The heart of the problem lies in how batteries are modeled. Traditional equivalent circuit models represent a cell as a network of resistors and capacitors, which is computationally cheap and therefore attractive for embedded battery management systems, but integer-order components carry hidden assumptions. Real batteries do not behave like ideal capacitors; their charge transfer and diffusion processes are distributed across time scales and across the physical volume of the electrodes, producing frequency-dependent impedance that lumped elements simply cannot replicate. The result is systematic voltage error, poor performance under varying temperatures, and the need for constant recalibration as the cell ages. Fractional-order models solve this by allowing non-integer exponents in the calculus that describes the system, capturing memory effects and non-exponential relaxation naturally. A key element is the constant phase element, whose exponent alpha slides continuously between the behavior of a resistor and a capacitor, and whose value of one half corresponds to the classic Warburg diffusion element seen in electrochemical impedance spectroscopy.

But fractional-order models come with a price: identifying their parameters is brutally difficult. The team adopted an R(RQ)(RQ)W structure, an advanced extension of classical second-order and PNGV models, in which an ohmic resistance is followed by two parallel resistor-constant-phase-element branches and a Warburg element. Each component maps onto a distinct region of the Nyquist plot, the standard graphical representation of impedance data. The ohmic resistance appears as the high-frequency intercept on the real axis, the two RQ branches form overlapping depressed semicircles in the mid-to-high frequency range corresponding to charge transfer and slower interfacial reactions, and the Warburg element produces the characteristic 45-degree low-frequency tail of solid-phase diffusion. Eight parameters must be extracted simultaneously, and they are strongly coupled, meaning an error in one drags the others off course. Metaheuristic optimizers, which explore search spaces without gradients, are the usual tool, but they frequently converge to local optima that look good numerically yet produce physically meaningless parameters.

To choose the best optimizer fairly, the researchers staged a systematic tournament of ten metaheuristic algorithms: three established methods, particle swarm optimization, grey wolf optimization, and artificial bee colony, against seven recent bio-inspired contenders including the sea lion, golden jackal, fox, honey badger, seahorse, sand cat, and marine predator algorithms. Each ran with an identical budget of 10,000 swarm members and 100 iterations across 20 independent trials. Scoring by root mean square error, mean absolute percentage error, and runtime, the Marine Predators Algorithm, which mimics the Lévy flight and Brownian motion strategies of foraging marine animals, emerged as the most robust performer. A Friedman statistical ranking placed it first with an average rank of 2.00, narrowly ahead of the Honey Badger Algorithm, and its mean runtime of 114 seconds comfortably beat both the honey badger’s 150 seconds and the artificial bee colony’s punishing 2,518 seconds. Yet even the winner showed troubling discrepancies for the second semicircle’s resistance and capacitance and the Warburg coefficient, confirming that raw optimization power alone was insufficient.

The first fix is elegantly physical. Instead of letting the optimizer start from arbitrary bounds, the team derived initial parameter estimates directly from the geometry of the Nyquist plot. A parallel resistor-capacitor circuit traces a perfect semicircle of radius R over two; replacing the capacitor with a constant phase element rotates that semicircle by an angle theta that depends analytically on the fractional exponent alpha, following the relation theta equals 90 degrees times one minus alpha. By extending the measured arcs to full circles, reading off their real-axis intercepts, and measuring the rotation angles of their apexes, the researchers could calculate the ohmic resistance, both branch resistances, both constant phase element exponents and capacitance offsets, and even the Warburg value from the lowest-frequency data point. Six of the eight parameters were thus estimated before optimization even began, shrinking the search space from bounds spanning twenty-fold in each direction to a tight window around physically consistent values.

The second fix attacks dimensionality. Rather than asking the optimizer to juggle all eight parameters at once, the two-stage refinement alternates between metaheuristic optimization of six parameters and analytical recovery of the remaining two. Once the first stage converges on R0, R1, Q1, alpha1, alpha2, and W1, the residual impedance, obtained by subtracting those known contributions from the measured spectrum, is fitted by global least squares to extract R2 and Q2 over the second branch’s frequency band. A Trust Region Reflective algorithm then polishes the estimates, and the cycle repeats for ten iterations with bounds that tighten by a factor of two at each stage. The swarm size drops from 10,000 in the brute-force baseline to just 2,000 in the first stage and 100 thereafter, cutting the iteration count by roughly 80 percent while improving precision.

