mlsanityData to 7 Oct 2026

Machine learning pipeline for battery state of health estimation

arXiv 2102.00837 ↗First posted 1 Feb 2021cs.LGcited by 28

Not on the map. Why?

Abstract (arXiv)

Lithium-ion batteries are ubiquitous in modern day applications ranging from portable electronics to electric vehicles. Irrespective of the application, reliable real-time estimation of battery state of health (SOH) by on-board computers is crucial to the safe operation of the battery, ultimately safeguarding asset integrity. In this paper, we design and evaluate a machine learning pipeline for estimation of battery capacity fade - a metric of battery health - on 179 cells cycled under various conditions. The pipeline estimates battery SOH with an associated confidence interval by using two parametric and two non-parametric algorithms. Using segments of charge voltage and current curves, the pipeline engineers 30 features, performs automatic feature selection and calibrates the algorithms. When deployed on cells operated under the fast-charging protocol, the best model achieves a root mean squared percent error of 0.45%. This work provides insights into the design of scalable data-driven models for battery SOH estimation, emphasising the value of confidence bounds around the prediction. The pipeline methodology combines experimental data with machine learning modelling and can be generalized to other critical components that require real-time estimation of SOH.

Key facts. Cited by 28 papers in the map (data through 7 Oct 2026). Builds on 9 of its 90 references that are in the map. First posted on arXiv 1 Feb 2021.

Builds on

9 of 90 references.

References · 9

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  3. 2018Accurate Uncertainties for Deep Learning Using Calibrated Regressioncited by 417
  4. 2018Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learningcited by 219
  5. 2017EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examplescited by 154
  6. 2013Confidence Intervals for Random Forests: The Jackknife and the Infinitesimal Jackknifecited by 43
  7. 2013Correlation and variable importance in random forestscited by 41
  8. 2017Attacking the Madry Defense Model with L1-based Adversarial Examplescited by 21
  9. 2014Why walking the dog takes time: Frechet distance has no strongly subquadratic algorithms unless SETH failscited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 4; 2023: 8; 2024: 4; 2025: 7; 2026: 5.

28 papers.

Sort:

By area:

  1. 2023Uncertainty Quantification in Machine Learning for Engineering Design and Health Prognostics: A Tutorialcited by 27
  2. 2023Distribution-free risk assessment of regression-based machine learning algorithmscited by 6
  3. 2022Dynaformer: A Deep Learning Model for Ageing-aware Battery Discharge Predictioncited by 4
  4. 2025PulseBat: A field-accessible dataset for second-life battery diagnostics from realistic histories using multidimensional rapid pulse testcited by 4
  5. 2025BatteryLife: A Comprehensive Dataset and Benchmark for Battery Life Predictioncited by 4
  6. 2022Evaluating feasibility of batteries for second-life applications using machine learningcited by 3
  7. 2024Adapting Amidst Degradation: Cross Domain Li-ion Battery Health Estimation via Physics-Guided Test-Time Trainingcited by 3
  8. 2024Health Index Estimation Through Integration of General Knowledge with Unsupervised Learningcited by 3
  9. 2023Lithium-ion Battery Online Knee Onset Detection by Matrix Profilecited by 1
  10. 2023A Self-attention Knowledge Domain Adaptation Network for Commercial Lithium-ion Batteries State-of-health Estimation under Shallow Cyclescited by 1
  11. 2023Accurate battery lifetime prediction across diverse aging conditions with deep learningcited by 1
  12. 2025Diagnostic-free onboard battery health assessmentcited by 1
16 more
  1. 2025Fast and Generalizable parameter-embedded Neural Operators for Lithium-Ion Battery Simulationcited by 1
  2. 2026From paper to benchmark: agentic, framework-based reproduction of under-specified methods in machine health intelligencecited by 1
  3. 2022Multivariate Prediction Intervals for Random Forestscited by 0
  4. 2022Random Search Hyper-Parameter Tuning: Expected Improvement Estimation and the Corresponding Lower Boundcited by 0
  5. 2023Toward High-Performance Energy and Power Battery Cells with Machine Learning-based Optimization of Electrode Manufacturingcited by 0
  6. 2023Depth analysis of battery performance based on a data-driven approachcited by 0
  7. 2023Health diagnosis and recuperation of aged Li-ion batteries with data analytics and equivalent circuit modelingcited by 0
  8. 2024Machine Learning based prediction of Vanadium Redox Flow Battery temperature rise under different charge-discharge conditionscited by 0
  9. 2024Coupling Machine Learning with Ontology for Robotics Applicationscited by 0
  10. 2025Chemistry-aware battery degradation prediction under simulated real-world cyclic protocolscited by 0
  11. 2025Learning to fuse: dynamic integration of multi-source data for accurate battery lifespan predictioncited by 0
  12. 2025Multiscale Dual-path Feature Aggregation Network for Remaining Useful Life Prediction of Lithium-Ion Batteriescited by 0
  13. 2026Excluding the Target Domain Improves Extrapolation: Deconfounded Hierarchical Physics Constraintscited by 0
  14. 2026Charging Phase Health Indicators for Battery State-of-Health Estimation: A Systematic Comparison of CC, CV, and Combined Approaches under Cross-Battery Validationcited by 0
  15. 2026PiDDM: Physics-Informed Differentiable Degradation Modeling for Lithium-Ion Battery State-of-Health Predictioncited by 0
  16. 2026Early Cycle Charge Trajectory Generative Prediction and Full Life Cycle Health Management of Iron-Chromium Flow Batteries Based on FlowBD-E1cited by 0