Journal of Tsinghua University(Science and Technology) >
Research on a multiparameter fire detection method for aircraft cargo compartment based on an improved self-attention mechanism
Received date: 2024-07-03
Online published: 2025-03-27
Copyright
Objective: With the rapid advancement of the aviation industry, ensuring aircraft safety, particularly in sensitive areas like cargo holds, is of paramount importance. Fires in aircraft cargo can be triggered by various factors, such as electrical malfunctions, hazardous materials, or environmental conditions, and pose significant threats to passengers and crew. Given the growing complexity of fire detection in these confined spaces, more reliable and accurate fire detection methods are urgently needed. Traditional fire detection systems, which primarily depend on single-sensor technologies, like smoke or heat detectors, have long been criticized for their high false alarm rates and limited accuracy. These deficiencies often result in delayed responses or unnecessary interventions, which ultimately compromise operational safety and efficiency. Therefore, this study aims to develop an innovative fire detection system that can overcome the limitations of conventional methods while meeting the advanced safety standards of modern aviation. Methods: To tackle these challenges, this research introduces an improved multiparameter fire detection method leveraging an advanced self-attention mechanism within the Transformer model architecture. The approach integrates data from multiple sensors, including carbon monoxide, smoke, humidity, and temperature sensors, to capture a wide range of environmental parameters in aircraft cargo holds. Data are gathered by simulating realistic fire scenarios within a laboratory setting, ensuring that the system is trained on diverse datasets that reflect the unpredictable nature of fire development in cargo spaces. The core of the proposed method is a Transformer-based model that incorporates two key innovations: local attention mechanism and multiscale feature extraction. The local attention mechanism addresses the computational complexity of processing long sequences of input data by dividing the data into smaller, manageable windows. This allows the model to focus on localized features without the burden of analyzing the entire sequence at once, making it more efficient and suitable for real-time applications. Furthermore, the multiscale feature extraction module processes data in parallel across different time windows, capturing short-term fluctuations and long-term trends, which is crucial for detecting gradual fires, such as slow-burning or smoldering fires, that traditional systems may miss. Results: The proposed method was rigorously evaluated through a series of experiments on a fire detection dataset designed to mimic real-world conditions in aircraft cargo holds. A range of hyperparameters, including sequence lengths, activation functions, dropout rates, and optimizers, was tested to fine-tune model classification performance. Results revealed that the optimized model significantly outperformed traditional approaches, such as convolutional neural networks, recurrent neural networks, and long short-term memory networks, in terms of classification accuracy, particularly under challenging conditions involving noisy or incomplete sensor data. The model excelled at distinguishing between fire and non-fire events, showcasing its superior ability to handle real-world fire scenarios. Moreover, the Transformer's intrinsic parallel computing capability reduced training times, making it a practical solution for time-sensitive fire detection applications in aviation. Conclusions: This study presents a novel multiparameter fire detection system that integrates an improved self-attention mechanism with local attention and multiscale feature extraction, offering several advantages over traditional models. The proposed method achieves higher accuracy, lower computational complexity, and faster training times, making it highly suitable for deployment in aircraft cargo hold fire detection systems. The promising results from the laboratory-based experiments suggest that this method can be readily adapted to real-world operational settings. Future research will focus on further validating the model's performance in live environments, aiming to extend its applicability to other safety-critical domains beyond aviation, such as industrial safety and transportation systems.
Key words: fire detection; machine learning; self-attention; Transformer model
Haibin WANG , Zhihui ZHANG , Zonghao BU , Zishan GAO , Quanyi LIU . Research on a multiparameter fire detection method for aircraft cargo compartment based on an improved self-attention mechanism[J]. Journal of Tsinghua University(Science and Technology), 2025 , 65(4) : 777 -785 . DOI: 10.16511/j.cnki.qhdxxb.2025.27.014
| 1 |
|
| 2 |
何永勃, 董玉珊, 雷建, 等. 飞机货舱火灾多传感器探测方法研究[J]. 中国安全科学学报, 2018, 28 (5): 74- 79.
|
| 3 |
ADITYANAND G. Aircraft cargo compartment multi-sensor smoke detection algorithm development: DOT/FAA/AR-07/58 [R]. Virginia: U.S. Department of Transportation Federal Aviation Administration, 2008.
|
| 4 |
周志刚. 智能家居防火系统的算法研究[D]. 北京: 中国地质大学(北京), 2014.
ZHOU Z G. The algorithm research on fire protection system of smart home [D]. Beijing: China University of Geosciences (Beijing), 2014. (in Chinese)
|
| 5 |
LIU M H, ZENG A L, LAI Q X, et al. T-WaveNet: Tree-structured wavelet neural network for sensor-based time series analysis [EB/OL]. (2020-12-10)[2024-06-10]. https://doi.org/10.48550/arXiv.2012.05456.
|
| 6 |
XIAO Z W, XU X, XING H L, et al. RNTS: Robust neural temporal search for time series classification [C]//2021 International Joint Conference on Neural Networks (IJCNN). Shenzhen, China: IEEE, 2021: 1-8.
|
| 7 |
YANG J B, NGUYEN M N, SAN P P, et al. Deep convolutional neural networks on multichannel time series for human activity recognition [C]//Proceedings of the 24th International Conference on Artificial Intelligence. Buenos Aires, Argentina: AAAI Press, 2015: 3995-4001.
|
| 8 |
|
| 9 |
ZHANG X C, GAO Y F, LIN J, et al. TapNet: Multivariate time series classification with attentional prototypical network [C]//Proceedings of the 34th AAAI Conference on Artificial Intelligence. New York, USA: AAAI Press, 2020: 6845-6852.
|
| 10 |
VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook, USA: Curran Associates Inc., 2017: 6000-6010.
|
| 11 |
DEVLIN J, CHANG M W, LEE K, et al. BERT: Pre-training of deep bidirectional transformers for language understanding [C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Minneapolis, USA: Association for Computational Linguistics, 2019: 4171-4186.
|
| 12 |
|
| 13 |
YUAN Y, XUN G X, MA F L, et al. A novel channel-aware attention framework for multi-channel EEG seizure detection via multi-view deep learning [C]//2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI). Las Vegas, USA: IEEE, 2018: 206-209.
|
| 14 |
LIANG Y X, KE S Y, ZHANG J B, et al. GeoMAN: Multi-level attention networks for geo-sensory time series prediction [C]//Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. Stockholm, Sweden: AAAI Press, 2018: 3428-3434.
|
| 15 |
ZHOU H Y, ZHANG S H, PENG J Q, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting [J/OL]. (2021-03-28)[2024-06-10]. https://doi.org/10.48550/arXiv.2012.07436.
|
| 16 |
WEN Q S, ZHOU T, ZHANG C L, et al. Transformers in time series: A survey [C]//Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence. Macao, China: Curran Associates, Inc., 2022: 6778-6786.
|
| 17 |
WANG Z G, YAN W Z, OATES T. Time series classification from scratch with deep neural networks: A strong baseline [C]//2017 International Joint Conference on Neural Networks (IJCNN). Anchorage, USA: IEEE, 2017: 1578-1585.
|
| 18 |
HAO Y F, CAO H P. A new attention mechanism to classify multivariate time series [C]//Proceedings of the 29th International Joint Conferences on Artificial Intelligence. Yokohama, Japan: International Joint Conferences on Artificial Intelligence, 2020: 1999-2005.
|
/
| 〈 |
|
〉 |