Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “battery aging”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
NASA NTRS2017

Effects of Acute Exposures to Carbon Dioxide upon Cognitive Functions

Carbon dioxide (CO2) originates from human metabolism and typically remains about 10-fold higher in concentration on the International Space Station (ISS) than at the earth's surface. There have been recurring complaints by crew members of episodes of "mental viscosity" adversely affecting their performance, and there is evidence from the ISS that associates CO2 levels with reports of headaches by crewmembers. Consequently, flight rules have been employed to control CO2 below 3 mm Hg, which is well below the existing Spacecraft Maximum Allowable Concentration (SMAC) of 10 mm Hg for 24-hour exposures, and 5.3 mm Hg for exposures of 7 to 180 days. Headaches, while sometime debilitating themselves, are also symptoms that can provide evidence that physiological defense mechanisms have been breached, and there is evidence that CO2 has effects at levels below the threshold for headaches. This concern appears to be substantiated in reports that CO2 at concentrations below 2 mm Hg substantially reduced some cognitive functions that are associated with the ability to make complex decisions in conditions that are characterized by volatility, uncertainty, complexity, ambiguity, and delayed feedback. These are conditions that could be encountered by crews in off-nominal situations or during the first missions beyond low earth orbit. Therefore, we set out to determine if decision-making under volatile, uncertain, confusing and ambiguous circumstances, where feedback is delayed or absent, is correlated with low levels of CO2 during acute exposures (several hours) in crew-like subjects and to determine if additional cognitive domains are sensitive to concentrations of CO2 at, or below, current ISS levels by using a test battery that is currently available onboard ISS. We enrolled 22 volunteers (8 females, 14 males) between the ages of 30-55 (38.8 +/- 7.0) years whose training and professional experience reflect that of the astronaut corps. Subjects were divided among 4 study groups. Exposures occurred in the 20-foot environmental chamber in Bldg 7 at JSC. Each group was exposed to each of four concentrations of CO2 (600, 1200, 2500, and 5000 ppm), and the exposure order was balanced across the groups which were randomly assigned to 4 possible complete exposure orders. Study participants and investigators were blinded to the exposure order. Each exposure lasted 4 hours and occurred on the same day each week for 4 consecutive weeks. Testing included the Strategic Management Simulation (SMS) methodology and the Cognition battery of psychometric measures that are being utilized aboard the ISS. Subjects participated in a familiarization session one week prior to the start of exposure sessions. Each exposure session followed the schedule noted in Figure 1 below. On the morning of each exposure, subjects completed the Cognition battery in a conference room at indoor ambient CO2 levels. Subjects then entered the exposure chamber and were acclimated to the environment for 15 minutes before completing another round of the Cognition battery. After a 20 minute rest, subject completed the approx.80 minute SMS. Subjects were allowed another 20 minute rest period before completing the final Cognition test battery in the exposure chamber. Subjects then returned to the conference room where they completed one final round of Cognition. Data from both testing methodologies are currently being analyzed and will be presented. Study investigators have not yet been unblinded to the coded exposure concentrations, but preliminary results suggest differences.

Ryder, V. E.↗
DOE OSTI2021

Early Battery Performance Prediction for Mixed Use Charging Profiles Using Hierarchal Machine Learning

A key step limiting how fast batteries can be deployed is the time necessary to provide evaluation and validation of performance. Using data analysis approaches, such as machine learning, the validation process can be accelerated. However, questions on the validity of projecting models trained on limited data or simple cycling profiles, such as constant current cycling, to real-world scenarios with complex loads remains. Here, we present the ability to predict performance with less than 1.2% mean absolute percent error when trained on cells aged using complex electric vehicle discharge profiles, and either AC Level 2 charge or DC Fast charge profiles, using only the first 45 cycles, namely 5% of the total testing time. While error is low across the projections, this study also highlights that battery lifetime analysis using only cycling data may not extrapolate safely to certain real-world conditions due to the impact of calendar degradation.

