Open Access

ARTICLE

A Joint Estimation Method of SOC and SOH for Lithium-ion Battery Considering Cyber-Attacks Based on GA-BP

Tianqing Yuan^1,2, Na Li^1,2, Hao Sun³, Sen Tan^4,*

1 Key Laboratory of Modern Power System Simulation and Control & Renewable Energy Technology, Ministry of Education, Northeast Electric Power University, Jilin, 132012, China
2 Department of Electrical Engineering, Northeast Electric Power University, Jilin, 132012, China
3 College of Information and Control Engineering, Jilin Institute of Chemical Technology, Jilin, 132022, China
4 Center for Research on Microgrids, Department of Energy, Aalborg University, Aalborg, 9220, Denmark

* Corresponding Author: Sen Tan. Email:

Computers, Materials & Continua 2024, 80(3), 4497-4512. https://doi.org/10.32604/cmc.2024.056061

Received 13 July 2024; Accepted 14 August 2024; Issue published 12 September 2024

Abstract

To improve the estimation accuracy of state of charge (SOC) and state of health (SOH) for lithium-ion batteries, in this paper, a joint estimation method of SOC and SOH at charging cut-off voltage based on genetic algorithm (GA) combined with back propagation (BP) neural network is proposed, the research addresses the issue of data manipulation resulting from cyber-attacks. Firstly, anomalous data stemming from cyber-attacks are identified and eliminated using the isolated forest algorithm, followed by data restoration. Secondly, the incremental capacity (IC) curve is derived from the restored data using the Kalman filtering algorithm, with the peak of the IC curve (ICP) and its corresponding voltage serving as the health factor (HF). Thirdly, the GA-BP neural network is applied to map the relationship between HF, constant current charging time, and SOH, facilitating the estimation of SOH based on HF. Finally, SOC estimation at the charging cut-off voltage is calculated by inputting the SOH estimation value into the trained model to determine the constant current charging time, and by updating the maximum available capacity. Experiments show that the root mean squared error of the joint estimation results does not exceed 1%, which proves that the proposed method can estimate the SOC and SOH accurately and stably even in the presence of false data injection attacks.

---

Keywords

---