The validation was unusually thorough. On 50 synthetic impedance profiles generated by Latin Hypercube Sampling, where the true parameter values were known exactly, the full framework achieved an overall mean absolute percentage error of just 7.27 percent, compared with a catastrophic 24,462 percent for standard marine predator optimization and an even worse result for two-stage refinement without informed initialization. The mean root mean square error fell to 1.74 times ten to the minus five, a 92.4 percent reduction relative to the baseline, with a median near ten to the minus six. Scatter plots of estimated versus true parameters clustered tightly along the ideal 45-degree line for all eight quantities, while competing methods showed dispersion spanning orders of magnitude. Crucially, the team demonstrated a subtle pitfall: methods 1 and 2 sometimes achieved low numerical error while their simulated Nyquist plots visibly failed to match the target data, proof that a low objective function value does not guarantee a physically accurate model.

Robustness testing added further confidence. Under 1 percent Gaussian noise the overall parameter error rose only from 7.27 to 12.92 percent, and under 2 percent noise to 18.41 percent, with no instability or divergence. Removing the five lowest-frequency data points, simulating instruments that cannot capture the Warburg region, barely changed the error at 7.38 percent, because the initialization relies on mid-to-high-frequency arc geometry. A transfer test on the simpler R(RQ)W structure showed the initialization concept generalizes across model topologies when the correct structure is known, though the authors are careful to note that automatic model-structure selection from raw data remains an open challenge. Against three state-of-the-art published methods, chaotic adaptive fractional particle swarm optimization, hybrid genetic-particle swarm optimization, and complex-order beetle swarm optimization, the new framework delivered curve-fitting errors one to two orders of magnitude lower, with the rivals showing parameter errors often exceeding several hundred percent for the second branch and Warburg elements.

Finally, the method proved itself on real hardware. The team measured electrochemical impedance spectra from commercial LG INR21700M50LT nickel-manganese-cobalt cells in three states: pristine, initially degraded at 90 percent state of health, and significantly aged at 80 percent state of health and 50 percent state of charge. Across 20 independent runs per condition, the full framework reduced mean root mean square error by 38.0, 32.5, and 76.6 percent respectively compared with the standard optimizer, and achieved parameter coefficients of variation as low as 0.016 percent, meaning repeated identifications landed in the same parameter basin with remarkable consistency. That repeatability is what makes the parameters trustworthy inputs for state of charge and state of health estimation, the foundation of reliable battery diagnosis, degradation tracking, and second-life applications. The authors acknowledge their framework assumes the two impedance arcs remain geometrically distinct, and extending it to overlapping arcs and adaptive topology selection is future work. But the central lesson stands: injecting electrochemical knowledge into the starting point of an optimization can accomplish what brute-force computation cannot, at a fraction of the cost.

Subject of Research: Physics-informed parameter identification for fractional-order lithium-ion battery models using electrochemical impedance spectroscopy

Article Title: Physics-informed parameter identification of fractional-order battery models using Nyquist-based initialization and two-stage metaheuristic refinement

Article References: Wang, S.-C., & Liu, Y.-H. (2026). Physics-informed parameter identification of fractional-order battery models using Nyquist-based initialization and two-stage metaheuristic refinement. Results in Engineering, 32, Article 113425. https://doi.org/10.1016/j.rineng.2026.113425

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113425

Keywords: lithium-ion batteries, fractional-order models, electrochemical impedance spectroscopy, parameter identification, metaheuristic optimization, Marine Predators Algorithm, Nyquist plot, battery management systems, constant phase element, state of charge, state of health, equivalent circuit model

To appreciate why this matters, it helps to understand what formation actually is. When a lithium-ion cell is assembled, it is not yet a functional battery. It must first be charged and discharged in carefully controlled cycles, a process called formation, during which the solid electrolyte interphase, a critical protective layer on the anode, is established. This stage is among the most expensive and time-consuming steps in battery manufacturing, and manufacturers around the world use different formation protocols, varying the charging currents, voltages, and cycle patterns, in pursuit of faster production or better performance. The trouble is that these protocol differences leave distinct fingerprints on the cell’s long-term degradation behavior. A prediction model trained on batteries formed one way may fail badly when applied to batteries formed another way, which is precisely the problem the research team set out to solve.