25 ENERGY STORAGE↗
NASA NTRS1994

Nickel cadmium battery operations and performance

The Earth Radiation Budget Satellite (ERBS), Compton Gamma Ray Observatory (CGRO), Upper Atmosphere Research Satellite (UARS), and Extreme Ultraviolet Explorer (EUVE) spacecraft are operated from NASA's Goddard Space Flight Center (GSFC) in Greenbelt, Maryland. On-board power subsystems for each satellite employ NASA Standard 50 Ampere-hour (Ah) nickel-cadmium batteries in a parallel configuration. To date, these batteries have exhibited degradation over periods from several months (anomalous behavior, UARS and CGRO (MPS-1); to little if any, EUVE) to several years (old age, normal behavior, ERBS). Since the onset of degraded performance, each mission's Flight Operations Team (FOT), under the direction of their cognizant GSFC Project Personnel and Space Power Application Branch's Engineers has closely monitored the battery performance and implemented several charge control schemes in an effort to extend battery life. Various software and hardware solutions have been developed to minimize battery overcharge. Each of the four sections of this paper covers a brief overview of each mission's operational battery management and its associated spacecraft battery performance. Also included are new operational procedures developed on-orbit that may be of special interest to future mission definition and development.

Rao, Gopalakrishna↗
DOE OSTI2021

PyBLM (Python-based Battery Life Model) [SWR-21-50]

PyBLM is a battery life model to predict the capacity fade of two home battery energy storage systems, manufactured by LG and Tesla. The battery life model is built from experimental test data, and accounts for calendar and cycling aging mechanisms as functions of cell environmental conditions and use. PyBLM is developed in Python and formatted to work in conjunction with another NREL software named TEMPEST (Thermo-Electric Model for Powering Energy Storage Technologies).

Mishra, Partha↗
NASA NTRS1988

Temperature Pill

Ingestible Thermal Monitoring System was developed at Johns Hopkins University as means of getting internal temperature readings for treatments of such emergency conditions as dangerously low (hypothermia) and dangerously high (hyperthermia) body temperatures. ITMS's accuracy is off no more than one hundredth of a degree and provides the only means of obtaining deep body temperature. System has additional applicability in fertility monitoring and some aspects of surgery, critical care obstetrics, metabolic disease treatment, gerontology (aging) and food processing research. Three-quarter inch silicone capsule contains telemetry system, micro battery, and a quartz crystal temperature sensor inserted vaginally, rectally, or swallowed.

Source record↗
NASA NTRS2018

The Mars Reconnaissance Orbiter Mission: 2018 Status

The Mars Reconnaissance Orbiter (MRO) continues its dual-mode mission of scientific exploration and programmatic support. Now in its 13th year of flight (and its 12th year in its low-altitude, sun-synchronous orbit around Mars), the spacecraft continues in its normal flight mode and its science instruments continue to acquire a treasure trove of data which is being used to study the Martian environment (surface, sub-surface, and atmosphere). To date (July 2018), more than 333 TB of scientific data have been returned to Earth. Building onto the 2017 IAC paper: “The Mars Reconnaissance Orbiter Mission: Continuing a Record of Exploration from Mars Orbit,” [1] this paper provides a 2018 status update on the MRO Mission. A signature event since the 2017 writing is the onset and evolution of a planet-encircling dust event (PEDE), the first such event since 2007. Additional recent science highlights for MRO include the discovery of mid-latitude ice cliffs, polar cap characterization, daily global mapping of the 2018 PEDE dust activity and the atmospheric thermal response, further characterization of the tantalizing recurring slope lineae (RSL), and continued detection of surface changes. Programmatically, UHF relay support for the Mars Science Laboratory (MSL) rover Curiosity and the Mars Exploration Rover (MER) rover Opportunity continues at a pace that provides the engineers and scientists operating those surface vehicles timely information to support their mission planning and science analysis. Most of the landing site characterization requests for the NASA Mars 2020 rover and the ESA ExoMars Rover/Surface Platform (RSP) missions have been completed, with final acquisitions delayed due to the obscuring dust haze from the current PEDE. In parallel, the flight team has faced new engineering challenges as the spacecraft has aged. The team has developed and is implementing actions that are aimed at stretching spacecraft battery life. A new all-stellar capability using the spacecraft’s star trackers now allows for normal spacecraft operations without IMUs. In addition to its current relay duties, MRO will be synchronized to provide critical communications support for the InSight Entry, Descent, and Landing (EDL) event on November 26, 2018, and will be readied to support InSight’s critical commissioning phase with UHF relay support and environmental characterization. A new split-pass relay capability has been developed which will allow MRO to relay with two surface vehicles in “close proximity” on the same overflight, i.e. the InSight lander and the Curiosity rover. Extended for another year of operations by NASA, this paper will highlight recent scientific progress and describe actions that will extend MRO’s life well into the 2020’s as a key element of the Mars Exploration Program.

