Experimental Validation of an Adaptive Series-Parallel Recombination Battery-Balancing Architecture Using Second-Life Lithium-Ion Cells

Electrical and Electronic Engineering Advanced cell balancing strategies for scalable lithium-ion battery management in electric vehicles

Advanced cell balancing strategies for scalable lithium-ion battery management in electric vehicles

The growing demand for efficient and reliable energy storage in electric vehicles has led to significant advancements in battery management systems (BMS), particularly in cell balancing. While existing research has primarily focused on balancing techniques using a small number of cells (typically 2 to 4), or lead-acid batteries, and employed algorithms like Kalman filters and state of charge (SoC)-based estimations for passive resistor-based or simple inductor/capacitor-based balancing, these approaches often face limitations in energy efficiency and scalability. In contrast, this study presents a comprehensive analysis of 13 lithium-ion cells using both passive and active balancing, as well as an advanced active staggered cell balancing strategy. The proposed system reallocates the charge more efficiently among cells, minimizing thermal loss and improving overall energy utilization. Using MATLAB®/Simulink® simulations, we evaluate and compare the simultaneous start and staggered start methods under various load conditions. Results show that active balancing methods can reduce energy loss from 15–25 W (typical in passive systems) to 3–7 W and decrease balancing time from 120–180 minutes to 45–80 minutes. Notably, the staggered start method delivers superior thermal management and extended battery life, confirming its suitability for high-performance and scalable BMS in electric vehicles.

Optimization of LIFEPO4 Battery Pack for Electric Vehicles Application Using Active Balancer

Cell-to-cell imbalance in series-connected LiFePO4 packs limits usable capacity and can introduce avoidable balancing losses, particularly when equalization relies on dissipative methods. This study quantifies the impact of adding a bidirectional active balancer to an electric-vehicle LiFePO4 battery pack that is otherwise managed by a Battery Management System (BMS). Two experimental configurations were evaluated under identical charge-discharge cycles: (1) BMS-only operation and (2) BMS operation combined with an active balancer (balancing current capability up to 5 A) that redistributes charge from higher-voltage cells to lower-voltage cells. Cell-voltage and pack-parameter data were acquired via the BMS communication interface using a Raspberry Pi, stored in a database, and visualized in real time through a Node-RED dashboard, enabling synchronized, cycle-by-cycle inspection of equalization behavior. Performance was assessed using charge-discharge coulombic efficiency together with the reduction in cell-voltage spread during charging and discharging. The measured coulombic efficiency with BMS-only operation is 87.50%, indicating significant loss during equalization, whereas adding the active balancer increases coulombic efficiency to 99.55% (a 12.05% improvement) and produces substantially tighter voltage uniformity. The results demonstrate that active balancing can markedly reduce balancing losses in LiFePO4 packs while providing a reproducible instrumentation workflow for laboratory and training use.

An Online Active Balancing Technique for Homogeneous and Heterogeneous Battery Packs

With the wide-scale deployment of battery packs in a variety of applications, some life cycle challenges are coming to light. These challenges include extending their operational life as a pack given their increasing cell-level imbalances and repurposing their cells into new battery packs to give them a second life, e.g., in grid storage. This paper presents a new circuit topology addressing both issues using active (controlled and nondissipative) cell-to-cell balancing for online operation, i.e., while the battery energy storage system is in use. The proposed circuit design is fast and safe for balancing, relying on current control to target each individual cell’s maximum charging and discharging current, while taking into account the pack current. The design has the lowest number of switches and circuit components compared to the state-of-the-art techniques, and is also flexible, allowing for the addition or replacement of cells in series. Its practical hierarchical control system enables speed, reliability, and reconfigurable limits in real-time operation on the individual cells. Experimental validation is carried out on homogeneous and heterogeneous packs in online operation, and the results demonstrate the speed and efficacy of the proposed technique.

Adaptive Control of Dual-Phase Bidirectional Flyback Converters for Efficient Cell Balancing in Lithium-Ion Battery Packs: A Comprehensive Review

The intensive development of electric vehicle (EV) technology, renewable energy systems, and stationary energy storage solutions has amplified the demand for advanced Battery Management Systems (BMS). The imbalance in cells within lithium-ion battery packs, due to manufacturing tolerances, varying aging, and thermal gradients, reduces available capacity, cycle life, and can cause thermal runaway. Active charge equalization with DC–DC converters has become a recent research focus among various balancing techniques because it has a better capability of redistributing energy. This paper gives a detailed study of converter-based cell-balancing topologies with a specific focus on the bidirectional flyback converter and the interleaved two-phase variant. Non-isolated topologies (buck–boost, Cuk converter topology, interleaved buck–boost) and isolated topologies (flyback, push–pull, dual-active bridge, LLC resonant) are compared concerning functional efficiency, component reduction, galvanic isolation, scalability, and bidirectional capability. The concept of soft-switching, including zero-voltage switching (ZVS) and zero-current switching (ZCS), and their circuit realizations are discussed. Advanced control models and artificial intelligence (AI) for the estimation of state-of-charge (SoC) and real-time optimization are mentioned. Thermal issues, scalability, reliability, and wide-bandgap semiconductor devices (SiC/GaN) are discussed.

Dynamic Duty Modulation to Enhance Voltage Equalization Performance of the Advanced Cell Balancer

Cell balancing is a critical requirement for ensuring the safety and longevity of supercapacitor (SC)-based energy storage systems (ESSs). While conventional series-parallel active balancing methods predominantly utilize a fixed 50% duty ratio, this static control mechanism fundamentally limits the energy transfer rate, particularly under severe initial voltage imbalances and non-ideal cell variations. To accelerate the equalization process without incurring the cost, volume, and complexity penalties of hardware modifications, this paper proposes a dynamic duty modulation (DDM) technique. The proposed DDM algorithm adaptively adjusts the switching duty ratios in real-time based on the instantaneous voltage deviations of individual cells. A discrete-time mathematical model is derived to evaluate the convergence behavior, analytically proving that the dynamic approach requires significantly fewer switching cycles than the conventional fixed-duty method. The mathematical superiority and practical feasibility of the proposed technique are strictly validated through comprehensive PSIM simulations. Quantitative simulation results confirm that the proposed DDM reduced the balancing time by over 48% and proportionally decreased the total number of switching cycles compared to the conventional method, providing an enhanced dynamic performance suitable for advanced battery management systems (BMSs).