In modern energy storage systems, the use of series-connected lithium battery modules is common to achieve higher voltage levels. However, inconsistencies among individual cells within these modules inevitably lead to imbalances in state of charge (SOC), which can significantly reduce the overall efficiency, lifespan, and safety of the battery system. To address these challenges, we have developed a hierarchical active equalization topology that leverages SOC as the key variable for control. This innovative approach utilizes bidirectional Buck-boost converters for intra-group balancing and bidirectional flyback converters for inter-group balancing, enabling simultaneous equalization within and between groups. By incorporating fuzzy logic control to dynamically adjust the equalization current, we aim to optimize energy transfer and improve equalization efficiency. This article details the design, simulation, and validation of this topology, demonstrating its superiority over traditional methods in reducing equalization time and enhancing the performance of energy storage lithium battery systems.
The growing demand for efficient energy storage solutions has driven the widespread adoption of lithium battery packs in applications such as electric vehicles, renewable energy integration, and grid stabilization. Energy storage lithium battery systems typically consist of multiple cells connected in series to meet voltage requirements. However, due to variations in manufacturing, temperature, and aging, the SOC of individual cells can diverge over time, leading to reduced capacity, power output, and potential safety hazards like thermal runaway. Passive equalization methods, which dissipate excess energy as heat, are simple but inefficient. In contrast, active equalization techniques transfer energy between cells, offering higher efficiency and better performance. Our research focuses on an active equalization strategy based on SOC, as it directly reflects the remaining energy in each cell, providing a more accurate measure of imbalance than voltage or capacity alone.
Traditional active equalization topologies, such as the single-layer Buck-boost converter, allow energy transfer only between adjacent cells, resulting in longer paths and increased losses for non-adjacent cells. To overcome these limitations, we propose a hierarchical topology that combines intra-group and inter-group equalization. This design reduces the equalization path length and enables direct energy transfer between distant cells, thereby accelerating the balancing process. The core components include bidirectional Buck-boost circuits for balancing pairs of cells within a group and bidirectional flyback circuits for balancing between groups. Additionally, we employ a fuzzy logic controller to regulate the equalization current based on SOC differences, ensuring adaptive and efficient operation. Through simulations in Matlab/Simulink, we compare our hierarchical topology with the conventional single-layer Buck-boost approach, showing significant improvements in equalization speed and effectiveness for energy storage lithium battery packs.
The hierarchical equalization topology is designed to address the inconsistencies in energy storage lithium battery systems by dividing the battery pack into groups. For instance, in a six-cell series configuration, the cells are grouped into pairs, with each pair balanced internally using a bidirectional Buck-boost converter. Meanwhile, the groups themselves are balanced using bidirectional flyback converters. This structure allows for concurrent equalization at multiple levels, reducing the overall time required to achieve SOC uniformity. The bidirectional Buck-boost converter for intra-group balancing operates in discontinuous conduction mode (DCM) to prevent inductor saturation and simplify control. It facilitates energy transfer between two adjacent cells by alternately charging and discharging an inductor through controlled switching. Similarly, the bidirectional flyback converter for inter-group balancing uses a coupled inductor to transfer energy between groups, providing electrical isolation and faster equalization due to its ability to handle higher power transfers.
In the bidirectional Buck-boost converter, the energy transfer process involves two phases: discharge and charge. For example, when cell B1 has a higher SOC than cell B2, the switch Q1 is closed during the first phase (0 < t < dT), allowing current to flow from B1 through the inductor L1, storing energy in the magnetic field. The inductor current during this phase is given by:
$$ I_{L1}(t) = \frac{U_{B1}}{L} t, \quad 0 < t < dT $$
where \( U_{B1} \) is the voltage of cell B1, L is the inductance, and dT is the on-time of the switch. The peak current at time dT is:
$$ I_{\text{peak}} = \frac{U_{B1}}{L} dT $$
During the second phase (dT < t < T), Q1 is turned off, and the inductor discharges through diode D2 to charge cell B2. The current decays linearly as:
$$ I_{L1}(t) = \frac{U_{B1}}{L} dT – \frac{U_{B2}}{L} (t – dT), \quad dT < t < T $$
This cycle repeats, transferring energy from the higher-SOC cell to the lower-SOC cell until their SOC values converge. The use of DCM ensures that the inductor fully demagnetizes in each cycle, avoiding core saturation and simplifying the control design.