Zurek, Richard W.↗
DOE OSTI2025

The Origin of Improved Performance in Boron‐Alloyed Silicon Nanoparticle‐Based Anodes for Lithium‐Ion Batteries

Stabilizing the solid electrolyte interphase (SEI) remains a key challenge for silicon‐based lithium‐ion battery anodes. Alloying silicon with secondary elements like boron has emerged as a promising strategy to improve the cycle life of silicon anodes, yet the underlying mechanism remains unclear. To address this knowledge gap, how boron concentration influences battery performance is systematically investigated. These results show a near‐monotonic increase in cycle lifetime with higher boron content, with boron‐rich electrodes significantly outperforming pure silicon. Additionally, silicon‐boron alloy anodes exhibit nearly three times longer calendar life than pure silicon. Through detailed mechanistic analysis, alternative contributing factors are systematically ruled out, and it is proposed that improved passivation arises from a strong permanent dipole at the nanoparticle surface. This dipole, formed by undercoordinated and highly Lewis acidic boron, creates a static, ion‐dense layer that stabilizes the electrochemical interface, reducing parasitic electrolyte decomposition and enhancing long‐term stability. These findings suggest that, within the SEI framework, the electric double layer is an important consideration in surface passivation. This insight provides an underexplored parameter space for optimizing silicon anodes in next‐generation lithium‐ion batteries.

25 ENERGY STORAGE↗
DOE OSTI2022

Anodic Interfacial Evolution in Extremely Fast Charged Lithium-Ion Batteries

Interfacial reaction mechanisms at the anode/separator interface play a central role in the performance and safety of lithium-ion batteries during fast charging. We report a mechanistic study on the evolution and interactions of the aging mechanisms at the anode/separator interface in lithium cobalt oxide/graphite pouch cells charged with variable charging rates (1–6C) over 10 cycles. In situ electrochemical measurements, including voltage relaxation, Coulombic efficiency, and direct current internal resistance, indicated an incremental lithium loss until the C rates were ≤5C. A substantial capacity fade is observed in the first few cycles of fast charging, but the magnitude of capacity fade progressively diminishes with the number of cycles, indicating a suppression in the lithium deposition mechanism. Post-mortem film thickness, X-ray photoelectron spectroscopy (XPS), and scanning electron microscopy (SEM) analyses were performed to elucidate the evolution of electrolyte decomposition, the solid–electrolyte interface (SEI), lithium plating, and film fracture mechanisms with C rate. XPS measurements confirmed an increasing lithium concentration in an SEI film with an increase in the C rate. SEM images showed a growth of dendritic lithium on the anode surface from 1C to 3C. Precrack formation leading to an interfacial film fracture was observed at higher C rates. Furthermore, a differential analysis of the discharge capacity indicated a possible two-phase delithiation from the anode and reduced cathodic lithiation due to lithium loss at high C rates.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Complementary Electrolyte Design for Li Metal Batteries in Electric Vehicle Applications

In this report a complementary electrolyte system with 0.8 M lithium bis(fluorosulfonylimide) (LiFSI) salt and 2 wt % lithium perchlorate (LiCLO4) additive in fluoroethylene carbonate (FEC)/ethyl methyl carbonate (EMC) solution enables not only stable cycling of lithium metal batteries (LMBs) with practical loading (<30 μm lithium anode, cathode loading > 4 mAh/cm 2 ) but also outstanding degradation stability toward the end of cycle life when compared to the conventional electrolyte. Although the use of LiFSI salt can increase the electrolyte conductivity and lengthen the cycle life of LMBs, the aged lithium anode morphology formed by the sacrificial decomposition of LiFSI is highly porous, leading to an abrupt cell capacity drop toward the end of cycling. Moreover, the inability to stop aluminum corrosion by the LiFSI-based electrolyte also causes cracking of the cathode tab during prolonged cycling. It is observed that a highly porous aged lithium consumed electrolyte at a higher rate, leading to the dry-out of electrolyte solvents. On the contrary, dense aged lithium anode morphology increased the localized current applied on the lithium, causing the formation of lithium dendrite. Thus, porosity control is the key to enhance battery performance. In this complementary system, LiClO 4 was introduced as an advanced additive to not only improve the capacity retention rate but also mitigate the abrupt capacity drop toward the end of cycle life because LiClO 4 acted as a pore astringent reducing the porosity of the aged lithium metal anode to the desired level. Moreover, the addition of LiClO 4 can also suppress the AI corrosion, allowing stable high-voltage cycling of LMBs. The synergistic effect of combining LiFSI salt and a LiClO 4 additive leads to an electrolyte system that can facilitate the application of high-energy LMBs with practical electrode loading.

