Abstract:The global shift towards electric vehicles (EVs) underscores the critical need for reliable battery performance and safety. Lithium-ion batteries, particularly Li-NMC (lithium nickel manganese cobalt oxide), are widely adopted for their balanced functional and performance characteristics. However, the advancement of batteries with higher nickel content and reduced manganese and cobalt introduces challenges, including increased susceptibility to thermal runaway and degradation, especially under abusive conditions like over-discharge. This study addresses significant research gaps by developing a machine learning (ML) algorithm for the early detection and predictive maintenance of over-discharged Li-NMC batteries. Current methods often fail to identify and mitigate the effects of continuous cycling, which can release harmful free radicals such as singlet oxygen ( 1 ) and superoxide ( ) that accelerate degradation. Our ML approach utilizes supervised learning, feature engineering, and model optimization, leveraging key input features such as voltage, time, and cycle count which are derived from extensive battery life testing. To validate our model, we conducted scanning electron microscopy energy-dispersive spectroscopy (SEM–EDS), galvanostatic charge–discharge (GCD) tests, and rate capability tests. The proposed ridge regression model achieved a mean absolute error (MAE) of 0.11422%, a mean squared error (MSE) of 0.02313%, and an R -squared ( R 2 ) value of 0.99, outperforming other models such as Decision Trees (DT), Recurrent Neural Networks (RNNs), Support Vector Machines (SVMs), Gradient Boosting (GB), and Lasso Regression. Our model addresses key shortcomings of existing methods, particularly in predicting degradation in precycled batteries subjected to fault induction. The insights gained contribute to a robust control strategy for EV battery management, enabling proactive maintenance, timely battery replacement, and enhanced system reliability and safety, effectively addressing the long-term challenges in battery health management.

Data-Driven Fault Diagnosis of Lithium-Ion Battery Overdischarge in Electric Vehicles

Data-Driven Fault Diagnosis of Lithium-Ion Battery Overdischarge in Electric Vehicles

Fault Mechanism Study on Li-Ion Battery at Over-Discharge and Its Diagnosis Approach

A Novel Method for Lithium-Ion Battery Fault Diagnosis of Electric Vehicle Based on Real-Time Voltage

Data-Driven Fault Diagnosis and Early Safety Warning for Lithium-Ion Batteries

Fault diagnosis of lithium-ion batteries based on voltage dip behavior

Detection of voltage fault in the battery system of electric vehicles using statistical analysis

A Comprehensive Signal-Based Fault Diagnosis Method for Lithium-Ion Batteries in Electric Vehicles

An Online Data-Driven Fault Diagnosis and Thermal Runaway Early Warning for Electric Vehicle Batteries

Review of Abnormality Detection and Fault Diagnosis Methods for Lithium-Ion Batteries

Fault diagnosis method for lithium-ion batteries in electric vehicles based on isolated forest algorithm

Integrating machine learning for health prediction and control in over-discharged Li-NMC battery systems

Sensor fault diagnosis modeling of lithium-ion batteries for electric vehicles

Fault Diagnosis and Detection for Battery System in Real-World Electric Vehicles Based on Long-term Feature Outlier Analysis

An Overdischarge Fault Detection Method for Lithium-ion Battery Module Based on Vibration Characteristics

An Incipient Multi-Fault Diagnosis Method for Lithium-Ion Battery Pack Based on Data-Driven with Incremental-Scale

Health Diagnosis for Lithium-Ion Battery by Combining Partial Incremental Capacity and Deep Belief Network During Insufficient Discharge Profile

A fast diagnosis method for accelerated degradation fault induced by overcharging of LiFePO4 batteries

An online fault diagnosis method for lithium-ion batteries based on signal decomposition and dimensionless indicators selection

Fault diagnosis method for lithium-ion batteries based on the combination of voltage prediction and Z-score

Fault Diagnosis Method for Lithium-Ion Battery Packs in Real-World Electric Vehicles Based on K-Means and the Fréchet Algorithm