Autors: Hristov K., Ganev, B. T., Grigorova, T. G.
Title: Industrial Fire Detection and Classification via ML-Enhanced Semiconductor Sensor Matrices
Keywords: Fire Detection, Gas sensor, Hazardous Substances, Machine learning

Abstract: Despite advancements in prevention protocols and safety instrumentation, industrial fires still claim thousands of lives each year and cost businesses worldwide billions of dollars. Ionization smoke detectors are widely adopted as a solution to this issue, due to their high sensitivity and affordability. However, the underlying technology has several notable limitations, related to false alarms and smoldering fire events. Our article explores the possibility of achieving improved results using multimodal semiconductor sensors, coupled with machine learning algorithms. The results revealed excellent detection capabilities and promising differentiation potential.

References

  1. R. Islam, M. I. Hossain, M. S. Rahman, S. Kabir, M. S. R. Sohan, and A. Shufian, “Smart IoT system for automatic detection and protection from indoor hazards: an experimental study,” in Proc. 2022 IEEE 10th Region 10 Humanitarian Technology Conf. (R10-HTC), Hyderabad, India, Sep. 2022.
  2. J. Fonollosa, A. Solórzano, and S. Marco, “Chemical sensor systems and associated algorithms for fire detection: a review,” Sensors, vol. 18, no. 2, art. 553, 2018.
  3. C. D. Litton and M. Hertzberg, Principles of Ionization Smoke Detection: Development of a New Sensor for Combustion‑Generated Submicrometer Particulates, Rep. PB271920, Bureau of Mines, U.S. Dep. of the Interior, 1977.
  4. R. W. Bukowski et al., “Performance of Home Smoke Alarms: Analysis of the Response of Several Available Technologies in Residential Fire Settings,” NIST Tech. Note 1455‑1, Gaithersburg, MD, Dec. 2007.
  5. Ministry of Housing, Communities & Local Government (UK), Economic and Social Costs of Fire: Control and Response Costs Including False Alarms, UK Gov., June 2020.
  6. R. N. Pietraru, M. Nicolae, Ș. Mocanu, and D. Merezeanu, “Easy-to-Use MOX-Based VOC Sensors for Efficient Indoor Air Quality Monitoring,” Sensors, vol. 24, no. 8, p. 2501, 2024.
  7. T. Kim et al., “Ultralow-Power Single-Sensor-Based E-Nose System Powered by Duty Cycling and Deep Learning for Real-Time Gas Identification,” ACS Sensors June, 2024, https://doi.org/10.1021/acssensors.4c00471
  8. M. Verriele et al., “The metrological qualification of commercial VOC multi-sensor systems: A mandatory step prior to indoor air quality monitoring,” IEEE Sensors J., 2024, doi:10.1109/JSEN.2024.3368285.
  9. S. Shaharuddin, K. N. A. Maulud, S. A. F. S. Abdul Rahman, A. I. Che Ani, and B. Pradhan, “The role of IoT sensors in smart building context for indoor fire hazard scenario: a systematic review,” Internet of Things, vol. 22, 2023, art. 100803.
  10. Bosch Sensortec, “BME688 Digital low power gas, pressure, temperature & humidity sensor with AI – Datasheet,” Rev. 1.3, Feb. 2024.
  11. I. H. Parlak, M. Milli, and N. S. Milli, “Machine learning–based detection of olive oil adulteration using BME688 gas sensor matrix,” Food Anal. Methods, 2025, doi:10.1007/s12161-025-02803-0.
  12. Y. Zhao et al., “Impact of a Diverse Combination of Metal Oxide Gas Sensors on Machine Learning–Based Gas Recognition in Mixed Gases,” ACS Omega, vol. 6, no. 19, pp. 12345–12356, May 2021.
  13. M. El Barkani, N. Benamar, H. Talei, and M. Bagaa, “Gas leakage detection using Tiny Machine Learning,” Electronics, vol. 13, no. 23, p. 4768, 2024.
  14. Hensel S., Marinov M.B., Koch M., Arnaudov D., Evaluation of deep learning-based neural network methods for cloud detection and segmentation, (2021) Energies, 14 (19), art. no. 6156, DOI: 10.3390/en14196156

Issue

2025 34th International Scientific Conference Electronics, ET 2025 - Proceedings, pp. 1-5, 2025, Bulgaria, https://doi.org/10.1109/ET66806.2025.11204167

Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus