October 5, 2025Journal of Energy Storage6 citationsOpen Access

Fast estimation of lithium-ion battery state of health using time series classification

Q: What is the clinical evidence from this study?

Study design: Other. Population: Lithium-ion battery State of Health. Intervention: Multivariate Time Series Classification (TSC) based on Random Interval Classifier (RIC) vs. Traditional machine learning approaches. Primary outcome: State of Health (SoH) prediction error.

View Full Paper

Ask AI

Bookmark

Population

Lithium-ion batteries (Oxford and NASA benchmark datasets)

Comparison

Multivariate Time Series Classification approach… vs Traditional machine learning approaches relying…

Design

Other

Key result

A multivariate Time Series Classification approach using 2%-3% of partial charging cycle data achieved battery State of Health prediction errors of about 1% and reduced diagnostic time 30-fold.

Authors

MMMoreno Marzolla FMFrancesco Mercuri

Discussion

Member takes

Overview

A novel multivariate time series classification approach accurately estimates lithium-ion battery state of health using only 2-3% of charge cycle data, reducing diagnostic time 30-fold.

Structured PICO

Population

Lithium-ion batteries (Oxford and NASA benchmark datasets)

Intervention

Multivariate Time Series Classification (TSC) approach based on the Random Interval Classifier (RIC) using partial charging cycle data (temperature and voltage)

Comparator

Traditional machine learning approaches relying on full charge-discharge cycle data

Outcome

State of Health (SoH) estimation accuracy and diagnostic time

A novel multivariate time series classification approach accurately estimates lithium-ion battery state of health using only 2-3% of charge cycle data, reducing diagnostic time 30-fold.

Cite This Study

Marzolla et al. (2025) studied Lithium-ion battery State of Health. Multivariate Time Series Classification (TSC) based on Random Interval Classifier (RIC) vs. Traditional machine learning approaches was evaluated on State of Health (SoH) prediction error. A multivariate Time Series Classification approach using 2%-3% of partial charging cycle data achieved battery State of Health prediction errors of about 1% and reduced diagnostic time 30-fold.

synapsesocial.com/papers/6aa2a492a841a277b27f6a50 https://doi.org/10.1016/j.est.2025.118747

View Full Paper

Ask AI

Bookmark

October 5, 2025Journal of Energy Storage6 citationsOpen Access

Fast estimation of lithium-ion battery state of health using time series classification

View Full Paper

Ask AI

Bookmark

Population

Lithium-ion batteries (Oxford and NASA benchmark datasets)

Comparison

Multivariate Time Series Classification approach… vs Traditional machine learning approaches relying…

Design

Other

Key result

A multivariate Time Series Classification approach using 2%-3% of partial charging cycle data achieved battery State of Health prediction errors of about 1% and reduced diagnostic time 30-fold.

Authors

MMMoreno Marzolla FMFrancesco Mercuri

Discussion

Member takes

Overview

A novel multivariate time series classification approach accurately estimates lithium-ion battery state of health using only 2-3% of charge cycle data, reducing diagnostic time 30-fold.

Structured PICO

Population

Lithium-ion batteries (Oxford and NASA benchmark datasets)

Intervention

Multivariate Time Series Classification (TSC) approach based on the Random Interval Classifier (RIC) using partial charging cycle data (temperature and voltage)

Comparator

Traditional machine learning approaches relying on full charge-discharge cycle data

Outcome

State of Health (SoH) estimation accuracy and diagnostic time

A novel multivariate time series classification approach accurately estimates lithium-ion battery state of health using only 2-3% of charge cycle data, reducing diagnostic time 30-fold.

Cite This Study

synapsesocial.com/papers/6aa2a492a841a277b27f6a50 https://doi.org/10.1016/j.est.2025.118747

View Full Paper

Ask AI

Bookmark