For inter-group balancing, the bidirectional flyback converter operates similarly. Consider two groups, BH (cells B1 and B2) and BL (cells B3 and B4), with BH having a higher average SOC. During the first phase (0 < t < dT), switch Q3 is closed, energizing the primary winding T1 of the coupled inductor from BH. The primary current is:
$$ I_{T1}(t) = \frac{U_{BH}}{L_{T1}} t, \quad 0 < t < dT $$
where \( U_{BH} \) is the voltage of group BH, and \( L_{T1} \) is the primary inductance. The peak currents in the primary and secondary windings are:
$$ I_{\text{peak1}} = \frac{U_{BH}}{L_{T1}} dT $$
$$ I_{\text{peak2}} = \frac{N_1}{N_2} \frac{U_{BH}}{L_{T1}} dT $$
where \( N_1 \) and \( N_2 \) are the turns ratios of the primary and secondary windings. In the second phase (dT < t < T), Q3 is turned off, and the secondary winding T2 discharges through diode D4 to charge group BL. The secondary current is:
$$ I_{T2}(t) = \frac{N_1}{N_2} \frac{U_{BH}}{L_{T1}} dT – \frac{U_{BL}}{L_{T2}} (t – dT), \quad dT < t < T $$
This process enables efficient energy transfer between groups, reducing SOC disparities across the entire battery pack. The hierarchical topology thus shortens the equalization path and improves the overall efficiency of the energy storage lithium battery system.
The control strategy for the hierarchical equalization system is based on SOC, as it provides a direct indicator of the energy state in each cell. We implement a fuzzy logic controller to dynamically adjust the equalization current, ensuring optimal performance under varying conditions. The fuzzy controller has two inputs: the difference in SOC between the cells being balanced (ΔS) and the deviation of their average SOC from the overall pack average (Sd). The output is the equalization current (I). The input and output variables are fuzzified into linguistic terms such as Very Small (VS), Small (S), Medium (M), Big (B), and Very Big (VB). The fuzzy rules are designed to prioritize fast equalization when SOC differences are large, while using smaller currents to prevent overcharging or over-discharging when differences are small.
The overall battery pack SOC average is calculated as:
$$ S_{\text{ave}} = \frac{1}{n} \sum_{i=1}^{n} S_i $$
where \( S_i \) is the SOC of cell i, and n is the number of cells. For two cells i and j being balanced, their average SOC is:
$$ S_a = \frac{S_i + S_j}{2} $$
The input variable Sd is defined as:
$$ S_d = S_{\text{ave}} – S_a $$
and ΔS is:
$$ \Delta S = S_i – S_j $$
The fuzzy controller uses these inputs to determine the appropriate equalization current. The rule base is summarized in Table 1, which maps combinations of Sd and ΔS to output current levels. For example, if both Sd and ΔS are large, a large current is applied to quickly reduce the imbalance. Conversely, if both are small, a small current is used to avoid excessive stress on the cells.
| ΔS \ Sd | VS | S | M | B | VB |
|---|---|---|---|---|---|
| VS | VS | S | M | B | B |
| S | S | M | B | B | VB |
| M | M | B | B | VB | VB |
| B | S | M | B | B | VB |
| VB | VS | S | M | B | B |
The output current is defuzzified using the centroid method to obtain a crisp value:
$$ I = \frac{\int z \phi(z) \, dz}{\int \phi(z) \, dz} $$
where z is the output variable and φ(z) is its membership function. This approach allows the equalization system to adapt to changing conditions, enhancing the robustness and efficiency of the energy storage lithium battery management.