Traditional approaches to remaining useful life prediction, the researchers note, often suffer from poor accuracy and weak robustness when batteries have undergone different formation protocols. The core difficulty is that formation-induced differences create systematic biases in the early-cycle data that prediction models rely on. Features extracted from charge-discharge curves may look deceptively similar across cells while hiding fundamentally different aging trajectories. The new method confronts this head-on with a two-pronged strategy: an unscented transformation tailored for variable formation protocols, and a hybrid framework combining an Unscented Particle Filter with a Gated Recurrent Unit neural network, complete with a protocol-sensitivity compensation module designed to mitigate the bias that protocol differences inject into predictions.

The first component of the framework is an exponential empirical degradation model, fitted using an Unscented Particle Filter, that captures the underlying aging mechanism of the battery. Empirical degradation models describe how capacity fades over cycles using mathematical functions, and the particle filter is a Bayesian inference technique that continuously updates the model’s parameters as new measurements arrive. The unscented variant is particularly well suited to this task because it propagates uncertainty through nonlinear transformations without requiring the model to be linearized, a step that can introduce serious errors in battery systems where degradation is anything but linear. By anchoring the prediction in a physically motivated degradation model, the framework avoids the worst failure mode of purely data-driven approaches: extrapolating confidently into regions where the training data offers no guidance.

The second component is where the deep learning enters. The researchers constructed a Gated Recurrent Unit neural network, a streamlined relative of the long short-term memory architecture that has become a workhorse of time-series prediction. The network takes key features from constant-current and constant-voltage charge-discharge cycles as its inputs, treats the battery’s state of health as a hidden state that evolves over time, and outputs the remaining useful life under varying formation protocols. Gated recurrent units are prized for their ability to capture temporal dependencies in sequential data while remaining computationally lighter than full LSTM networks, making them attractive for applications where predictions must be updated continuously as new cycling data streams in.

What makes the architecture genuinely novel is the way these two components collaborate. Rather than letting the neural network operate alone, the framework integrates empirical modeling with data-driven prediction, allowing the particle-filter-based degradation model to refine the network’s RUL estimates while the protocol-sensitivity compensation module adapts dynamically to variations in formation conditions. In effect, the empirical model provides a physically grounded skeleton, the neural network fleshes out the complex, protocol-dependent details, and the compensation module corrects for the systematic distortions that different formation histories introduce. This model-data fusion philosophy reflects a broader trend in battery prognostics, where researchers increasingly recognize that neither pure physics nor pure machine learning is sufficient on its own.

Validation was carried out on public data from the University of Michigan Battery Lab, specifically the fast formation study dataset published through the university’s Deep Blue Data repository. This dataset is a valuable testing ground precisely because it contains batteries cycled under different formation protocols, allowing the researchers to systematically quantify and model the impact of protocol differences on long-term degradation and remaining useful life prediction. The results were striking: the proposed method achieved an R-squared value of 99.41 percent, meaning it explained nearly all of the variance in the observed degradation trajectories, and a root mean square error of 2.23 percent, indicating tight agreement between predicted and actual values. The team reports that the method significantly outperformed comparison methods in both state of health and remaining useful life prediction accuracy, demonstrating superior cross-protocol generalization and robustness compared with existing hybrid approaches.

The practical implications of this work extend well beyond the laboratory. Accurate remaining useful life prediction is essential for determining appropriate battery replacement times, a decision with enormous consequences for safety, cost, and sustainability. In electric vehicles, overestimating a battery’s life risks stranding drivers with a depleted pack, while underestimating it wastes money by replacing cells that still have serviceable capacity. In aerospace and space applications, where some of the co-authors’ institutions operate, the stakes are even higher, since battery failure can compromise entire missions. A prediction framework that remains accurate regardless of how the cells were formed could allow manufacturers to mix batteries from different production lines or formation batches into the same fleet management system without sacrificing prognostic reliability, something current methods struggle to guarantee.