25 ENERGY STORAGE↗
DOE OSTI2025

Lithium-Ion Batteries with Lithium Manganese Iron Phosphate Cathodes and Lithium Titanate Anodes: Linking Electrode Dynamics to Cell Performance

This study examines the electrochemical performance, impedance behavior, and aging mechanisms of lithium-ion cells pairing lithium manganese iron phosphate (LiMn 0.6 Fe 0.4 PO 4 , LMFP64) cathodes with lithium titanate (Li 4 Ti 5 O 12 , LTO) anodes. LMFP64 half-cells display distinct Fe and Mn redox plateaus at ∼3.5 and ∼4.1 V, respectively, and deliver excellent rate capability. Three-electrode measurements show that LMFP64 dominates cell impedance, which changes sharply across the Fe–Mn redox transition and decreases with increasing temperature (30 °C–50 °C). Voltage hysteresis arises mainly from the cathode, with higher C-rates shortening redox plateaus and broadening the transition region. Elevated temperature lowers impedance, especially near the Mn-redox and transition regions, but accelerates capacity fade. Post-cycling analyses of electrodes confirm structural stability of LMFP64 and indicate loss of lithium inventory (LLI) as the source of capacity fade. This LLI causes potential slippage between the LMFP64 and LTO electrodes, which decreases the Fe-redox plateau width. Long-term cycling highlights the critical role of electrode overhang: cathode overhang mitigates capacity fade by buffering lithium loss, whereas anode overhang cells experience faster capacity decline due to uncompensated lithium depletion. These results clarify LMFP64/LTO battery impedance evolution, thermal effects, and degradation pathways and provide insights for their applications in electric vehicles.

Area Specific Impedance↗
DOE OSTI2022

Extreme fast charge aging: Effect of electrode loading and NMC composition on inhomogeneous degradation in graphite bulk and electrode/electrolyte interface

Empowering extreme fast charging (XFC) requires a comprehensive understanding of its application with advanced anode and cathode materials in lithium-ion batteries. No report exists for the full extent of limitations for the anode with crosstalk effect from paired cathode as well as Li plating due to electrode loading under XFC. In this study, a combination of cell testing and multiple length characterization is used to investigate XFC aging mechanism in cells with a low loading of 1.5 mAh cm -2 and high loading of 2.5 mAh cm -2 for graphite (Gr)/Ni-rich LiNi x Mn y Co 1-x-y O 2 (NMCs). Operando XRD mappings show 1.5 mAh cm -2 loadings result in higher strain in graphite for all three cathode types. Among the three NMC cathodes, the graphite from NMC532 and NMC811 cells show comparable strain. Scanning electron microscopy (SEM) images show distinct differences between 6-C-charged anodes in two loadings. Significantly increased electrode thickness can be seen due to more damage in the graphite bulk and accumulation of the electrolyte decomposition products in electrode pores. X-ray photoelectron spectroscopy (XPS) reveals both cathode chemistry and Li plating influence the non-uniform SEI composition on graphite surface. We report higher Ni content in NMC811 promotes the higher levels of salt decomposition on the SEI and formation of higher mass of electrolyte aging products.

