Interpreting CNN-RNN Hybrid Model-Based Ensemble Learning with Explainable Artificial Intelligence to Predict the Performance of Li-Ion Batteries in Drone Flights
Source abstract
Abstract
Li-ion batteries are important in modern technology, especially for drones, due to their high energy density, long cycle life, and lightweight properties. Predicting their performance is crucial for enhancing drone flight safety, optimizing operations, and reducing costs. This involves using advanced techniques like machine learning (e.g., Convolutional Neural Network-CNNs, Recurrent Neural Network-RNNs), statistical modeling (e.g., Kalman Filtering), and explainable AI (e.g., SHAP, LIME, PDP) to forecast battery behavior, extend battery life, and improve drone efficiency. The study aims to develop a CNN-RNN-based ensemble model, enhanced with explainable AI, to predict key battery metrics during drone flights. The model’s predictions will aid in enhancing battery performance via continuous, data-driven monitoring, improve drone safety, optimize operations, and reduce greenhouse gas emissions through advanced recycling methods. In the present study, comparisons are made for the behaviors of two different drone Li-ion batteries, numbered 92 and 129. The ensemble model in Drone 92 showed the best performance with MAE (0.00032), RMSE (0.00067), and R2 (0.98665) scores. Similarly, the ensemble model in Drone 129 showed the best performance with MAE (0.00030), RMSE (0.00044), and R2 (0.98094) performance metrics. Similar performance results are obtained in the two predictions. However, drone 129 has a minimally lower error rate. When the Partial Dependence Plots results, which are one of the explainable AI (XAI) techniques, are interpreted with the decision tree algorithm, the effect of the Current (A) value on the model estimations in both drone flights is quite evident. When the current value is around −4, the model is more sensitive and shows more changes. This study will establish benchmarks for future research and foster advancements in drone and battery technologies through extensive testing.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Battery Technologies Research · Electric Vehicles and Infrastructure · Advancements in Battery Materials
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer and Instructional Technologies Education, Gazi University, Ankara 06560, Türkiye
Faculty of Engineering, Department of Computer Engineering, Adıyaman University, Adıyaman 02040, Türkiye
Mobilers Team, Sivas Cumhuriyet University, Sivas 58050, Türkiye
Faculty of Engineering, Department of Computer Engineering, Gazi University, Ankara 06570, Türkiye
Battery Research Laboratory, Faculty of Engineering and Natural Sciences, Sivas University of Science and Technology, Sivas 58010, Türkiye
References (64)
Show source-provided references
1. Afroze, S., Reza, M.S., Kuterbekov, K., Kabyshev, A., Kubenova, M.M., Bekmyrza, K.Z., and Azad, A.K. (2023). Emerging and recycling of Li-ion batteries to aid in energy storage, a review. Recycling, 8.10.3390/recycling8030048 ↗
2. The challenges and opportunities of battery-powered flight10.1038/s41586-021-04139-1 ↗
3. Oyucu, S., Dümen, S., Duru, İ., Aksöz, A., and Biçer, E. (2024). Discharge Capacity Estimation for Li-Ion Batteries: A Comparative Study. Symmetry, 16.10.3390/sym16040436 ↗
4. Insights and reviews on battery lifetime prediction from research to practice10.1016/j.jechem.2024.03.013 ↗
5. Fahmy, H., Swief, R., Hasanien, H., Alharbi, M., Maldonado, J., and Jurado, F. (2023). Hybrid State of Charge Estimation of Lithium-Ion Battery Using the Coulomb Counting Method and an Adaptive Unscented Kalman Filter. Energies, 16.10.3390/en16145558 ↗
6. Shoriat Ullah, M., and Seo, K. (2022). Prediction of lithium-ion battery capacity by functional principal component analysis of monitoring data. Appl. Sci., 12.10.3390/app12094296 ↗
7. Lithium-ion batteries remaining useful life prediction based on a mixture of empirical mode decomposition and ARIMA model10.1016/j.microrel.2016.07.151 ↗
