A look at traditional and optimized methods of battery charging.
In Part 1 of this series, we introduced the battery management system (BMS) and explained the battery modeling process. In Part 2, we discussed battery state estimation. In this final part, we’ll take a look at battery charging methods.
Battery Charging
A battery is discharged when its voltage is lower than the cut-off voltage or when the battery state of charge is below 20 percent. At this point, it is imperative to stop the discharging process and recharge the battery. Over-discharging and overcharging a battery can affect its condition considerably, as doing so dramatically accelerates battery degradation.
Developing a proper battery charging method is an essential part of the BMS. The method is based on accurate battery estimations for state of charge (SOC), state of health (SOH) and temperature. The proper battery charging approach facilitates efficient battery charging from the initial to the final SOC battery state, as well as protects the battery from overheating, prolonging its life span, and improving capacity utilization.
Temperature is a dominant factor affecting battery charging performance. High temperature decreases the life cycle of Li-ion batteries, and charging is not recommended in below freezing conditions. Table 1 shows the influence of temperature on charging performance for different battery types.
Table 1. Charging performance of different battery types. (Data based on Liu et al., 2018.)
Traditional Battery Charging Methods
There are four commonly used and popular charging methods:
- constant current (CC) charging
- constant-voltage (CV) charging
- constant-current-constant-voltage (CC-CV) charging
- multi-stage constant-current (MCC) charging
CC charging is a simple method that uses a small constant current to charge the battery during the whole charging process. CC charging stops whena predefined value is reached. This method is widely used for charging NiCd or NiMH batteries, as well as Li-ion batteries. The charging current rate is the most important factor, and it can significantly influence the battery’s behavior. For this reason, the main challenge of CC charging is setting a suitable charging current value that will satisfy both charging time and capacity utilization. A high charging current provides a quick charge but also significantly affects the battery’s aging process. A low charging current provides high capacity utilization but also produces a very slow charge, which is inconvenient for EV applications.
Another method is CV charging, which regulates a predefined constant voltage to charge batteries. Its main advantage is that it circumvents overvoltages and irreversible side reactions, thus prolonging battery life. Since the voltage is constant, the charging current decreases as the battery charges. A high current value is required to provide a constant terminal voltage at anearly stage of the charging process. A high charging current from 15 percent to 80 percent SOC provides fast charging, butthe high current stresses the battery and can cause battery lattice collapse and pole breaking.
The main challenge for CV charging is selecting a proper voltage value that will balance the charging speed, electrolyte decomposition, and capacity utilization. Generally, the CV charging method is efficient for speedy charging, but it damages the battery capacity. The negative effect is caused by an increased charging current at a low battery SOC (at the beginning of the charging process), where the current value is significantly higher than the nominal battery current. The high battery current causes the battery lattice frame to collapse and contributes to the pulverization of the active battery pole substance.
The CC-CV charging method is a hybrid approach that combines the two previously mentioned charging methods. It uses CC charging in the first charging stage, and when the voltage reaches the maximum safe threshold value, the charging process shifts tothe CV charging method. The charging process is complete when the current levels off or when full battery capacity is reached. The charging time is mainly defined by the constant current value (CC mode), while the capacity utilization is predominantly influenced by the constant voltage value (CV mode).
CC-CV charging was initially used to charge lead-acid batteries and, later, to charge Li-ion batteries. Li-ion batteries require a much longer CC mode.
The CC-CV charging method is more efficient than either the CC or CV methods individually, and as such it is used as the reference for comparison with the latest charging methods.
The main challenge with CC-CV charging is defining suitable constant values for each mode. The suitable current value will provide a balance between charging performance and battery safety. Having a current that is too high or too low can cause negative effects as previously discussed.
The MCC charging method includes several constant current stages, where the current is gradually decreased as the terminal voltage reaches a default voltage threshold. The charging process continues until the battery reaches the terminal conditions. The MCC method is shown in Figure 2.
The MCC method is suitable for charging the following battery types: lead-acid, NiMH, and Li-ion batteries. With equal initial current values, the MCC charging process takes a bit more time compared to the CC-CV charging method.
Comparison of Traditional Charging Methods
Table 2 summarizes the features of the four traditional charging methods.
Table 2. Comparison of the traditional charging methods.