25 ENERGY STORAGE↗
DOE OSTI2020

Improving Electronic Conductivity of Layered Oxides through the Formation of Two-Dimensional Heterointerface for Intercalation Batteries

Synthetic strategies for the improvement in electronic conductivities and electrochemical stabilities of transition metal oxide cathodes, which are limiting factors in the performance of commercial intercalation batteries, are required for next-generation, high-performance battery systems. Furthermore, the chemical preintercalation approach, consisting of a combined sequence of a sol–gel process, extended aging, and a hydrothermal treatment, is a versatile, wet synthesis technique that allows for the incorporation of a polar species between the layers of transition metal oxides. Here, formation of a layered 2D δ-C x V 2 O 5 · n H 2 O heterostructure occurs via chemical preintercalation of dopamine molecules between bilayers of vanadium oxide followed by the hydrothermal treatment of the precipitate, leading to carbonization of the organic molecules. The presence of carbon layers within the structure has been confirmed via a combined analysis of scanning electron microscopy, X-ray diffraction, thermogravimetric analysis, Raman spectroscopy, X-ray photoelectron spectroscopy, electrochemical impedance spectroscopy, four-probe conductivity measurements, and scanning transmission electron microscopy characterization. 2D δ-C x V 2 O 5 · n H 2 O heterostructure electrodes demonstrated significantly improved electrochemical performance, particularly at higher current densities, in Li-ion cells. The heterostructure electrodes exhibited 75% of the capacity retention when the current was changed from 20 mA g –1 (206 mAh g –1 ) to 300 mA g –1 (155 mAh g –1 ), while the reference δ-V 2 O 5 · n H 2 O electrodes exhibited only 10% capacity retention in the same experiment. Remarkably, 2D δ-C x V 2 O 5 · n H 2 O heterostructure electrodes demonstrated significantly improved capacity retention (94% after 30 cycles) for bilayered vanadium oxide electrodes in Li-ion cells during galvanostatic cycling at 20 mA g –1 . The improved electrochemical performance, in both extended cycling and rate capability studies, of the 2D δ-C x V 2 O 5 · n H 2 O heterostructure electrodes in the Li-ion system is ascribed to the intermittent formation of carbon layers within the bilayered structure, which leads to increased electronic conductivity and improved structural stability of the heterostructure compared to the reference δ-V 2 O 5 · n H 2 O electrodes.

25 ENERGY STORAGE↗
DOE OSTI2021

Structure and dynamics of small polyimide oligomers with silicon as a function of aging

The effect of UV curing and shearing on the structure and behavior of a polyimide (PI) binder as it disperses silicon particles in a battery electrode slurry was investigated. PI dispersant effectiveness increases with UV curing time, which controls the overall binder molecular weight. Here, the shear force during electrode casting causes higher molecular weight PI to agglomerate, resulting in battery anodes with poorly dispersed Si particles that do not cycle well. It is hypothesized that when PI binder is added above a critical amount, it conformally coats the silicon particles and greatly impedes Li ion transport. There is an “interzonal region” for binder loading where it disperses silicon well and provides a coverage that facilitates Li transport through the anode material and into the silicon particles. These results have implications in ensuring reproducible electrode manufacturing and increasing cell performance by optimizing the PI structure and coordination with the silicon precursor.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Performance and lifetime of intercalative water deionization cells for mono- and divalent ion removal

We report intercalative deionization (IDI) uses two cation intercalation electrodes separated by an anion exchange membrane in a symmetric cell design that has the potential to deliver electrochemically desalinated water in an energy- and water-efficient way. Here, we define and measure metrics to describe the performance and lifetime of IDI cells and compare them for NaCl and CaCl 2 feed solutions. With 20 mM NaCl, NiHCF/AEM/NiHCF flow cells achieve 10 mM average concentration change at a productivity of 20 l/h/m 2 and 5 mM average concentration change at 130 l/h/m 2 . In both cases the cells are operated at a 3C current rate and consume ~30 Wh/m 3 of energy. With 10 mM CaCl 2 , the specific capacity and salt removal of IDI flow cells is ~4 times lower. NiHCF/NiHCF beaker cells with CaCl 2 electrolyte suffer from strong capacity fade, while the same cells with NaCl electrolyte achieve 500 cycles without any capacity fade. Our post-mortem analysis using X-ray diffraction, secondary electron microscopy, energy dispersive X-ray spectroscopy, synchrotron-based X-ray absorption spectroscopy and micro X-ray fluorescence mapping reveals that NiHCF dissolves upon repeated intercalation with Ca 2+ , releasing residual K + , Ni 2+ and Fe(CN) 6 3 , which precipitates as a crystalline decomposition product on the electrodes. This side reaction deprives the active material NiHCF of charge compensating Fe, and thus accounts for the observed capacity fade.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS2011

