A novel battery SOC estimation method based on random search optimized LSTM neural network

Xuqing Chai, Shihao Li and Fengwei Liang

Energy, 2024, vol. 306, issue C

Abstract: Battery state of charge (SOC) estimation is crucial for assessing electric vehicle safety and evaluating the remaining driving range. Owing to the complexity, variability of operating conditions, and the highly nonlinear internal mechanisms of batteries, accurate SOC estimation remains a focal point of current research. Therefore, this paper proposes a random search optimization-based Long Short-Term Memory (RS-LSTM) neural network for precise SOC estimation. The paper firstly uses the CALCE dataset, extracting discharge capacity and discharge energy as critical features from six battery parameters by employing the random forest algorithm. The Look-back, Epoch, Batch size, and Learning rate parameters in the LSTM neural network optimized by random search algorithm. The study result reveals optimal settings (Look back: 45, Epoch: 177, Batch size: 64, Learning rate: 0.0026) achieving superior estimation accuracy, evidenced by mean average error（MAE） and root mean square error（RMSE）of 0.221 % and 0.262 %, respectively. Furthermore, the method's superiority, effectiveness, robustness, and applicability were verified by conducting tests across various estimation methods, various SOC estimation intervals, various temperature conditions, the addition of Gaussian noise, and tests on experimental and real-world vehicle data. The research process demonstrates that the proposed method has superior precision and indicates promising potential for future applications.

Keywords: Lithium-ion batteries; SOC estimation; Random search; Random forest dimensionality reduction; LSTM neural network (search for similar items in EconPapers)
Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544224023570
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:306:y:2024:i:c:s0360544224023570

DOI: 10.1016/j.energy.2024.132583

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().