Modeling Battery Energy Storage Systems Based on Remaining Useful Lifetime through Regression Algorithms and Binary Classifiers
Source abstract
Abstract
This research work implements an initial methodology for the assessment of Battery Energy Storage Systems (BESSs) based on Remaining Useful Lifetime (RUL), and its main contribution is the modeling and estimation of Health and Charge indicators through regression algorithms and binary classifiers during the battery’s operation. Linear Regression, Ridge Regression, and Lasso Regression are the main algorithms for modeling the State of Health (SOH), while Decision Tree, Naïve Bayes, and Logistic Regression are implemented as binary classifiers to estimate the charge and discharge during battery operation. Additional data science techniques are executed to provide feature selection, validation, and metrics of performance. The results show that binary classifiers achieve a remarkable accuracy, around 95% for charge and discharge predictions, which is supported by experimental battery measurements. Similarly, regression algorithms achieve accuracy results around 97% and provide a basis for determining the Remaining Useful Lifetime (RUL) according to the End-of-Life (EOL) criteria of a BESS.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Battery Technologies Research · IoT-based Smart Home Systems · Fault Detection and Control Systems
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electrical Power Engineering and Mechatronics, Tallinn University of Technology, 19086 Tallinn, Estonia
Funding
Estonian Research Council
References (14)
Show source-provided references
1. Liao, C., Li, H., and Wang, L. (2009, January 7–10). A dynamic equivalent circuit model of LiFePO4 cathode material for lithium ion batteries on hybrid electric vehicles. Proceedings of the 2009 IEEE Vehicle Power and Propulsion Conference, Dearborn, MI, USA.10.1109/VPPC.2009.5289681 ↗
2. Rahn, C.D., and Wang, C.-Y. (2013). Battery System Engineering, John Wiley & Sons, Ltd.10.1002/9781118517048 ↗
3. Pop, V., Bergveld, H.J., Notten PH, L., and Regtien, P.P. (2005). State-of-the-Art of Battery State-of-Charge Determination, Springer. Measurement Science and Technology.10.1088/0957-0233/16/12/R01 ↗
4. Critical review of state of health estimation methods of Li-ion batteries for real applications10.1016/j.rser.2015.11.042 ↗
5. Early-Stage end-of-Life prediction of lithium-Ion battery using empirical mode decomposition and particle filter
6. Macintosh, A. (2023, April 24). Li-Ion Battery Aging Datasets. Ames Research Center, Available online: https://c3.ndc.nasa.gov/dashlink/members/38/.
7. Decision tree methods: Applications for classification and prediction
8. An Introduction to Logistic Regression Analysis and Reporting10.1080/00220670209598786 ↗
9. Belyadi, H., and Haghighat, A. (2021). Machine Learning Guide for Oil and Gas Using Python, Gulf Professional Publishing.
10. Rish, I. (2001). En Empirical Study of the Naïve Bayes Classifier, T.J. Watson Research Center.
11. Cross-Validation
12. Status of land cover classification accuracy assessment10.1016/S0034-4257(01)00295-4 ↗
13. Murphy, K.P. (2012). Machine Learning: A Probabilistic Perspective, The MIT Press.
14. Hastie, T., Tibshirani, R., and Friedman, J. (2017). The Elements of Statistical Learning: Data Mining, Inference, and Prediction, Springer. [2nd ed.].