International Journal of Engineering
Trends and Technology

Research Article | Open Access | Download PDF
Volume 74 | Issue 6 | Year 2026 | Article Id. IJETT-V74I6P104 | DOI : https://doi.org/10.14445/22315381/IJETT-V74I6P104

Exploring Second Life Potential of Retired EV Batteries: Data-Driven Criteria Development using Beta Variational Autoencoders and GMM Clustering

Louie Jhyms Acidillo, Ralph Harold C. Cabanes, Joel Geralla, Kimjay B. Ihada, JV Mark T. Pasilaban, Gian Carlo Villahermosa6, Jestoni P. Tan, Emerita M. Tan, Chona R. Dagatan9, Donald R. Lalican

Received Revised Accepted Published
21 Nov 2025 05 Mar 2026 08 May 2026 27 Jun 2026

Citation :

Louie Jhyms Acidillo, Ralph Harold C. Cabanes, Joel Geralla, Kimjay B. Ihada, JV Mark T. Pasilaban, Gian Carlo Villahermosa6, Jestoni P. Tan, Emerita M. Tan, Chona R. Dagatan9, Donald R. Lalican, "Exploring Second Life Potential of Retired EV Batteries: Data-Driven Criteria Development using Beta Variational Autoencoders and GMM Clustering," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 6, pp. 47-65, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I6P104

Abstract

The increasing need for environmentally friendly ways to dispose of and reuse Electric Vehicle (EV) batteries shows how important it is to do a good end-of-life assessment. Second-life applications offer economic and environmental advantages; however, the classification of decommissioned batteries continues to pose difficulties. This research employs a data-driven methodology utilizing a β-Variational Autoencoder (β-VAE) in conjunction with Gaussian Mixture Model (GMM) clustering to assess battery health classification. From the UNIBO Powertools Dataset, six engineered features were extracted: normalized Capacity, delta capacity (ΔSoH), capacity ratio, average voltage, temperature, and cycle count. We used β-VAE to cluster the data by standardizing it and compressing it into a latent space. The silhouette score, Davies–Bouldin index, Calinski–Harabasz index, Bayesian Information Criterion (BIC), and Akaike Information Criterion (AIC) were all used to measure how well the clustering worked. The β-VAE has an MAE of 0.3104 and an R² of 0.3384. GMM clustering found two good clusters (k = 2) with a silhouette score of 0.826 and a Davies-Bouldin index of 0.263. We turned the cluster-wise means into percentage thresholds to sort batteries into three groups: Healthy, Second-Life, and Disposal. For instance, Second-Life batteries had a voltage of at least 71.08 percent and a temperature of no more than 10.99 percent. The suggested β-VAE-GMM framework is a strong and easy-to-understand way to figure out when a battery has reached the end of its life. Setting clear, data-driven criteria for second-life eligibility helps circular economy projects.

Keywords

Battery health assessment, Second-life batteries, Β-Vae, Gaussian mixture model, Circular economy.

References

[1] Jeremy Neubauer, and Ahmad Pesaran, “The Ability of Battery Second-Use Strategies to Impact Plug-in Electric Vehicle Prices and Serve Utility Energy Storage Applications,” Journal of Power Sources, vol. 196, no. 23, pp. 10351-10358, 2011.
[CrossRef] [Google Scholar] [Publisher Link]

[2] E. Martinez-Laserna et al., “Evaluation of Lithium-Ion Battery Second Life Performance and Degradation,” 2016 IEEE Energy Conversion Congress and Exposition (ECCE), Milwaukee, WI, USA, pp. 1-7, 2016.
[CrossRef] [Google Scholar] [Publisher Link]

[3] Dipti Kamath et al., “Economic and Environmental Feasibility of Second-Life Lithium-Ion Batteries as Fast-Charging Energy Storage,” Environmental Science and Technology, vol. 54, no. 11, pp. 6878-6887, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Diederik P. Kingma, and Max Welling, “Auto-Encoding Variational Bayes,” arXiv preprint, pp. 1-14, 2013.
[Google Scholar]

[5] Hauke Engel, Patrick Hertzke, and Giulia Siccardo, “Second-Life EV Batteries: The Newest Value Pool in Energy Storage,” McKinsey and Company, vol. 30, 2019.
[Google Scholar]