To validate the proposed hierarchical topology, we conducted simulations in Matlab/Simulink using a model of a six-cell energy storage lithium battery pack. Each cell had a nominal voltage of 3.2 V and a capacity of 100 Ah. The initial SOC values were set to 58.1%, 59%, 60%, 61%, 62%, and 63% to create an imbalanced condition. The equalization threshold was set to 0.1%, meaning that equalization would start when the SOC difference between any pair of cells or groups exceeded this value. The inductors for the Buck-boost and flyback converters were set to 0.1 mH, and the switching frequency was 10 kHz. We compared the performance of our hierarchical topology with a traditional single-layer Buck-boost topology under three scenarios: static equalization, discharge equalization, and charge equalization.
In the static equalization test, the battery pack was left idle, and the SOC convergence was monitored. The traditional Buck-boost topology took approximately 4,718 seconds to equalize the cells, whereas the hierarchical topology achieved equalization in about 3,028 seconds, a reduction of 35.8%. This demonstrates the effectiveness of simultaneous intra-group and inter-group balancing in reducing equalization time for energy storage lithium battery systems.
For the discharge equalization test, the battery pack was discharged at a constant current of 3 A. The traditional topology required around 5,118 seconds to reach SOC balance, while the hierarchical topology only needed 3,184 seconds, a 37.8% improvement. Similarly, during charge equalization at 3 A, the traditional method took 5,314 seconds, compared to 3,864 seconds for the hierarchical approach, a 27.2% reduction. These results highlight the superiority of the hierarchical topology in dynamic operating conditions, making it highly suitable for real-world applications of energy storage lithium battery systems.
The simulation results are summarized in Table 2, which compares the equalization times for both topologies across different tests. The hierarchical topology consistently outperforms the traditional approach, thanks to its shorter energy transfer paths and adaptive fuzzy control. This leads to faster convergence, reduced energy loss, and improved overall efficiency.
| Test Scenario | Traditional Buck-boost Time (s) | Hierarchical Topology Time (s) | Reduction (%) |
|---|---|---|---|
| Static Equalization | 4718 | 3028 | 35.8 |
| Discharge Equalization | 5118 | 3184 | 37.8 |
| Charge Equalization | 5314 | 3864 | 27.2 |
In conclusion, the hierarchical active equalization topology based on SOC offers a significant advancement in managing inconsistencies in energy storage lithium battery systems. By combining bidirectional Buck-boost converters for intra-group balancing and bidirectional flyback converters for inter-group balancing, along with fuzzy logic control for current regulation, this approach reduces equalization time by at least 27.2% compared to traditional single-layer topologies. The simulations confirm its effectiveness in static, discharge, and charge scenarios, demonstrating improved efficiency and faster SOC convergence. This research provides a reliable solution for enhancing the performance and safety of energy storage lithium battery packs, with potential applications in electric vehicles, renewable energy storage, and smart grids. Future work could explore scalability to larger battery packs and integration with real-time battery management systems.
The development of efficient equalization techniques is crucial for the widespread adoption of energy storage lithium battery technology. As the demand for high-capacity and long-lasting battery systems grows, addressing SOC imbalances becomes increasingly important. Our hierarchical topology not only improves equalization speed but also reduces stress on individual cells, prolonging their lifespan. The use of fuzzy control adds an layer of intelligence, allowing the system to adapt to varying operational conditions. This makes it particularly suitable for dynamic environments where battery usage patterns change frequently. Overall, this study contributes to the ongoing efforts to optimize energy storage lithium battery management, paving the way for more reliable and sustainable energy solutions.
In summary, the key contributions of this work include the design of a novel hierarchical equalization topology, the implementation of a fuzzy logic-based control strategy, and comprehensive validation through simulation. The results underscore the potential of this approach to enhance the performance of energy storage lithium battery systems, making it a valuable reference for researchers and engineers in the field. As battery technology continues to evolve, such innovations will play a vital role in meeting the energy storage challenges of the future.