There is also a manufacturing angle that could prove commercially significant. Formation is one of the most energy-intensive and capital-heavy stages of cell production, and the industry has been actively exploring faster formation protocols to cut costs and increase throughput. One reason manufacturers hesitate to adopt aggressive fast-formation schemes is uncertainty about how those choices will affect long-term life and whether existing quality-control and prognostic tools will still work. A prediction method that explicitly compensates for protocol sensitivity could lower that barrier, giving manufacturers the confidence to experiment with formation strategies that shorten production times without losing the ability to forecast field performance. The research team, which includes authors affiliated with CATL, the world’s largest battery manufacturer, is well positioned to understand these industrial pressures.

The study, led by Gaige Chen and Haitao Meng, who contributed equally, with Yahong Yang, Xiaoshen Zhang, Wei Deng, and Jinsong Liu, was supported by the National Natural Science Foundation of China and builds on the group’s earlier work on transfer learning for state of health prediction under cross-formation protocols. As batteries proliferate through every corner of the energy economy, tools like this hybrid UPF-GRU framework point toward a future in which the lifespan of every cell can be forecast accurately from its earliest cycles, regardless of the manufacturing choices that shaped it. That kind of predictive certainty could reshape how batteries are designed, produced, warranted, and retired, turning one of the industry’s most persistent blind spots into a solved problem.

Subject of Research: Machine learning-based remaining useful life prediction of lithium-ion batteries across different formation protocols

Article Title: Remaining useful life prediction of lithium-ion batteries across formation protocols

Article References: Chen, G., Meng, H., Yang, Y., Zhang, X., Deng, W., & Liu, J. (2026). Remaining useful life prediction of lithium-ion batteries across formation protocols. Ionics. https://doi.org/10.1007/s11581-026-07472-5

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07472-5

Keywords: lithium-ion batteries, remaining useful life, formation protocols, unscented particle filter, gated recurrent unit, state of health, battery degradation, machine learning, battery manufacturing, prognostics, CATL, University of Michigan Battery Lab

Predictive maintenance rests on a deceptively simple bargain. Algorithms trained on operational data learn to estimate a battery’s state of charge, state of health and remaining useful life, allowing operators to schedule interventions before failures occur rather than after. But the models behind these estimates are typically trained under a set of assumptions: that the battery will perform a defined duty, that environmental conditions will resemble those seen during training, and that the monitoring pipeline feeding data to the algorithm will stay consistent. The Perspective’s central claim is that all three assumptions are increasingly false for modern storage assets, and that the resulting mismatch, which the authors call operational drift, is a structural failure mode rather than an occasional nuisance.

The sources of drift are more varied than they might first appear. Application requirements evolve as grid markets change and new revenue streams emerge, so a system installed for peak shaving may later be dispatched for frequency regulation with very different charge and discharge profiles. Duty cycles shift accordingly, and research has shown that dynamic cycling can actually enhance battery lifetime, meaning that degradation trajectories can bend in either direction when usage patterns change. Environmental exposure varies too, from extreme cold climates that demand heating and cooling loads to be factored into system sizing, to temperature differences that strongly influence which degradation mechanisms dominate. Even the monitoring pipeline itself can drift, as sensor biases, random errors, low sampling rates and intermittent data unavailability alter the statistical character of the very signals the algorithms depend on.

To bring order to this complexity, the authors classify drift by its origin in evolving asset roles and then organize its effects into three layered functional mismatches. The first is a data-layer mismatch: the distribution of inputs seen in deployment no longer matches the training distribution, a phenomenon statisticians call covariate shift. The second is a model-layer mismatch, where the relationship the algorithm learned between features and outcomes no longer holds, for example when a new duty cycle activates degradation pathways such as lithium plating that were rare during training. The third is a temporal-layer mismatch, arising because degradation is not stationary in time; knee points on capacity-fade curves mark transitions from gradual to accelerated decline, and a model calibrated on early-life behavior can be badly wrong after such a transition.

