Machine learning pipeline for battery state-of-health estimation (2021-02-01T00:00:00.000000Z)

TL;DR

This work designs and evaluates a machine learning pipeline for estimation of battery capacity fade—a metric of battery health—on 179 cells cycled under various conditions, and provides insights into the design of scalable data-driven models for battery SOH estimation, emphasizing the value of confidence bounds around the prediction.

Abstract

Lithium-ion batteries are ubiquitous in 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 Article, 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 error of 0.45%. This work provides insights into the design of scalable data-driven models for battery SOH estimation, emphasizing the value of confidence bounds around the prediction. The pipeline methodology combines experimental data with machine learning modelling and could be applied to other critical components that require real-time estimation of SOH. Rechargeable lithium-ion batteries play a crucial role in many modern-day applications, including portable electronics and electric vehicles, but they degrade over time. To ensure safe operation, a battery’s ‘state of health’ should be monitored in real time, and this machine learning pipeline, tested on a variety of charging conditions, can provide such an online estimation of battery state of health.

Authors

D. Roman

1 papers

Saurabh Saxena

1 papers

V. Robu

1 papers

TL;DR

Abstract

Authors

References90 items

Comparative Analysis of Features for Determining State of Health in Lithium-Ion Batteries

Closed-loop optimization of fast-charging protocols for batteries with machine learning

Energy storage: The future enabled by nanomaterials

Online State-of-Health Estimation for Li-Ion Battery Using Partial Charging Segment Based on Support Vector Machine

Machine Learning With Python

Data-driven prediction of battery cycle life before capacity degradation

CHAPTER 13. Challenges and Opportunities in Lithium-ion Battery Supply

SOH Estimation for Li-ion Batteries Based on Features of IC Curves and Multi-output Gaussian Process Regression Method

Lithium-Ion Battery State of Health Monitoring Based on Ensemble Learning

State of health estimation of lithium-ion batteries based on the constant voltage charging curve

Random forest regression for online capacity estimation of lithium-ion batteries

Understanding heterogeneous electrocatalytic carbon dioxide reduction through operando techniques

Modelling

Using recursive feature elimination in random forest to account for correlated variables in high dimensional data

A Stacked Autoencoder-Based Deep Neural Network for Achieving Gearbox Fault Diagnosis

Remaining capacity estimation of lithium-ion batteries based on the constant voltage charging profile

Understanding ageing in Li-ion batteries: a chemical issue.

Accurate Uncertainties for Deep Learning Using Calibrated Regression

A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve

Manipulating Machine Learning: Poisoning Attacks and Countermeasures for Regression Learning

Gaussian Process Regression for In Situ Capacity Estimation of Lithium-Ion Batteries

Attacking the Madry Defense Model with L1-based Adversarial Examples

A review of lithium-ion battery state of charge estimation and management system in electric vehicle applications: Challenges and recommendations

EAD: Elastic-Net Attacks to Deep Neural Networks via Adversarial Examples

Diagnosis of Electric Vehicle Batteries Using Recurrent Neural Networks

Degradation diagnostics for lithium ion cells

Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles

Designing high-energy lithium-sulfur batteries.

Understanding undesirable anode lithium plating issues in lithium-ion batteries

Charging protocols for lithium-ion batteries and their impact on cycle life—An experimental study with different 18650 high-power cells

Battery Health Prognosis for Electric Vehicles Using Sample Entropy and Sparse Bayesian Predictive Modeling

Understanding electrochemical potentials of cathode materials in rechargeable batteries

Deep feature synthesis: Towards automating data science endeavors

Challenges and recommendations

Online Capacity Estimation of Lithium-Ion Batteries Based on Novel Feature Extraction and Adaptive Multi-Kernel Relevance Vector Machine

An Efficient Algorithm for Calculating the Exact Hausdorff Distance

Online identification of lithium-ion battery parameters based on an improved equivalent-circuit model and its implementation on battery state-of-power prediction

Critical review of on-board capacity estimation techniques for lithium-ion batteries in electric and hybrid electric vehicles

Explaining and Harnessing Adversarial Examples

A support vector machine-based state-of-health estimation method for lithium-ion batteries under electric vehicle operation

Adaptation of an Electrochemistry-based Li-Ion Battery Model to Account for Deterioration Observed Under Randomized Use

Determination of lithium-ion battery state-of-health based on constant-voltage charge phase

Why Walking the Dog Takes Time: Frechet Distance Has No Strongly Subquadratic Algorithms Unless SETH Fails

Confidence intervals for random forests: the jackknife and the infinitesimal jackknife

A review on lithium-ion battery ageing mechanisms and estimations for automotive applications

Correlation and variable importance in random forests

Electrochemistry-based Battery Modeling for Prognostics

On-board state of health monitoring of lithium-ion batteries using incremental capacity analysis with support vector regression

Model based identification of aging parameters in lithium ion batteries

Experimental investigation of the lithium-ion battery impedance characteristic at various conditions and aging states and its influence on the application

Random Search for Hyper-Parameter Optimization

A comparative study of equivalent circuit models for Li-ion batteries

A review on prognostics and health monitoring of Li-ion battery

Characterization of high-power lithium-ion batteries by electrochemical impedance spectroscopy. II: Modelling

Scikit-learn: Machine Learning in Python

Generalised Entropy of Curves for the Analysis and Classification of Dynamical Systems

Prognostics Methods for Battery Health Monitoring Using a Bayesian Framework

Metrics for evaluating performance of prognostic techniques

Prognostics in Battery Health Management

An Improved Battery Characterization Method Using a Two-Pulse Load Test

Studies on the degradation of Li-ion batteries by the use of microreference electrodes

An integrated approach to battery health monitoring using bayesian regression and state estimation

Curve Matching, Time Warping, and Light Fields: New Algorithms for Computing Similarity between Curves

Study of the charging process of a LiCoO2-based Li-ion battery

Incremental Capacity Analysis and Close-to-Equilibrium OCV Measurements to Quantify Capacity Fade in Commercial Rechargeable Lithium Batteries

Pattern Recognition and Machine Learning

Characterizing aging effects of lithium ion batteries by impedance spectroscopy

Fast Gaussian Process Regression using KD-Trees

The ARTEMIS European driving cycles for measuring car pollutant emissions.

Gaussian Processes in Machine Learning

Gene Selection for Cancer Classification using Support Vector Machines

Factors Which Limit the Cycle Life of Rechargeable Lithium (Metal) Batteries

Prognostics, the real issues involved with predicting life remaining

The effect of sample size and variability of data on the comparative performance of artificial neural networks and regression

Modeling of Galvanostatic Charge and Discharge of the Lithium/Polymer/Insertion Cell

Isotonic Median Regression: A Linear Programming Approach