8. Oyucu, S., Ersöz, B., Sağıroğlu, Ş., Aksöz, A., and Biçer, E. (2024). Optimizing Lithium-Ion Battery Performance: Integrating Machine Learning and Explainable AI for Enhanced Energy Management. Sustainability, 16.10.3390/su16114755 ↗
9. Explainability-driven model improvement for SOH estimation of lithium-ion battery10.1016/j.ress.2022.109046 ↗
10. Oyucu, S., Doğan, F., Aksöz, A., and Biçer, E. (2024). Comparative Analysis of Commonly Used Machine Learning Approaches for Li-Ion Battery Performance Prediction and Management in Electric Vehicles. Appl. Sci., 14.10.3390/app14062306 ↗
11. Marri, I., Petkovski, E., Cristaldi, L., and Faifer, M. (2023). Comparing Machine Learning Strategies for SoH Estimation of Lithium-Ion Batteries Using a Feature-Based Approach. Energies, 16.10.3390/en16114423 ↗
12. Ersöz, B., Sağıroğlu, Ş., and Bülbül, H.İ. (2022, January 18–21). A short review on explainable artificial intelligence in renewable energy and resources. Proceedings of the 2022 11th International Conference on Renewable Energy Research and Application (ICRERA), Istanbul, Turkey.10.1109/ICRERA55966.2022.9922870 ↗
13. Molnar, C., Freiesleben, T., König, G., Herbinger, J., Reisinger, T., Casalicchio, G., Wright, M.N., and Bischl, B. (2023). Relating the partial dependence plot and permutation feature importance to the data generating process. World Conference on Explainable Artificial Intelligence, Springer.10.1007/978-3-031-44064-9_24 ↗
14. Au–Ag alloy nanoparticles with tunable cavity for plasmon-enhanced photocatalytic H2 evolution10.1016/j.jechem.2020.01.005 ↗
15. Issues and challenges facing rechargeable lithium batteries10.1038/35104644 ↗
16. Electrochemical model based observer design for a lithium-ion battery10.1109/TCST.2011.2178604 ↗
17. Model-based electrochemical estimation and constraint management for pulse operation of lithium ion batteries10.1109/TCST.2009.2027023 ↗
18. Design of primary and secondary cells: II. An equation describing battery discharge10.1149/1.2423659 ↗
19. Dynamic lithium-ion battery model for system simulation10.1109/TCAPT.2002.803653 ↗
20. Accurate electrical battery model capable of predicting runtime and IV performance10.1109/TEC.2006.874229 ↗
21. Selection and performance-degradation modeling of LiMO2/Li4Ti5O12 and LiFePO4/C battery cells as suitable energy storage systems for grid integration with wind power plants: An example for the primary frequency regulation service
22. Characterization of high-power lithium-ion batteries by electrochemical impedance spectroscopy. II: Modelling10.1016/j.jpowsour.2010.07.071 ↗
23. An analytical model for predicting the remaining battery capacity of lithium-ion batteries10.1109/TVLSI.2006.876094 ↗
24. A hybrid battery model capable of capturing dynamic circuit characteristics and nonlinear capacity effects10.1109/TEC.2011.2167014 ↗
25. Stroe, D.-I. (2014). Lifetime Models for Lithium Ion Batteries Used in Virtual Power Plants. [Ph.D. Thesis, Department of Energy Technology, Aalborg University].
26. A review of the state of health for lithium-ion batteries: Research status and suggestions10.1016/j.jclepro.2020.120813 ↗
27. A review on the state of health estimation methods of lead-acid batteries10.1016/j.jpowsour.2021.230710 ↗
28. A review of non-probabilistic machine learning-based state of health estimation techniques for Lithium-ion battery10.1016/j.apenergy.2021.117346 ↗
29. A critical review of improved deep learning methods for the remaining useful life prediction of lithium-ion batteries10.1016/j.egyr.2021.08.182 ↗
30. Recent advances on cooperative control of heterogeneous multi-agent systems subject to constraints: A survey10.1080/21642583.2022.2074169 ↗
31. H∞ state estimation for memristive neural networks with randomly occurring DoS attacks10.1080/21642583.2022.2048322 ↗
32. Hassoun, M. (1995). Fundamentals of Artificial Neural Networks, The MIT Press.10.1109/JPROC.1996.503146 ↗
33. Cross trajectory gaussian process regression model for battery health prediction10.35833/MPCE.2019.000142 ↗
34. Adaptive sliding window LSTM NN based RUL prediction for lithium-ion batteries integrating LTSA feature reconstruction10.1016/j.neucom.2021.09.025 ↗
