Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{206581,
        author = {Abhinand Haridas E and Hiran P P and Muhammad Nasim C K and Pranav Joy and Prof. Abhijnha B N},
        title = {Fire Detection System in EV Vehicles Using LSTM},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {no},
        pages = {9-14},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=206581},
        abstract = {Electric vehicles (EVs) depend on lithium-ion batteries. While these batteries are efficient, they can create safety problems like thermal runaway, which can cause fire or explosion. Traditional Battery Management Systems (BMS) use threshold-based methods that only detect faults after serious issues occur. This paper suggests an AI-based early fire detection system that uses Long Short-Term Memory (LSTM) networks to improve battery safety. The system analyzes time-series battery data, including temperature, voltage, current, and state of charge (SOC). It also uses derived features like temperature rate of change (dT/dt), voltage variation (dV/dt), internal resistance, and power to improve prediction accuracy. The collected data is preprocessed and normalized with MinMaxScaler and converted into sequences that fit LSTM modeling. The LSTM model learns patterns of normal and abnormal battery behavior over time and predicts the likelihood of fire occurrence. Using predefined thresholds, the system classifies battery conditions as normal, warning, or high-risk. When a high fire risk is detected, the system sends alerts and simulates battery isolation to prevent further damage. Experimental results show that the proposed system can effectively spot early signs of thermal instability and provide timely warnings. This approach boosts the reliability and safety of EV battery systems by allowing predictive monitoring instead of reacting after problems arise. The proposed model can be expanded for real-time applications and integrated with advanced BMS for improved vehicle safety.},
        keywords = {.},
        month = {July},
        }