What question did this study set out to answer?

The research aims to develop and evaluate a data-driven framework for predicting the State of Charge (SOC) of electric vehicle batteries.

April 18, 2026

Health monitoring systems for EV battery

View Full Paper

Ask AI

Bookmark

Key result

Random Forest Regressor provides more consistent EV battery SOC predictions than RC-VAE across three chemistries.

Why the study?

Electric vehicle battery health monitoring is crucial to guaranteeing dependability, performance, and safety.

Population

Publicly accessible datasets of Lithium-ion, Lithium Polymer, and Lead-acid battery chemistries

Comparison

Random Forest Regressor vs Recurrent Conditional Variational Autoencoder

Design

Machine learning model evaluation study

Authors

SSSasmita C SUniversity of Arts AMAnbuselvi MathivananSri Sri University SPSaravanan PalaniswamyASA College

Discussion

Member takes

Implication

Validates ML-driven SOC prediction across battery chemistries in EVs; extends single-chemistry models to multi-chemistry real-world use.

Key Points

The research aims to develop and evaluate a data-driven framework for predicting the State of Charge (SOC) of electric vehicle batteries.
Utilized publicly available datasets across three battery chemistries: Lithium-ion, Lithium Polymer, and Lead-acid.
Implemented a Random Forest Regressor for SOC estimation and compared its performance with a Recurrent Conditional Variational Autoencoder.
Evaluated using metrics like Mean Absolute Error, Root Mean Squared Error, and coefficient of determination (R²).
Integrated a voice-activated chatbot interface for user interaction regarding SOC information and advice.
The Random Forest model provided more consistent and reliable SOC predictions than the RC-VAE.
The performance of the RC-VAE was notably poorer under certain datasets and scaling conditions.
The chatbot interface enhanced user accessibility to SOC information in electric vehicle battery monitoring.

Structured PICO

Population

Datasets of three distinct battery chemistries (Lithium-ion, Lithium Polymer, and Lead-acid) from Mendeley Data, CALCE Battery Research Group, and open-source GitHub repositories

Intervention

Random Forest Regressor (RFR) for State of Charge (SOC) estimation, combined with a voice-activated chatbot interface

Comparator

Recurrent Conditional Variational Autoencoder (RC-VAE)

Outcome

State of Charge (SOC) estimation performance evaluated using Mean Absolute Error, Root Mean Squared Error, and coefficient of determination (R²)

A Random Forest Regressor provides robust and consistent State of Charge predictions across different EV battery chemistries compared to RC-VAE.

Cite This Study

S et al. (2026) studied this question. The Random Forest Regressor provides more consistent and dependable State of Charge predictions across three EV battery chemistries compared to the RC-VAE model.

synapsesocial.com/papers/69e31fcb40886becb653ef6a https://doi.org/10.1051/e3sconf/202670503001/pdf

View Full Paper

Ask AI

Bookmark

Health monitoring systems for EV battery

Discussion

Member takes

Implication

Key Points

Structured PICO

Cite This Study

Health monitoring systems for EV battery

Discussion

Member takes

Implication

Key Points

Structured PICO

Cite This Study

Also Consider

Also Consider