Machine-Learning for Battery Health Diagnosis

Gasper, Paul (ORCID:0000000188349458); Schiek, Andrew (ORCID:0000000321717820); Smith, Kandler (ORCID:0000000170110377); Yoshida, Shuhei; Shimonishi, Yuta

DOE OSTI · 1984841

Machine-Learning for Battery Health Diagnosis

Gasper, Paul (ORCID:0000000188349458)·Schiek, Andrew (ORCID:0000000321717820)·Smith, Kandler (ORCID:0000000170110377)·Yoshida, Shuhei·Shimonishi, Yuta

View original source ↗Read source document ↗Export metadata ↓

Abstract

Battery health diagnosis often requires time-consuming measurements in laboratory conditions, but these types of measurements are unsuitable for use in real-world application such as electric vehicles. Instead, rapid measurements that can be conducted at varying environmental conditions need to be used to monitor battery health. Machine-learning techniques can then be utilized to connect these rapid measurements to health diagnostic information recorded in the lab. This presentation is a part of a tutorial on using machine-learning methods to analyze and predict battery state.

Explore related subjects

battery battery health degradation electrochemical impedance spectroscopy ENERGY STORAGE,MATHEMATICS AND COMPUTING machine-learning

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Gasper, Paul (ORCID:0000000188349458), Schiek, Andrew (ORCID:0000000321717820), Smith, Kandler (ORCID:0000000170110377), Yoshida, Shuhei, Shimonishi, Yuta. 2023-06-07. Machine-Learning for Battery Health Diagnosis. https://www.osti.gov/biblio/1984841

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related reports

DOE OSTI2026

Revealing Local Structures through Machine-Learning-Fused Multimodal Spectroscopy

Atomistic structures of materials offer valuable insights into their functionality. Determining these structures remains a fundamental challenge in materials science, especially for systems with defects. While both experimental and computational methods exist, each has limitations in resolving nanoscale structures. Core-level spectroscopies, such as X-ray absorption (XAS) or electron energy-loss spectroscopies (EELS), have been used to determine the local bonding environment and structure of materials. Recently, machine learning (ML) methods have been applied to extract structural and bonding information from XAS/EELS data. However, frameworks relying solely on a single data stream, defined as characterization data derived from a single element using one technique, are often insufficient because multiple local environments can yield similar spectral features, making it challenging to differentiate between competing structural hypotheses. Here, in this work, we address this challenge by integrating multimodal ab initio simulations, experimental data acquisition, and ML techniques for structure characterization. Our goal is to determine local structures and properties using EELS and XAS data from multiple elements and edges. To showcase our approach, we use various lithium nickel manganese cobalt (NMC) oxide compounds which are used for lithium ion batteries, including those with oxygen vacancies and antisite defects, as the sample material system. We successfully inferred local element content, ranging from lithium to transition metals, with quantitative agreement with experimental data. Beyond local element inference, we find that ML model based on multimodal spectroscopic data is able to determine whether local defects such as oxygen vacancy and antisites are present, a task which is impossible for single mode spectra or other experimental techniques. Furthermore, our framework is able to provide physical interpretability, bridging spectroscopy with the local atomic and electronic structures.

battery↗

DOE OSTI2025

The Solid Electrolyte Interphase Dispersion Can Predict Cycle and Calendar Lifetimes in Silicon Anodes for Lithium-Ion Batteries

The solid electrolyte interphase (SEI) plays a critical role in lithium-ion battery (LIB) anodes. It is responsible for passivating the reactive surface of lithiated anodes against degradation of the electrolyte which enables long cycle and calendar lifetimes for LIBs. This role is especially important in high energy density anodes like silicon, that undergo massive volumetric changes during electrochemical cycling. The mechanism by which the SEI performs this role, however, is not clear which makes designing an SEI to passivate silicon anodes impossible. Through decades of research, dozens of chemical species have been identified within the SEI ranging from inorganic solids to polymeric coatings all of which simultaneously exist in a 'mosaic' composition at the anode surface. This mosaic creates a highly dispersive environment in which electrostatic screening of the anode surface from the electrolyte is not always complete. Here, we present an electrochemical technique to directly test the dispersion at the anode surface. We use this technique to screen more than 20 different electrolytes against silicon anodes and find correlations between both the cycle life and calendar life. The insights from this study offer a new framework to think about the SEI and a rapid screening method to test novel electrode/electrolyte combinations.

battery↗

DOE OSTI2024

Investigation of Nonflammable Electrolytes for Behind-the-Meter Storage Batteries

Behind-the-Meter Storage (BTMS) is a battery-based, stationary energy storage system that is connected to the residential or industrial customer's side of the electrical grid utility service meter. BTMS systems enable consumers to (A) economically schedule charging and usage of stored energy, (B) store and use energy from on-site generation, especially from inconstant, renewable sources like solar and wind, and (C) avoid overloading the grid during peak hours via supplementation with stored energy. BTMS battery performance requirements and general priorities differentiate from other applications, like EVs, which has prompted the development of batteries with tailored electrode and electrolyte materials. These materials prioritize low cost, avoiding critical materials; longevity, achieving 8000 cycle and 20-year shelf lives; and importantly, high safety. Nonflammable electrolytes show promise to improve safety by mitigating thermal runaway, yet often come with sacrifices to battery performance. In this presentation (1) primary categories of nonflammable electrolytes will be discussed; (2) a rational design of experiment will be presented for efficient performance evaluation of several nonflammables electrolyte in BTMS-relevant battery chemistry, Li4Ti5O12- and LiNi0.90Mn0.10O2; and (3) preliminary results will be presented.

battery↗