a source of propulsion. One of the key factors in EV development is battery management and time required to fully charge the battery. Constant-Current (CC) and Constant-Voltage (CV) are the most common charging strategies used for the charging of batteries. High charging current can cause the temperature of the battery to rise and shorten the battery life. At low State of Charge (SoC) levels, the open circuit voltage (OCV) of the battery is low and CV strategy may cause high charging current, adversely affecting the battery life. Therefore, CCCV is the most commonly used charging strategy, in which battery is charged by CC at low SOC levels and after the OCV reaches a limit, the CV method is applied for charging. A comparison between conventional CCCV profile charging and optimized CCCV profile charging is done. Three factors are considered in the paper for optimization, i.e. charging time, temperature rise and energy losses during charging. A cost function is calculated for all the factors and weights are assigned for each parameter. This will give the manufacturer, the ability to set the priority for optimization parameters. This optimization problem is complex and non-linear in nature as the battery can be charged with different currents at different SOC levels. Evolutionary algorithms (EA) are general population-based algorithms inspired by the biological evolution of life form in nature. In this paper, genetic algorithm, a form of EA, is applied for finding the best solution to the optimization problem in charging of Lithium-Ion battery.