Portable Multigas Monitors for International Space Station

The Environmental Health System (EHS) on International Space Station (ISS) includes portable instruments to measure various cabin gases that acutely impact crew health. These hand-held devices measure oxygen, carbon dioxide, carbon monoxide, hydrogen chloride and hydrogen cyanide. The oxygen and carbon dioxide units also serve to back up key functions of the Major Constituent Analyzers. Wherever possible, commercial off-the-shelf (COTS) devices are employed by EHS to save development and sustaining costs. COTS hardware designed for general terrestrial applications however has limitations such as no pressure compensation, limited life of the active sensor, calibration drift, battery issues, unpredictable vendor support and obsolescence. The EHS fleet (inflight and ground inventory) of instruments is both aging and dwindling in number. With the retirement of the US Space Shuttle, maintenance of on-orbit equipment becomes all the more difficult. A project is underway to search for gas monitoring technology that is highly reliable and stable for years. Tunable Diode Laser Spectroscopy (TDLS) seems to be the front-runner technology, but generally is not yet commercially available in portable form. NASA has fostered the development of TDLS through the Small Business Innovative Research (SBIR) program. A number of gases of interest to the aerospace and submarine communities can be addressed by TDLS including the list mentioned above plus hydrogen fluoride, ammonia and water (humidity). There are several different forms of TDLS including photoacoustic and direct absorption spectroscopy using various multipass cell geometries. This paper describes the history of portable gas monitoring on NASA spacecraft and provides a status of the development of TDLS based instruments. Planned TDLS flight experiments on ISS could lead both to operational use on ISS and important roles in future Exploration spacecraft and habitats.

Mudgett, Paul D.↗
DOE OSTI2024

Approximate Dynamic Programming With Enhanced Off-Policy Learning for Coordinating Distributed Energy Resources

Herein this paper proposes an innovative approximate dynamic programming (ADP) method for distributed energy resource coordination with the loss of life of battery energy storage system (BESS) explicitly modeled. The dispatch policy is designed to account for both calendrical and cyclical aging effects on BESS, explicitly modeling the impacts of ambient temperature on BESS lifespan. The proposed ADP employs an adaptive critic method and enhanced off-policy deterministic policy gradient (DPG) strategy, addressing the limitations of the on-policy gradient-based ADP approaches, including inadequate exploration, low data usage, and computational complexity. In particular, a customized policy is proposed to guide the algorithm to explore some promising decisions and thereby improve exploration capability and learning efficiency compared to conventional DPG-based learning approaches, which may struggle to find a global optimum due to random noisy action-based exploration or require expert demonstration with extra effort. The proposed method is illustrated using the IEEE 123-node system and compared with the existing ADP methods to prove solution accuracy and demonstrate the effects of incorporating degradation models into control design. Case studies showed that the proposed ADP effectively coordinates DERs with a 10 times smaller optimization gap compared to existing methods, and the incorporation of the BESS life loss model ensures the expected lifespan and results in significant cost savings.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Early calendar life and health prediction of silicon batteries via machine learning with uncertainty quantification

Lithium-ion batteries with silicon anodes promise high energy density but are limited by calendar lifetime. Reducing the long iteration time to obtain experimental results requires predicting calendar lifetime early in a cell's life. In this study, we demonstrate that lightweight machine learning models with feature engineering can provide calendar lifetime estimates from early electrochemical signals. After 1 month of electrochemical aging, the best models achieve 10% error in calendar-life prediction and can separate "bad" from "good" lifetime cells with a mean F1 score of 0.857. As battery systems exhibit inherent variability, four methods for uncertainty quantification are compared, and confidence intervals are demonstrated with an uncertainty of +-3.6 months in lifetime prediction. A feature importance analysis indicates that early patterns in voltage decay are the strongest indicators of calendar lifetime. Finally, this modeling approach has high error when generalizing to new electrode chemistries or testing conditions but with appropriately low confidence.

25 ENERGY STORAGE↗