Each mismatch layer, the authors argue, calls for a targeted transfer strategy rather than a one-size-fits-all fix. At the data layer, domain adaptation techniques can align source and target distributions, using statistical distances such as maximum mean discrepancy or Wasserstein distance to quantify and correct the shift. Adversarial variants go further, learning features that a domain classifier cannot distinguish, so that a model trained on laboratory cells or on one fleet can generalize to another. At the model layer, transfer learning approaches range from freezing a backbone network and attaching small trainable adapter layers, to physics-informed schemes that embed electrochemical knowledge, such as Arrhenius temperature dependence, directly into activation functions. Residual learning offers another elegant option: instead of learning the full mapping, the model learns only the difference between a physics-based reference prediction and observed behavior, which keeps adaptations small and interpretable.

The temporal layer demands perhaps the most delicate machinery, because adapting a model to new conditions can destroy what it already knows. The Perspective explicitly assesses candidate strategies against four failure risks: negative transfer, where adaptation to a new domain actually degrades performance; catastrophic forgetting, where learning new behavior erases old competence; poorly calibrated uncertainty, where the model’s confidence no longer reflects its true reliability; and low sample efficiency, where adaptation demands more target data than operators can realistically collect. Continual learning methods, including regularization-based approaches and bilevel-optimized schemes, have been shown to preserve earlier knowledge while absorbing new operating conditions, and meta-learning can adapt to a fresh battery with only a handful of target measurements.

What makes the argument timely is the sheer scale of the stakes. Grid-scale storage is expanding rapidly as renewable generation grows, and batteries increasingly provide multiple services simultaneously, from arbitrage and peak shaving to regulation and reliability support. Studies of vehicle-to-grid operation show that the dominant degradation mechanism determines whether extra energy throughput is worth the days of life lost, which means maintenance decisions are inseparable from market strategy. Meanwhile, retired electric vehicle batteries are flowing into second-life deployments where their histories are heterogeneous and often poorly documented, and where a model trained on first-life data may face a target domain it has never seen. The authors position transferable predictive maintenance as a life-cycle-spanning capability that connects manufacturing, first-life operation and second-life deployment into one continuous chain of health information.

The framework also carries implications that reach back to the factory floor. Research on transfer learning for battery cell manufacturing suggests that models trained during formation and early testing can predict lifetime outcomes, and that low-state-of-charge resistance measured immediately after manufacturing is a rapid predictor of how formation protocols will shape long-term durability. If such early-life signals could be carried forward through a battery’s passports and reused when the cell changes roles, the cold-start problem of second-life health estimation would shrink dramatically. Recent work on history-free prediction of second-life degradation trajectories from a single partial cycle, using interpretable mixture-of-experts models, hints at how much can be recovered even from sparse information, but the authors contend that systematic transferability would make such feats routine rather than exceptional.

None of this can be built on algorithms alone, and the Perspective is unusually concrete about the infrastructure required. The authors call for data infrastructures that preserve context across the life cycle, foundation models for batteries that encode broad electrochemical and operational knowledge and can be cheaply adapted to new tasks, and open benchmarks that test transferability under realistic drift rather than stationary laboratory conditions. Initiatives toward unified battery data descriptions and large-scale real-world datasets of retired cells point in this direction, as does work on federated learning, which allows fleets owned by different parties to collaborate on health estimation without sharing raw data. Privacy-preserving collaboration matters because the data that would make transferable maintenance most powerful, spanning manufacturers, fleet operators and recyclers, is exactly the data that commercial interests are most reluctant to pool.

The authors close with regulatory recommendations, arguing that standards should require calibrated uncertainty alongside point predictions and should encourage the audit dimensions their framework defines, so that operators can verify not just whether a model is accurate but whether it remains trustworthy as conditions drift. If they are right, the payoff extends well fewer failures and longer asset lives: batteries that can document their own health across changing roles would become easier to resell, repurpose and recycle, tightening the loop of a circular battery economy. In a decade when storage assets are expected to outlive the markets they were built for, the ability of their maintenance algorithms to move with them may prove as important as the chemistry inside the cells. Transferable predictive maintenance, the authors suggest, is the missing link between the batteries we build and the many lives we ask them to live.

