With the rapid expansion of electric vehicles (EVs) and energy storage systems (ESS), ensuring the operational safety of lithium-ion batteries has become a critical technical challenge. Conventional battery management systems (BMS) primarily rely on threshold-based rule logic, which is limited in detecting coupled anomalies and early-stage degradation patterns. In this study, a deep learning-based framework for multivariate anomaly detection is proposed using BMS sensor data, including voltage, current, temperature, state of charge (SOC), and state of health (SOH). Five representative fault scenarios were defined, including thermal runaway precursors, cell voltage imbalance, SOC inconsistency, internal resistance increase, and communication delay. The proposed CNN-LSTM model was compared with conventional Rule-based methods and machine learning models, including Isolation Forest, Autoencoder, and LSTM. Experimental results show that the proposed model achieved the highest performance, with an F1-score of 0.885, an AUC of 0.94, and a detection delay of 8.1 s. In contrast, the Rule-based method exhibited a significantly higher false negative rate of 42.0%, indicating limitations in detecting complex anomaly patterns. These results demonstrate that the proposed spatiotemporal deep learning approach can significantly improve the accuracy and responsiveness of battery anomaly detection. Furthermore, the proposed method is expected to contribute to enhancing safety, reliability, and predictive diagnostics in next-generation intelligent BMS platforms.

This study proposes an optimization strategy for the over-current protection (OCP) parameters of a lithium iron phosphate (LiFePO₄, LFP) battery system used in electric golf carts operating under high motor-load conditions. Real-world hillclimbing tests were conducted under four clearly defined payload/passenger conditions to analyze the transient discharge-current pro-file, voltage sag, and cell-temperature response. The maximum discharge current reached -238.2 A under the 200 kg cargopayload and one-passenger condition, and the current interval exceeding 150 A lasted up to 27 s. The maximum instantaneous power was 11.05 kW. Thermal analysis showed that the cell-temperature rise was within 2°C and the maximum measured cell temperature was 22.3°C. Linear regression of voltage and current yielded R² = 0.9368 and dV/dI = 0.0126 Ω, which was used as the DC internalresistance estimate. Based on these quantitative results and the cell specification limit of 300 A continuous discharge, the OCP threshold was reviewed from 250 A to 280 A to improve driving continuity while remaining below the allowable continuous-discharge current. EIS-based SOH estimation and the AI-BMS variable protection logic are presented as an extension framework for reflecting temperature and aging effects in future OCP-setting decisions.

Lithium-ion batteries used for IT, automobiles, and industrial energy-storage devices have battery management systems (BMS) to protect the battery from abnormal voltage, current, and temperature environments, as well as safety devices like, current interruption device (CID), fuse, and vent to obtain positive temperature coefficient (PTC). Nonetheless, there are harmful to human health and property and damage the brand image of the manufacturer because of smoke, fire, and explosion of lithium battery packs. In this paper, we propose a systematic protection algorithm combining battery temperature, over-current, and interconnection between protection elements to prevent copper deposition, internal short circuit, and separator shrinkage due to frequent and instantaneous over-current discharges. The parameters of the proposed algorithm are suggested to utilize the experimental data in consideration of battery pack operating conditions and malicious conditions.

A large number of lithium-ion batteries are arranged in series and parallel in battery packs, such as those in electric vehicles or energy storage systems. As battery packs age, their output power and energy density drop because of voltage deviation, constant and non-uniform exposure to abnormal environments, and increased contact resistance between batteries; this reduces application system efficiency. Despite the balancing circuit and logic of the battery management system, the output of the battery pack is concentrated in the most severely aged unit cell and the output is frequently limited by power derating. In this study, we implemented a cell imbalance detection algorithm and selected parameters to detect a sudden decrease in battery pack output. In addition, we propose a method to increase efficiency by applying the measured testing values considering the operating conditions and abnormal conditions of the battery pack.

The lithium-ion battery pack of an electric vehicle (EV) deserves to be considered for an alternative use within smart-grid infrastructure. Despite the long automotive service life, EV batteries retain over 70~80% of their initial capacity. These battery packs must be managed for their reliability and safety. Therefore, a battery management system (BMS) should use specific algorithms to measure and estimate the status of the battery. Most importantly, the BMS of a grid-connected energy storage system (ESS) must ensure that the lithium-ion battery does not catch fire or explode due to an internal short from uncontrolled dendrite growth. In other words, the BMS of a lithium-ion battery pack should be capable of detecting the battery’s status based on the electrochemical reaction continuously until the end of the battery’s lifespan. In this paper, we propose a new protection algorithm for a dendritic lithium battery. The proposed algorithm has applied a parameter from battery pack aging results and has control power managing.

High-power lithium batteries are suitable for equipment with high power output needs, such as for ESS’s initial start-up. However, their management cost is increased by the installation of air-conditioning to minimize the risk of explosion due to internal temperature rise and also by a restriction on the number of charge/discharge cycles. High-capacity flow batteries, on the other hand, have many advantages. They can be used for over 20 years due to their low management costs, resulting from no risk of explosion and a high number of charge/discharge cycles. In this paper, we propose an ESS based on hybrid batteries that uses a lithium iron phosphate battery (LiFePO) at the initial startup and a vanadium redox flow battery (VRFB) from the end of the transient period, with a bi-directional PCS to operate two batteries with different DC voltage levels and using an efficient energy management control algorithm.