The Precision SOC Estimation for Fire Prevention of the EES Using ANN

• Published in: 2021 10th International Conference on Renewable Energy Research and Application (ICRERA) pp. 231 - 234
• Main Authors: Jung, Ga-Eun, Baek, JiKook, Liu, Jianyong, Dao, Van Quan, Dinh, Minh-Chau, Kim, Chang Soon, Lee, Myung-Kwan, Bae, JungHyo
• Format: Conference Proceeding
• Language: English
• Published: IEEE 26-09-2021
• Subjects: Artificial neural network; Artificial neural networks; Batteries; Circuit breakers; Electrical energy storage; Estimation; Fire prevention; Predictive models; State of charge; Surge protection; Voltage measurement
• ISSN: 2572-6013
• DOI: 10.1109/ICRERA52334.2021.9598750

Recently, the spread of Electrical energy storage(EES) system in Korea has increased and a fire accident has also occurred. To apply reliable EES to the grid, it is important to estimate the precision State of Charge (SOC) and apply an additional battery protection system. This paper proposed a model that can evaluate precise SOC through Artificial Neural Network (ANN) and a battery protection system against external surge for fire prevention. The proposed fire prevention method includes an ANN model for precise SOC prediction and a circuit breaker system in case of external surge intrusion, which is applied to the Battery management system(BMS) design. In case of the ANN model, the actual SOC and the predicted SOC of the LiB model were compared to prove the validity. As a result, the ANN showed the maximum and average errors of 2.1% and 0.5%, respectively, and the accuracy was 99.5%. In the battery protection system test for fire prevention, it was confirmed that the circuit breaker was successfully performed when an external surge intruded. These results can be effective to design a BMS system for EES fire protection.