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

Battery Lab 2.0 includes an impressive 4,000 square feet of research and production space, strategically located on Ellsworth Road in Ann Arbor. This additional space is not only dedicated to academic researchers but is also designed to accommodate the needs of industry researchers, fostering a collaborative environment where pioneering ideas can thrive. The facility boasts a suite of state-of-the-art machinery, including an automatic laser welder that adheres to industry standards for the assembly of battery modules and packs. One of the standout features of Battery Lab 2.0 is its cutting-edge lithium-ion battery production line, which is believed to be the highest-capacity pilot-scale line based at a university, featuring an impressive three megawatt-hours of production capability.

The opening of Battery Lab 2.0 is a response to the rapidly growing demand for advanced battery technologies that have far outpaced the capacity of the original U-M Battery Lab. Alan Taub, the director of the U-M Electric Vehicle Center and a prominent figure in materials science and mechanical engineering, emphasizes that while the original lab has played a crucial role in facilitating battery advancements for electric vehicles, the increasing demand necessitated the expansion. This new facility ensures that researchers have the necessary space and resources to continue innovating and building the technologies that will shape the future.

Researchers at Battery Lab 2.0 have access to a diverse array of capabilities, allowing for the prototyping, testing, and manufacturing of multiple types of batteries, including lithium-ion, solid-state, and next-generation rechargeable batteries. The lab can accommodate both pouch and cylindrical formats, which are essential for applications spanning from grid storage technologies to personal electronics, backup power systems, drones, defense technologies, and, of course, electric vehicles. This versatility is one of the key factors that positions Battery Lab 2.0 as a central hub for battery innovation and research.

Moreover, the unique environment offered by makerspaces like the U-M Battery Lab provides invaluable opportunities for companies lacking in-house labs or manufacturing capabilities. This is particularly beneficial for emerging startups and organizations looking to test their ideas, customize prototypes, and ultimately bring their products to market. Established automotive giants, such as Ford and Mercedes-Benz, as well as innovative startups like MIT-spinout Nanoramic, have leveraged the lab’s resources to enhance their research and development efforts, showcasing the lab’s capacity to cater to a broad spectrum of industry needs.

The response from battery innovators has been overwhelmingly positive. Greg Less, the director of the battery lab, underscores the essential role that Battery Lab 2.0 plays in bridging the so-called “valley of death,” which refers to the challenging period that separates promising concepts from commercial production. By providing the requisite infrastructure, advanced equipment, and specialized expertise necessary to validate new products, the lab enables researchers and companies to conduct thorough testing and prove that their technologies are market-ready.

A notable achievement highlighted by Nanoramic—the maker of advanced battery technologies—speaks to the vital role that the lab has played in advancing their research and development initiatives. The company successfully raised $44 million in a financing round co-led by GM Ventures and participated in by Samsung Ventures, demonstrating how the collaborations fostered within the lab can lead to significant capital acquisition and commercial success. Such collaborations between academia and industry are critical in driving forward the frontier of battery technology.

Since its original opening in 2014 on the University of Michigan’s North Campus, the U-M Battery Lab has been a magnet for researchers from around the world, consistently reaching full booking capacity due to the demand for its vast resources. This $9 million facility’s strategic location is particularly advantageous, providing easy access to automotive industries and facilitating on-site collaboration. This proximity not only benefits external partners but also enriches the academic environment, allowing for the mutual exchange of ideas and innovations.

As Battery Lab 2.0 begins its operation, its state-of-the-art features—including a 4,000-square-foot dry room, an increased mixing volume that extends up to 30 liters, and an advanced coating system—position it firmly at the cutting edge of battery research and development. The ability to produce both 21700 format cylindrical batteries and 100×150 millimeter format pouch batteries enables researchers to work with formats that are representative of modern automotive applications, ensuring that the research conducted is highly applicable to real-world technologies.

The establishment of Battery Lab 2.0 is designed as an interim solution until the construction of a $60 million battery lab on-campus is completed. This future facility will be part of the U-M Transportation Research Institute, with funding secured through a $130 million state contract that established the Electric Vehicle Center. Such investments reflect a formidable commitment from the state of Michigan to support the advancement of electric vehicle technology and battery research.

By situating Battery Lab 2.0 just off the University of Michigan campus, near the southern border of Ann Arbor, the facility enhances accessibility for industrial users, ensuring efficient collaboration between academia and industry. Enhanced access is crucial for stimulating innovation, allowing businesses to tap into academic resources while researchers gain insight from industry practices and needs.

The new equipment featured in Battery Lab 2.0 was manufactured by Energy Tech Solution, known for their expertise in creating advanced battery machinery. This manufacturing partnership underscores the collaborative effort between different stakeholders in the battery research ecosystem, highlighting the intersection of academia, industry, and technology as all work in unison to address the evolving challenges of battery technology.

As the global landscape for battery technology continues to evolve rapidly, the expansion of the U-M Battery Lab into Battery Lab 2.0 signifies a crucial leap forward in the United States’ commitment to being at the forefront of battery innovation. By fostering collaboration between industry and academia, the lab stands poised to spearhead the development of the next generation of battery technologies that will undoubtedly serve as the backbone for future electric vehicles and a broad range of other applications.

The establishment of this cutting-edge facility serves as a powerful statement regarding the importance of research and development in ensuring that the United States remains a leader in advanced battery technology. The University of Michigan’s Battery Lab 2.0 not only embodies a significant investment in infrastructure and talent but also represents a collaborative spirit that will likely yield transformative outcomes for the automotive industry and beyond. As we look toward the future, the innovations emerging from this lab will surely play a pivotal role in shaping our sustainable energy landscape.

Subject of Research: Next-generation battery technology
Article Title: The University of Michigan’s Battery Lab 2.0: Pioneering Advanced Battery Research
News Publication Date: October 2023
Web References: https://umbatterylab.engin.umich.edu/
References: Various sources on U-M Battery Lab and academic publications
Image Credits: University of Michigan Battery Lab Photos