Subject of Research: Transferable predictive maintenance for battery energy storage systems under operational drift

Article Title: Transferable predictive maintenance for battery energy storage systems

Article References: Tao, S., Liang, C., Tavallaey, S. S., Thulin, N., Wik, T., Zio, E., Pecht, M. G., Mo, H., & Zou, C. (2026). Transferable predictive maintenance for battery energy storage systems. Nature Reviews Electrical Engineering. https://doi.org/10.1038/s44287-026-00337-z

Image Credits: AI Generated

DOI: 10.1038/s44287-026-00337-z

Keywords: battery energy storage, predictive maintenance, transfer learning, operational drift, domain adaptation, continual learning, state of health, second-life batteries, uncertainty quantification, foundation models, grid storage, battery degradation

The research, led by Kui Chen with colleagues including Xiaoying Zheng, Huai Gao, and Guangning Wu, centers on a multi-physics coupled model that links three intertwined processes inside a single lithium-ion cell: the electrochemical reactions that store and release charge, the thermal behavior that governs how heat is generated and distributed, and the growth of the solid electrolyte interphase, or SEI, a thin film on the graphite anode that thickens over a battery’s life and consumes cyclable lithium. By coupling these domains, the model can simulate not just how a cell performs on a given cycle, but how each cycle quietly reshapes the cell’s internal chemistry and future performance.

The SEI film deserves particular attention because it sits at the heart of battery aging. Formed during the first charge cycles when the electrolyte decomposes at the anode surface, the SEI is nominally protective, passivating the anode against further reaction. But it is never perfectly stable. Each cycle, and especially each cycle run hot or at high current, adds new layers of decomposition products, and every increment of film growth permanently removes lithium ions from circulation. The result is a slow, irreversible loss of capacity that engineers call calendar and cycle aging. The Chinese team’s model treats this growth explicitly rather than as an afterthought, allowing them to track how aging and heating feed back into one another.

The team’s simulations compared cells cycled at different C-rates, a measure of how quickly a battery is charged or discharged relative to its capacity. A 1 C rate means a full discharge in one hour; 4 C means the same energy is extracted in fifteen minutes, a regime closer to what a locomotive experiences during hard acceleration or regenerative braking recovery. The results were unambiguous. Higher discharge rates accelerated the decline of cell voltage, shortened the usable discharge duration, and promoted capacity loss. After 1000 simulated cycles, the state of health, a standard metric expressing remaining capacity as a fraction of the original, stood at 0.902 for cells cycled at 1 C, 0.899 at 2 C, and 0.896 at 4 C.

Those capacity differences may look modest, but the thermal consequences of fast cycling are anything but. At a 4 C discharge, the model showed the temperature-rise rate was approximately 2.29 times that observed at 1 C. Heat generation in a lithium-ion cell scales steeply with current because both ohmic losses and polarization losses grow as charge is pushed through the cell faster. In a compact traction pack, that heat must be removed quickly or it accumulates, and accumulated heat does more than threaten safety: it accelerates the very aging processes that erode capacity, creating a feedback loop in which hard use makes the battery both hotter and weaker.

The SEI growth data made that feedback vivid. Under 4 C cycling, the increment in SEI film thickness was approximately 2.23 times that at 0.5 C and 1.78 times that at 1 C. In other words, pushing a cell four times harder did not merely quadruple the wear proportionally; it compounded the chemical damage, thickening the interphase far faster than gentle cycling. Every micron of extra SEI represents lithium locked away from useful work, and the model’s ability to resolve this growth cycle by cycle gives engineers a way to predict where on a duty profile a locomotive battery pays the highest aging price.

Ambient temperature emerged as an even more powerful lever on degradation. When the team cycled cells at 40 degrees Celsius, the state of health fell to 0.824 after the same 1000 cycles, and the SOH loss was approximately 2.8 times that observed under the low-temperature condition. The finding aligns with a well-established body of evidence that elevated temperatures accelerate electrolyte decomposition and SEI growth, following Arrhenius-type kinetics in which reaction rates climb exponentially with temperature. For locomotives operating through hot summers or in warm climates, the implication is that thermal management is not merely a comfort feature but a primary determinant of battery lifetime and replacement economics.

