Optimizing Lithium-Ion Battery Performance: Integrating Machine Learning and Explainable AI for Enhanced Energy Management

Saadin Oyucu,Betül Ersöz,Şeref Sağıroğlu,Ahmet Aksöz,Emre Biçer

DOI: https://doi.org/10.3390/su16114755

IF: 3.9

2024-06-03

Sustainability

Abstract:Managing the capacity of lithium-ion batteries (LiBs) accurately, particularly in large-scale applications, enhances the cost-effectiveness of energy storage systems. Less frequent replacement or maintenance of LiBs results in cost savings in the long term. Therefore, in this study, AdaBoost, gradient boosting, XGBoost, LightGBM, CatBoost, and ensemble learning models were employed to predict the discharge capacity of LiBs. The prediction performances of each model were compared based on mean absolute error (MAE), mean squared error (MSE), and R-squared values. The research findings reveal that the LightGBM model exhibited the lowest MAE (0.103) and MSE (0.019) values and the highest R-squared (0.887) value, thus demonstrating the strongest correlation in predictions. Gradient boosting and XGBoost models showed similar performance levels but ranked just below LightGBM. The competitive performance of the ensemble model indicates that combining multiple models could lead to an overall performance improvement. Furthermore, the study incorporates an analysis of key features affecting model predictions using SHAP (Shapley additive explanations) values within the framework of explainable artificial intelligence (XAI). This analysis evaluates the impact of features such as temperature, cycle index, voltage, and current on predictions, revealing a significant effect of temperature on discharge capacity. The results of this study emphasize the potential of machine learning models in LiB management within the XAI framework and demonstrate how these technologies could play a strategic role in optimizing energy storage systems.

environmental sciences,environmental studies,green & sustainable science & technology

What problem does this paper attempt to address?

### What problem does this paper attempt to solve? This paper aims to optimize the performance of lithium-ion batteries (LiBs), especially in large-scale applications, by enhancing energy management systems through the combination of machine learning and explainable artificial intelligence (XAI) technologies. Specifically, the main objectives of the study include: 1. **Predicting the discharge capacity of lithium-ion batteries**: Using various machine learning algorithms (such as AdaBoost, Gradient Boosting, XGBoost, LightGBM, and CatBoost) to predict the discharge capacity of lithium-ion batteries and comparing the performance of different models. 2. **Improving model interpretability**: Utilizing SHAP values (Shapley Additive Explanations) within the explainable artificial intelligence framework to analyze key features affecting model predictions, such as temperature, cycle number, voltage, and current. 3. **Optimizing energy storage systems**: Demonstrating how these technologies can play a strategic role in lithium-ion battery management to optimize the performance and lifespan of energy storage systems. The study found that the LightGBM model had the lowest Mean Absolute Error (MAE) and Mean Squared Error (MSE), and the highest R-squared value, indicating its superior predictive performance. Additionally, ensemble models also showed good performance, suggesting that combining multiple models can further enhance overall performance. Through these methods, the researchers hope to improve the efficiency of lithium-ion battery management, reduce costs, and promote sustainable development.

Optimizing Lithium-Ion Battery Performance: Integrating Machine Learning and Explainable AI for Enhanced Energy Management

Discharge Capacity Estimation for Li-Ion Batteries: A Comparative Study

AI-Driven Digital Twin Model for Reliable Lithium-Ion Battery Discharge Capacity Predictions

Estimation of Lithium-ion Battery Discharge Capacity by Integrating Optimized Explainable-AI and Stacked LSTM Model

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

Machine Learning Strategies for Forecasting Discharge Capacity in Lithium-Ion Batteries with NCM Layered Cathode Material

Machine learning Approach to Battery Management and Balancing Techniques for Enhancing Electric Vehicle Battery Performance

Machine Learning Aided Predictions for Capacity Fade of Li-Ion Batteries

Life Prediction of Lithium Battery Based on the Grouping of Charge and Discharge Conditions in Logistics Scenario

Lithium-Ion Battery Health Prediction on Hybrid Vehicles Using Machine Learning Approach

Machine learning for predicting battery capacity for electric vehicles

Early prediction of battery degradation in grid-scale battery energy storage system using extreme gradient boosting algorithm

Calendar ageing modelling using machine learning: an experimental investigation on lithium ion battery chemistries

Python‐based scikit‐learn machine learning models for thermal and electrical performance prediction of high‐capacity lithium‐ion battery

Predictive machine learning in optimizing the performance of electric vehicle batteries: Techniques, challenges, and solutions

Comparison of prediction performance of lithium titanate oxide battery discharge capacity with machine learning methods

Optimizing electric bike battery management: Machine learning predictions of LiFePO4 temperature under varied conditions

Machine learning-based prediction of lithium-ion battery life cycle for capacity degradation modelling

A cloud-edge collaborative strategy for capacity prognostic of lithium-ion batteries based on dynamic weight allocation and machine learning

Enhancing Battery Management System Optimization: A Secure AI Modeling Approach

Prediction of remaining useful life and recycling node of lithium-ion batteries based on a hybrid method of LSTM and LightGBM