[6] Aki Takahashi, Anirudh Allam, and Simona Onori, “Evaluating the Feasibility of Batteries for Second-Life Applications using Machine Learning,” iScience, vol. 26, no. 4, pp. 1-15, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[7] Mohammed Hussein Saleh Mohammed Haram et al., “Feasibility of Utilising Second Life EV Batteries: Applications, Lifespan, Economics, Environmental Impact, Assessment, and Challenges,” Alexandria Engineering Journal, vol. 60, no. 5, pp. 4517-4536, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[8] S.M. Rezaul Karim, Debasish Sarker, and Md. Monirul Kabir, “Analyzing the Impact of Temperature on the PV Module Surface During Electricity Generation using Machine Learning Models,” Cleaner Energy Systems, vol. 9, pp. 1-7, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[9] Pushpak B. Patel, and Sanjay R. Vyas, “Evaluating Lithium-Ion Battery Performance through Mathematical Modeling and Simulation: Charging, Discharging, and Performance Parameter,” 2024 4th International Conference on Emerging Frontiers in Electrical and Electronic Technologies (ICEFEET), Patna, India, pp. 1-6, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[10] Arvind R. Singh et al., “Machine Learning-based Energy Management and Power Forecasting in Grid-Connected Microgrids with Multiple Distributed Energy Sources,” Scientific Reports, vol. 14, no. 1, pp. 1-23, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Alexander Wallis et al., “A Framework for Strategy Selection of Atomic Entities in the Holonic Smart Grid,” ENERGY 2020, The Tenth International Conference on Smart Grids, Green Communications and IT Energy-aware Technologies, Lisbon, Portugal, pp. 11-16, 2020.
[Google Scholar] [Publisher Link]

[12] Yixuan Wang et al., “Environmental Impact Assessment of Second Life and Recycling for LiFePO4 Power Batteries in China,” Journal of Environmental Management, vol. 314, pp. 1-9, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Zhijing Deng et al., “State of Health Estimation for Lithium-Ion Batteries based on Composite Multi-Scale Tsallis Entropy Algorithm and Hybrid Physics-Informed Neural Network,” Journal of Energy Storage, vol. 159, 2026.
[CrossRef] [Google Scholar] [Publisher Link]

[14] Guangzhao Zhang et al., “A Monofluoride Ether-based Electrolyte Solution for Fast-Charging and Low-Temperature Non-Aqueous Lithium Metal Batteries,” Nature Communications, vol. 14, no. 1, pp. 1-13, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[15] João Pedro Caldas Ferreira et al., “Discharge Behavior of Lithium Batteries,” CONTROLO 2024: Proceedings of the 16th APCA International Conference on Automatic Control and Soft Computing, Porto, Portugal, vol. 1325, pp. 431-443, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Kateřina Nováková et al., “Second-Life of Lithium-Ion Batteries from Electric Vehicles: Concept, Aging, Testing, and Applications,” Energies, vol. 16, no. 5, pp. 1-19, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Jinsong Yu et al., “Online State-of-Health Prediction of Lithium-Ion Batteries with Limited Labeled Data,” International Journal of Energy Research, vol. 44, no. 14, pp. 11345-11360, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[18] Mano Schmitz, and Julia Kowal, “A Deep Learning Approach for Online State of Health Estimation of Lithium-Ion Batteries using Partial Constant Current Charging Curves,” Batteries, vol. 10, no. 6, pp. 1-16, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[19] T.G.T.A. Bandara, “State of Health Estimation using Machine Learning for Li-Ion Batteries on Electric Vehicles,” 2021 IEEE Vehicle Power and Propulsion Conference (VPPC), Gijon, Spain, pp. 1-4, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[20] Alireza Valizadeh, and Mohammad Hossein Amirhosseini, “Machine Learning in Lithium-Ion Battery: Applications, Challenges, and Future Trends,” SN Computer Science, vol. 5, no. 6, pp. 1-17, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[21] Mohamed H. Al-Meer, “A Deep Learning Method for the Health State Prediction of Lithium-Ion Batteries based on LUT-Memory and Quantization,” World Electric Vehicle Journal, vol. 15, no. 2, pp. 1-20, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[22] Salil Bharany et al., “Wildfire Monitoring based on Energy Efficient Clustering Approach for FANETS,” Drones, vol. 6, no. 8, pp. 1-19, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[23] Irina Higgins et al., Beta-VAE: Learning basic Visual Concepts with a Constrained Variational Framework, ICLR, pp. 1-22, 2017. [Online]. Available: https://openreview.net/forum?id=Sy2fzU9gl

[24] Dongdong Cheng et al., “K-Means Clustering with Natural Density Peaks for Discovering Arbitrary-Shaped Clusters,” IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 8, pp. 11077-11090, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[25] Juner Zhu et al., “End-of-Life or Second Life Options for Retired Electric Vehicle Batteries,” Cell Reports Physical Science, vol. 2, no. 8, pp. 1-26, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[26] Yudong Wang et al., “Lithium-Ion Battery Screening by K-means with DBSCAN for Denoising,” Computers, Materials, and Continua, vol. 65, no. 3, pp. 2111-2122, 2020.
[CrossRef] [Google Scholar] [Publisher Link]

[27] Shirui Feng et al., “Health State Estimation of On-Board Lithium-Ion Batteries based on GMM-BID Model,” Sensors, vol. 22, no. 24, pp. 1-17, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[28] Christopher P. Burgess et al., “Understanding Disentangling in β-VAE,” arXiv preprint, pp. 1-11, 2018.
[CrossRef] [Google Scholar] [Publisher Link]

[29] Ricky T.Q. Chen et al., “Isolating Sources of Disentanglement in Variational Autoencoders,” Advances in Neural Information Processing Systems, vol. 31, pp. 1-11, 2018.
[Google Scholar] [Publisher Link]