Simple CC and CV charging methods have low implementation costs, but can cause charging problems such as battery lattice collapse and broken battery poles. Also, these methods pose a challenge in balancing battery capacity utilization and charging speed. Currently, the understanding of battery electrochemical reactions (e.g., lithium plating) is still insufficient and is the focus of research in EV applications.
Optimization of Battery Charging Methods
In order to overcome the existing issues with traditional charging methods, different optimized charging approaches have recently been proposed.
CV charging optimization is focused on the main charging performance factors: charging speed and temperature variation. One of the optimized CV approaches, described in Lee, Chuang, and Wang (2016), proposes a constant voltage charging with various current restrictions to limit variation in battery temperature. Lee and Park (2013) proposed a fast CV charging method by using a developed control scheme that is based on the battery’s internal impedance.
Since the hybrid CC-CV and MCC are quite efficient charging methods, researchers usually focus on optimizing the charging performance of these approaches. Essentially, charging parameters such as charging speed, energy loss, and capacity utilization mainly depend on the number of CC stages and the current values at each stage.
One optimized CC-CV approach proposed by Lee and Park (2013) for Li-ion batteries uses a cycle control algorithm with zero computational methods. In other research, Abdollahi, Han, Avvari, et al (2016) proposed a closed-form approach that considers the charging speed, energy loss, and temperature rise of the battery.
Results show that the charging process is improved through battery capacity and efficiency. MCC charging optimization needs to optimize the number of current stages in the MCC profile and current rates for each CC stage. Different optimization approaches have been suggested. For example, Asadi, Kaboli, Mohammadi, et al. (2012) propose a fuzzy logic controller that converts the charging quality parameters into a single fuzzy dual-response performance index. This approach includes a five-stage MCC charging pattern.
A second optimization approach for MCC charging is known as the Taguchi-based method, described in Liu Y H, Hsieh C H, Luo Y F (2011). This approach helps accelerate the charging process and prolong the life span of Li-ion batteries. The method uses a five-stage MCC charging pattern that is optimized by using the consecutive orthogonal array technique.
Additional optimization methods use computational intelligent technologies such as dynamic programming (DP), model predictive control (MPC), evolutionary algorithms and pseudo-spectral optimization.
The DP method optimizes charging profiles based on the proper battery models. The DP model is one of the most flexible methods to search battery charging profiles because it examines both nonlinear and time-varying parameters in battery models. However, the DP approach requires heavy computation.
The MPC model requires designing the battery model (electric or thermal model) according to how the charging behaviors (current, voltage and temperature) are predicted. This method will be enhanced with improvements in prediction accuracy and a better understanding of the effect that battery temperature and aging have on the model parameters.
The pseudo-spectral optimization approach is able to cover many complicated charging conditions, and it is considered a popular optimization method in real applications. However, the method is based on theoretical foundations and requires extensive battery information, which can be challenging in real EV applications.
Previous parts of the Battery Management Series:
Battery Management Systems – Part 1: Battery Modeling
Battery Management Systems – Part 2: Battery State Estimation
References:
[1] KailongLiu, Kang Li, Qiao Peng, Cheng Zhang, A brief review on key technologies in the battery management system of electric vehicles, January 9, 2018 Front. Mech. Eng.
[2] Lee K T, Chuang C C,Wang Y H, A low temperature increase transcutaneous battery charger for implantable medical devices. Journal of Mechanics in Medicine and Biology, 2016, 16(5): 1650069
[3] Lee Y D, Park S Y. Rapid charging strategy in the constant voltage mode for a high power Li-ion battery. In: Proceedings of 2013 IEEE Energy Conversion Congress and Exposition. Denver: IEEE, 2013, 4725–4731.
[4] Abdollahi A, Han X, Avvari G V, et al. Optimal battery charging, Part I: Minimizing time-to-charge, energy loss, and temperature rise for OCV-resistance battery model. Journal of Power Sources, 2016, 303: 388–398.
[5] Asadi H, Kaboli S H A, Mohammadi A, et al. Fuzzy-control-based five-step Li-ion battery charger by using AC impedance technique. In: Proceedings of Fourth International Conference on Machine Vision (ICMV 11). SPIE, 2012, 834939.
[6] Liu Y H, Hsieh C H, Luo Y F. Search for an optimal five-step charging pattern for Li-ion batteries using consecutive orthogonal arrays. IEEE Transactions on Energy Conversion, 2011, 26(2): 654–661.