35. Data-driven nonparametric Li-ion battery ageing model aiming at learning from real operation data–Part A: Storage operation10.1016/j.est.2020.101409 ↗
36. Modified Gaussian process regression models for cyclic capacity prediction of lithium-ion batteries10.1109/TTE.2019.2944802 ↗
37. Machine learning pipeline for battery state-of-health estimation10.1038/s42256-021-00312-3 ↗
38. Offline and online blended machine learning for lithium-ion battery health state estimation10.1109/TTE.2021.3129479 ↗
39. An overview of data-driven battery health estimation technology for battery management system10.1016/j.neucom.2023.02.031 ↗
40. An end-to-end neural network framework for state-of-health estimation and remaining useful life prediction of electric vehicle lithium batteries10.1016/j.rser.2021.111843 ↗
41. A Data-driven Auto-CNN-LSTM Prediction Model for Lithium-ion Battery Remaining Useful Life10.1109/TII.2020.3008223 ↗
42. Predicting battery end of life from solar off-grid system field data using machine learning10.1016/j.joule.2021.11.006 ↗
43. State-of-health estimators coupled to a random forest approach for lithium-ion battery aging factor ranking10.1016/j.jpowsour.2020.229154 ↗
44. Online diagnosis of state of health for lithium-ion batteries based on short-term charging profiles10.1016/j.jpowsour.2020.228478 ↗
45. Predicting future capacity of lithium-ion batteries using transfer learning method10.1016/j.est.2023.108120 ↗
46. Tian, Y., Wen, J., Yang, Y., Shi, Y., and Zeng, J. (2022). State-of-health prediction of lithium-ion batteries based on cnn-bilstm-am. Batteries, 8.10.3390/batteries8100155 ↗
47. Wang, S., Ren, P., Takyi-Aninakwa, P., Jin, S., and Fernandez, C. (2022). A critical review of improved deep convolutional neural network for multi-timescale state prediction of lithium-ion batteries. Energies, 15.10.3390/en15145053 ↗
48. A data-driven learning method for online prediction of drone battery discharge10.1016/j.ast.2022.107921 ↗
49. Data-efficient modeling for power consumption estimation of quadrotor operations using ensemble learning10.1016/j.ast.2023.108791 ↗
50. Safavi, V., Mohammadi Vaniar, A., Bazmohammadi, N., Vasquez, J.C., and Guerrero, J.M. (2024). Battery Remaining Useful Life Prediction Using Machine Learning Models: A Comparative Study. Information, 15.10.3390/info15030124 ↗
51. Chen, Y., Baek, D., Bocca, A., Macii, A., Macii, E., and Poncino, M. (2018, January 7–11). A case for a battery-aware model of drone energy consumption. Proceedings of the 2018 IEEE International Telecommunications Energy Conference (INTELEC), Turino, Italy.10.1109/INTLEC.2018.8612333 ↗
52. Thermal design and optimization of lithium ion batteries for unmanned aerial vehicles10.1002/est2.48 ↗
53. Rufino Júnior, C.A., Sanseverino, E.R., Gallo, P., Amaral, M.M., Koch, D., Kotak, Y., Diel, S., Walter, G., Schweiger, H.-G., and Zanin, H. (2024). Unraveling the Degradation Mechanisms of Lithium-Ion Batteries. Energies, 17.10.3390/en17143372 ↗
54. Battery Management System for Drones
55. Korthauer, R. (2018). Lithium-Ion Batteries: Basics and Applications, Springer.10.1007/978-3-662-53071-9 ↗
56. Health prognostics for lithium-ion batteries: Mechanisms, methods, and prospects10.1039/D2EE03019E ↗
57. CNN and RNN mixed model for image classification10.1051/matecconf/201927702001 ↗
58. Fake news detection: A hybrid CNN-RNN based deep learning approach
59. Hu, Y., Wong, Y., Wei, W., Du, Y., Kankanhalli, M., and Geng, W. (2018). A novel attention-based hybrid CNN-RNN architecture for sEMG-based gesture recognition. PLoS ONE, 13.10.1371/journal.pone.0206049 ↗
60. Ensemble of averages: Improving model selection and boosting performance in domain generalization
61. Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions10.1186/s40537-021-00444-8 ↗
62. Ensemble deep learning: A review10.1016/j.engappai.2022.105151 ↗
63. Duval, A. (2019). Explainable Artificial Intelligence (XAI), Mathematics Institute, The University of Warwick. MA4K9 Scholarly Report.
64. An explainable artificial intelligence framework for the deterioration risk prediction of hepatitis patients10.1007/s10916-021-01736-5 ↗