Perhaps the most novel contribution of the study is its treatment of spatial nonuniformity. A battery cell is not a point object; it is a three-dimensional structure in which the core, the surfaces, and the tabs can sit at markedly different temperatures during high-current operation. The researchers quantified this with an internal temperature nonuniformity coefficient, and under 4 C cycling the coefficient rose from 2.9 percent to approximately 5.2 percent. That means the hottest regions of the cell were aging measurably faster than the coolest, so the cell effectively contains a population of micro-environments with different degradation rates. Hot spots can seed local lithium depletion, accelerate local SEI growth, and in extreme scenarios contribute to the conditions that precede thermal runaway, the catastrophic self-heating failure mode that battery safety research works so hard to prevent.

Why does this matter for locomotives specifically? Rail traction is among the most punishing duty profiles any battery can face. A new-energy locomotive must deliver megawatt-scale bursts for acceleration, absorb regenerative braking energy, sustain long hauls, and do so across deserts, mountains, and humid subtropics, often with limited opportunity for active cooling. Unlike passenger electric vehicles, locomotives carry enormous battery packs whose replacement cost and downtime are substantial. A model that can predict how a given operating profile, climate, and cooling strategy translate into state of health over thousands of cycles is therefore a genuine engineering tool, not just an academic exercise. The authors position their work explicitly as a modeling basis for thermal management design and lifetime assessment of traction batteries.

The study also fits into a broader movement in battery science toward multi-physics and pseudo-two-dimensional electrochemical models, descendants of the classic Doyle-Fuller-Newman framework, increasingly coupled with aging submodels and, more recently, with machine learning for state estimation and remaining useful life prediction. What distinguishes the present work is its focus on the coupled thermal-aging response inside a single cell under conditions representative of rail service, and its explicit quantification of how C-rate, ambient temperature, and spatial nonuniformity interact. As electrified freight and passenger rail expand worldwide, findings like these, showing that a 4 C duty cycle heats a cell more than twice as fast as a 1 C cycle and that 40-degree operation can nearly triple capacity loss, will help engineers decide how hard a locomotive battery can safely be pushed, and how much cooling it needs to survive a working life measured in decades rather than years.

Subject of Research: Thermal response and SEI-driven aging of lithium-ion traction batteries under varying C-rates and ambient temperatures

Article Title: Thermal response and aging behavior of lithium-ion batteries for new-energy locomotives under complex operating conditions

Article References: Chen, K., Zheng, X., Gao, H., Tang, M., Wang, H., Long, Z., Wu, F., Gao, G., & Wu, G. (2026). Thermal response and aging behavior of lithium-ion batteries for new-energy locomotives under complex operating conditions. Ionics. https://doi.org/10.1007/s11581-026-07525-9

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07525-9

Keywords: lithium-ion battery, battery aging, SEI film growth, thermal management, state of health, C-rate, new-energy locomotive, electrochemical-thermal model, temperature nonuniformity, capacity fade, traction battery, multi-physics modeling

The researchers’ approach centers on electrochemical impedance spectroscopy, a technique that probes a battery by applying small alternating-current signals across a wide range of frequencies. Because different physical processes inside the cell respond at different frequencies, the resulting impedance spectrum acts like a fingerprint of battery aging. Slow degradation mechanisms such as the growth of the solid electrolyte interphase, loss of active material and charge transfer slowdowns each leave distinctive traces in the spectrum, which is why the method has long been favored for state of health estimation. The challenge is that these processes overlap in the frequency domain and that real-world measurements are contaminated by noise, both of which undermine straightforward regression models that try to map raw spectra directly onto battery health.

To overcome this entanglement, Song and Zhao turned to the distribution of relaxation times, or DRT, a mathematical deconvolution technique that transforms an impedance spectrum from the frequency domain into the time constant domain. Instead of overlapping frequency responses, the DRT produces a set of peaks, each corresponding to a distinct relaxation process within the cell. As a battery degrades, the positions, heights and shapes of these peaks shift in physically meaningful ways, effectively giving the model a cleaner and more interpretable set of inputs. The DRT framework has a long pedigree in electrochemistry, and recent work has refined its regularization and discretization, but its integration with modern deep learning architectures for battery diagnostics remains an active frontier.

