Bayesian learning for rapid prediction of lithium-ion battery-cycling protocols

نویسندگان

چکیده

Summary Advancing lithium-ion battery technology requires the optimization of cycling protocols. A new data-driven methodology is demonstrated for rapid, accurate prediction cycle life obtained by protocols using a single test lasting only 3 cycles, enabling rapid exploration protocol design spaces with orders magnitude reduction in testing time. We achieve this combining lifetime early hierarchical Bayesian model (HBM) to rapidly predict performance distributions without need extensive repetitive testing. The applied comprehensive dataset lithium-iron-phosphate/graphite comprising 29 different fast-charging HBM alone provides high protocol-lifetime performance, 6.5% overall average percent error, after one failure. By model, we error 8.8% 3-cycle test. In addition, generalizability approach lithium-manganese-cobalt-oxide/graphite cells.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Development of Lifetime Prediction Model of Lithium-Ion Battery Based on Minimizing Prediction Errors of Cycling and Operational Time Degradation Using Genetic Algorithm

Accurate lifetime prediction of lithium-ion batteries is a great challenge for the researchers and engineers involved in battery applications in electric vehicles and satellites.  In this study, a semi-empirical model is introduced to predict the capacity loss of lithium-ion batteries as a function of charge and discharge cycles, operational time, and temperature. The model parameters are obtai...

متن کامل

Electrochemical Characterization of Low-Cost Lithium-Iron Orthosilicate Samples as Cathode Materials of Lithium-Ion Battery

Lithium-iron-orthosilicate is one of the most promising cathode materials for Li-ion batteries due to its safety, environmental brightness and potentially low cost. In order to produce a low cost cathode material, Li2FeSiO4/C samples are synthesized via sol-gel (SG; one sample) and solid state (SS; two samples with different carbon content), starting from Fe (III) in the raw materials (lo...

متن کامل

Numerical investigation of the parameters of a prismatic lithium ion battery under load for electrical vehicle

Electric vehicles and hybrid electric vehicles are a suitable alternative for vehicles with hydrocarbons fuels to reduce pollution and fossil resources. The batteries operate as the driving force for these vehicles. One of the most critical parameters of the battery is computing the state of charge (SOC). The best range for SOC of lithium-ion battery is between 20% and 90%, and charging and dis...

متن کامل

Micrometric Growth of V2O5Hexagonal Nano-plates as an Active Material for Lithium Ion Battery Cathode Electrode

This manuscript reports the synthesis of V2O5 nanostructures using reflux method, without using additives such as surface reactants. The influence of reaction parameters like temperature and concentration on the growth of nanostructures have been investigated. It has been observed that the nanostructures are formed with a hexagonal nano-plate morphology, grown from a common core. The diameter o...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Joule

سال: 2021

ISSN: ['2542-4351', '2542-4785']

DOI: https://doi.org/10.1016/j.joule.2021.10.010