A Bayesian nonlinear random effects model for identification of defective batteries from lot samples

Edward Cripps, Michael Pecht

    Research output: Contribution to journalArticle

    9 Citations (Scopus)

    Abstract

    Numerous materials and processes go into the manufacture of lithium-ion batteries, resulting in variations across batteries' capacity fade measurements. Accounting for this variability is essential when determining whether batteries are performing satisfactorily. Motivated by a real manufacturing problem, this article presents an approach to assess whether lithium-ion batteries from a production lot are not representative of a healthy population of batteries from earlier production lots, and to determine, based on capacity fade data, the earliest stage (in terms of cycles) that battery anomalies can be identified. The approach involves the use of a double exponential function to describe nonlinear capacity fade data. To capture the variability of repeated measurements on a number of individual batteries, the double exponential function is then embedded as the individual batteries' trajectories in a Bayesian random effects model. The model allows for probabilistic predictions of capacity fading not only at the underlying mean process level but also at the individual battery level. The results show good predictive coverage for individual batteries and demonstrate that, for our data, non-healthy lithium-ion batteries can be identified in as few as 50 cycles.

    Original languageEnglish
    Pages (from-to)342-350
    Number of pages9
    JournalJournal of Power Sources
    Volume342
    DOIs
    Publication statusPublished - 28 Feb 2017

    Fingerprint

    electric batteries
    Identification (control systems)
    Exponential functions
    Trajectories
    lithium
    exponential functions
    Lithium-ion batteries
    ions
    cycles
    fading
    manufacturing
    trajectories
    anomalies
    predictions

    Cite this

    @article{1305a6397da14ed9986a597e639188f2,
    title = "A Bayesian nonlinear random effects model for identification of defective batteries from lot samples",
    abstract = "Numerous materials and processes go into the manufacture of lithium-ion batteries, resulting in variations across batteries' capacity fade measurements. Accounting for this variability is essential when determining whether batteries are performing satisfactorily. Motivated by a real manufacturing problem, this article presents an approach to assess whether lithium-ion batteries from a production lot are not representative of a healthy population of batteries from earlier production lots, and to determine, based on capacity fade data, the earliest stage (in terms of cycles) that battery anomalies can be identified. The approach involves the use of a double exponential function to describe nonlinear capacity fade data. To capture the variability of repeated measurements on a number of individual batteries, the double exponential function is then embedded as the individual batteries' trajectories in a Bayesian random effects model. The model allows for probabilistic predictions of capacity fading not only at the underlying mean process level but also at the individual battery level. The results show good predictive coverage for individual batteries and demonstrate that, for our data, non-healthy lithium-ion batteries can be identified in as few as 50 cycles.",
    keywords = "Bayesian statistics, Capacity fade, Lithium-ion battery, Markov chain Monte Carlo, Nonlinear random effects",
    author = "Edward Cripps and Michael Pecht",
    year = "2017",
    month = "2",
    day = "28",
    doi = "10.1016/j.jpowsour.2016.12.067",
    language = "English",
    volume = "342",
    pages = "342--350",
    journal = "Journal of Power Sources",
    issn = "0378-7753",
    publisher = "Pergamon",

    }

    A Bayesian nonlinear random effects model for identification of defective batteries from lot samples. / Cripps, Edward; Pecht, Michael.

    In: Journal of Power Sources, Vol. 342, 28.02.2017, p. 342-350.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - A Bayesian nonlinear random effects model for identification of defective batteries from lot samples

    AU - Cripps, Edward

    AU - Pecht, Michael

    PY - 2017/2/28

    Y1 - 2017/2/28

    N2 - Numerous materials and processes go into the manufacture of lithium-ion batteries, resulting in variations across batteries' capacity fade measurements. Accounting for this variability is essential when determining whether batteries are performing satisfactorily. Motivated by a real manufacturing problem, this article presents an approach to assess whether lithium-ion batteries from a production lot are not representative of a healthy population of batteries from earlier production lots, and to determine, based on capacity fade data, the earliest stage (in terms of cycles) that battery anomalies can be identified. The approach involves the use of a double exponential function to describe nonlinear capacity fade data. To capture the variability of repeated measurements on a number of individual batteries, the double exponential function is then embedded as the individual batteries' trajectories in a Bayesian random effects model. The model allows for probabilistic predictions of capacity fading not only at the underlying mean process level but also at the individual battery level. The results show good predictive coverage for individual batteries and demonstrate that, for our data, non-healthy lithium-ion batteries can be identified in as few as 50 cycles.

    AB - Numerous materials and processes go into the manufacture of lithium-ion batteries, resulting in variations across batteries' capacity fade measurements. Accounting for this variability is essential when determining whether batteries are performing satisfactorily. Motivated by a real manufacturing problem, this article presents an approach to assess whether lithium-ion batteries from a production lot are not representative of a healthy population of batteries from earlier production lots, and to determine, based on capacity fade data, the earliest stage (in terms of cycles) that battery anomalies can be identified. The approach involves the use of a double exponential function to describe nonlinear capacity fade data. To capture the variability of repeated measurements on a number of individual batteries, the double exponential function is then embedded as the individual batteries' trajectories in a Bayesian random effects model. The model allows for probabilistic predictions of capacity fading not only at the underlying mean process level but also at the individual battery level. The results show good predictive coverage for individual batteries and demonstrate that, for our data, non-healthy lithium-ion batteries can be identified in as few as 50 cycles.

    KW - Bayesian statistics

    KW - Capacity fade

    KW - Lithium-ion battery

    KW - Markov chain Monte Carlo

    KW - Nonlinear random effects

    UR - http://www.scopus.com/inward/record.url?scp=85007003377&partnerID=8YFLogxK

    U2 - 10.1016/j.jpowsour.2016.12.067

    DO - 10.1016/j.jpowsour.2016.12.067

    M3 - Article

    VL - 342

    SP - 342

    EP - 350

    JO - Journal of Power Sources

    JF - Journal of Power Sources

    SN - 0378-7753

    ER -