The new model, named DRT-MSFCNN, is a physics-feature-enhanced multi-scale convolutional neural network. Multi-scale convolutional branches allow the network to examine features of the DRT profile at several resolutions simultaneously, capturing both broad peak structures and fine morphological details that may signal specific aging mechanisms. In parallel, feature-wise linear modulation, or FiLM, modules inject statistical physical features derived from the profiles directly into the network’s internal processing. Rather than forcing the network to rediscover electrochemical structure from scratch, the architecture lets measured physics guide the learning process, a strategy that reduces sensitivity to noise and improves generalization to cells the model has never seen.

Attention modules complete the design, helping the network decide which portions of the DRT profile deserve the greatest weight when estimating health. Ablation experiments, in which components of the model are systematically removed, confirmed that multi-scale convolution, FiLM conditioning and attention each contribute complementary improvements, indicating that the architecture’s strength comes from the combination rather than any single trick.

What sets the evaluation apart is its rigor. The experiments used a publicly available impedance spectroscopy dataset and a battery-level, independent protocol called Leave-One-Battery-Out, or LOBO, in which the model is trained on all batteries except one and then tested on the withheld cell. This simulates the hardest and most realistic deployment scenario, estimating health for a battery whose aging history is entirely unknown. Many published methods are validated by splitting data randomly within the same batteries, a practice that inflates apparent accuracy, so the LOBO protocol provides a more honest benchmark of real-world capability.

The results were striking. DRT-MSFCNN achieved a macro-average root mean square error of 0.74 percent and a mean absolute error of 0.60 percent on the test batteries, meaning that on average its health estimates landed within roughly two-thirds of a percentage point of the true value. It outperformed both traditional machine learning baselines, such as methods built on support vector regression and random forests, and competing deep learning models that worked directly on impedance spectra or simpler feature sets. In a field where fractions of a percent translate into significant differences in how aggressively a battery can be used or when it should be retired, this margin matters.

The implications extend across the battery life cycle. For electric vehicles, accurate health tracking supports more reliable range prediction and safer warranty assessment. For second-life applications, where retired electric vehicle batteries are repurposed for stationary storage, fast and trustworthy health estimation determines whether a cell is worth reusing. Because impedance spectroscopy can in principle be performed quickly and even during operation, as recent operando studies have demonstrated, a robust estimation pipeline built on impedance data could eventually run continuously inside battery management systems rather than requiring lengthy laboratory characterization.

The study also illustrates a broader trend in artificial intelligence for the physical sciences: rather than replacing physical understanding with brute-force learning, the most effective models embed physics into the learning pipeline. By first decoupling the impedance spectrum into interpretable relaxation processes and then feeding those physics-rich profiles to a network designed to exploit their structure, the researchers obtained a system that is both more accurate and arguably more trustworthy than purely data-driven alternatives. The work was supported by the National Natural Science Foundation of China, and the underlying dataset is publicly available, allowing other groups to benchmark against it and accelerate progress.

Challenges remain before such methods reach commercial battery management systems, including the computational cost of DRT deconvolution and the need for impedance hardware on board vehicles, but the direction is clear. As battery fleets grow into the billions of cells, the ability to read a battery’s inner condition through a physics-informed neural lens could reshape how society maintains, reuses and ultimately trusts its energy storage infrastructure.

Subject of Research: Physics-enhanced state of health estimation of lithium-ion batteries using distribution of relaxation times and multi-scale deep learning on electrochemical impedance spectroscopy.

Article Title: Lithium-Ion battery SOH estimation via a physics-feature-enhanced method integrating DRT mechanism and multi-scale deep learning

Article References: Song, C., & Zhao, G. (2026). Lithium-Ion battery SOH estimation via a physics-feature-enhanced method integrating DRT mechanism and multi-scale deep learning. Ionics. https://doi.org/10.1007/s11581-026-07519-7

Image Credits: AI Generated

DOI: 10.1007/s11581-026-07519-7

Keywords: lithium-ion battery, state of health, electrochemical impedance spectroscopy, distribution of relaxation times, deep learning, convolutional neural network, FiLM, battery diagnostics, energy storage, machine learning, battery degradation, Guangxi University