Power electronics-enabled online battery impedance spectroscopy advances real-time monitoring for next-generation energy storage

The explosive growth of electric vehicles, renewable energy integration, and large-scale energy storage systems has placed lithium-ion batteries at the heart of the global transition to sustainable energy. Ensuring their safe, efficient, and long-lasting performance demands sophisticated battery management systems capable of continuously tracking critical states such as state of charge, state of health, aging, and potential faults. While conventional monitoring relies on temperature sensors, voltage-current profiling, ultrasonic probes, or embedded optical fibers, these methods often provide limited multidimensional insights, incur high implementation costs, or struggle with real-time applicability in operational environments. Electrochemical impedance spectroscopy stands out as a particularly powerful technique, offering rich, frequency-dependent information about internal electrochemical processes that reflect battery dynamics under varying conditions. However, traditional EIS measurements using dedicated electrochemical workstations remain confined to laboratory settings due to their expense, bulk, and lack of online capability.

To overcome these barriers, researchers have pioneered online battery impedance spectroscopy, which leverages the power electronic converters already present in battery systems—such as DC-DC converters in charging modules or DC-AC inverters in vehicle drivetrains—to generate the necessary small-signal perturbations without additional hardware. By injecting controlled voltage or current ripples through the existing circuit topology, measuring the resulting responses, and computing impedance spectra in real time, this approach transforms routine power conversion into a dual-purpose platform for advanced diagnostics. The method dramatically lowers costs, eliminates the need for specialized high-power perturbation sources, and enables in-situ, non-invasive monitoring that captures battery behavior under actual operating conditions, including dynamic loads and temperature swings.

A comprehensive review of the field highlights two primary implementation strategies: distributed measurements, where individual cells or modules are probed separately via localized converters, and centralized approaches that assess entire packs through master converters. Various circuit topologies and control techniques have been explored, ranging from ripple injection via DC-DC stages to more sophisticated AC perturbations generated by three-phase inverters. These innovations carefully balance the generation of diagnostic signals with minimal disruption to primary functions, often by utilizing energy storage elements as temporary “power pools” to absorb or supply the alternating components. The review also addresses system impacts, including potential effects on load stability, electromagnetic compatibility, and efficiency, alongside mitigation strategies such as filtered perturbations and adaptive control. Perturbation signal designs—sinusoidal sweeps, pseudo-random sequences, or multisine waveforms—pair with tailored processing methods like Fourier analysis, correlation techniques, or least-squares fitting to extract accurate impedance data, with validation steps ensuring reliability even when strict linearity and stability assumptions are relaxed under real-world variability.

The advantages of this power electronics-based online EIS are compelling. It delivers high-information-density diagnostics at a fraction of traditional costs, supports scalable data acquisition across battery packs, and integrates seamlessly into existing battery management architectures with only modest sampling rate enhancements and control refinements. Simulations and experimental validations across reviewed studies confirm robust impedance spectra acquisition with sufficient accuracy for state estimation, paving the way for earlier detection of degradation, lithium plating, or thermal runaway risks.

Looking to the future, online impedance spectroscopy promises to revolutionize battery health management in electric vehicles, grid storage, and renewable microgrids. Enhanced real-time processing could enable predictive maintenance, adaptive charging protocols, and dynamic equalization to extend service life and boost system reliability. Coupling impedance data with rapidly evolving machine learning algorithms holds particular potential for mapping complex impedance-state relationships under diverse conditions, unlocking more precise remaining useful life predictions and fault prognostics. Further research into higher-frequency sampling, noise-robust algorithms, and standardization of perturbation protocols will accelerate commercialization, while integration with cloud-based analytics could support fleet-level insights.

Ultimately, this review underscores a transformative shift in battery diagnostics: from offline laboratory analysis to embedded, online intelligence powered by the very converters that manage energy flow. By harnessing power electronics for dual-purpose impedance spectroscopy, the approach delivers a low-cost, high-value solution that enhances safety, efficiency, and longevity in next-generation energy systems. As intelligent algorithms continue to mature and impedance technology proliferates, it stands ready to play a pivotal role in realizing more resilient, sustainable, and electrified energy infrastructures worldwide.

Reference

Author: Boyang Li ^a b, Ding Luo ^a b, Min Zhou ^a b, Dong Jiang ^a b, An Li ^c

Title of original paper: A review of online battery impedance spectroscope acquisition method based on power electronic system

Article link: https://www.sciencedirect.com/science/article/pii/S2773153725000027

Journal: Green Energy and Intelligent Transportation

DOI: 10.1016/j.geits.2025.100252

Affiliations:

^a School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China

^b State Key Laboratory of Advanced Electromagnetic Technology, Huazhong University of Science and Technology, Wuhan 430074, China

^c Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore