Machine learning based prediction method for the heat release rate of a fire source
YANG Yunhao1, ZHANG Guowei1, ZHU Guoqing1, YUAN Diping1, HE Minghuan2,3
1. Shenzhen Research Institute, China University of Mining and Technology, Shenzhen 518057, China; 2. Ruinengsaite Technology (Shenzhen) Co., Ltd., Shenzhen 518118, China; 3. Jiangsu Firemana Safety Technology Co., Ltd., Xuzhou 221100, China
Abstract：[Objective] An accurate measurement of the heat release rate (HRR) of a fire source is crucial for thoroughly understanding the fire evolution process. However, the commonly used oxygen consumption method requires expensive equipment, leading to high operational costs. According to the energy conservation principle, heat released per unit of time during material combustion is closely related to the increase of the room temperature. Machine learning methods have demonstrated considerable potential for exploring relationships between several independent and dependent variables, and attempts have been made to predict fire parameters based on temperature. Therefore, based on previous studies, this paper proposes a comprehensive machine learning framework to predict the HRR using temperature data as input. Furthermore, feature selection techniques are innovatively introduced to obtain key location temperatures to maximize HRR prediction accuracy. First, fire scenarios with different parameters are simulated in an ISO 9705 room using the fire dynamics simulator (FDS) software. Temperature data at different locations are obtained by gridding thermocouples, and a fire database is constructed. Subsequently, feature selection using recursive feature elimination (RFE) algorithms based on least absolute shrinkage and selection operator (Lasso) and random forest (RF) is performed to obtain two different low-dimensional subsets from high-dimensional simulated temperature features. Control groups with the same number of features are established. Finally, the performance of three typical models, namely, linear regression (LR), K-nearest neighbors (KNN), and light gradient boosting machine (LightGBM), for predicting the HRR are compared using different feature subsets. The results show that using the subset obtained by RFE based on RF, the LightGBM model demonstrates the lowest root mean square error (RMSE) and mean absolute error (MAE) values of 23.89 kW and 15.49 kW, respectively, indicating the least difference between its predicted and observed values. Furthermore, regarding the coefficient of determination, the LightGBM model reaches the highest value of 0.991 6, close to 1, thereby demonstrating its superior fitting capability. This is primarily due to the complex nonlinear relationship between the trained temperature dataset and HRR. Compared to the KNN and LR models, the histogram algorithm and the leaf-wise growth strategy with a depth limit enable the LightGBM model to give full play to its advantages. Tree-based models, such as LightGBM and XGBoost, can also be used for algorithm model-level improvements in the future. Additionally, deep learning models, such as multilayer fully connected and convolutional neural networks, can be utilized for fitting complex mappings. Compared with LightGBM models trained with manual feature subsets, RFE based on RF decreases the prediction errors (RMSE and MAE values decrease by 46.54 % and 50.66 %, respectively). The coefficient of determination also increases by 2.1 %, validating that this feature selection method can considerably improve prediction accuracy over manual feature selection. This paper proposes a comprehensive machine learning framework to obtain thermocouple temperature through the FDS fire simulation and then combines the feature selection and prediction models for HRR prediction, substantially improving prediction accuracy over manual feature selection. This comprehensive framework has theoretical and practical significance, thereby opening new pathes for HRR prediction. Subsequent research studies will construct additional comprehensive fire databases, increase the number of feature parameters, or explore algorithm combinations to improve HRR prediction.
杨云浩, 张国维, 朱国庆, 袁狄平, 贺名欢. 基于机器学习的火源热释放速率预测方法[J]. 清华大学学报（自然科学版）, 2024, 64(5): 922-932.
YANG Yunhao, ZHANG Guowei, ZHU Guoqing, YUAN Diping, HE Minghuan. Machine learning based prediction method for the heat release rate of a fire source. Journal of Tsinghua University(Science and Technology), 2024, 64(5): 922-932.
[1] BABRAUSKAS V, PEACOCK R D. Heat release rate:The single most important variable in fire hazard[J]. Fire Safety Journal, 1992, 18(3):255-272. [2] 程远平, 陈亮, 张孟君. 火灾过程中火源热释放速率模型及其实验测试方法[J]. 火灾科学, 2002, 11(2):70-74. CHENG Y P, CHEN L, ZHANG M J. The models and experimental testing method of heat release rate of fuel during the development of fire[J]. Fire Safety Science, 2002, 11(2):70-74. (in Chinese) [3] THORNTON W M. XV. The relation of oxygen to the heat of combustion of organic compounds[J]. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 1917, 33(194):196-203. [4] HUGGETT C. Estimation of rate of heat release by means of oxygen consumption measurements[J]. Fire and Materials, 1980, 4(2):61-65. [5] 中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. 火灾试验表面制品的实体房间火试验方法:GB/T 25207-2010[S]. 北京:中国标准出版社, 2010. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Standardization Administration of the People's Republic of China. Fire tests:Full-scale room test for surface products:GB/T 25207-2010[S]. Beijing:Standards Press of China, 2010. (in Chinese) [6] 孙利民, 尚志强, 夏烨. 大数据背景下的桥梁结构健康监测研究现状与展望[J]. 中国公路学报, 2019, 32(11):1-20. SUN L M, SHANG Z Q, XIA Y. Development and prospect of bridge structural health monitoring in the context of big data[J]. China Journal of Highway and Transport, 2019, 32(11):1-20. (in Chinese) [7] 乔莹莹. 基于数值预测的机器学习相关算法综述[J]. 安阳工学院学报, 2017, 16(4):71-74. QIAO Y Y. Summarization of machine learning based on numerical prediction[J]. Journal of Anyang Institute of Technology, 2017, 16(4):71-74. (in Chinese) [8] 林岚, 王婧璇, 付振荣, 等. 脑老化中脑年龄预测模型研究综述[J]. 生物医学工程学杂志, 2019, 36(3):493-498. LIN L, WANG J X, FU Z R, et al. A review on brain age prediction in brain ageing[J]. Journal of Biomedical Engineering, 2019, 36(3):493-498. (in Chinese) [9] 高超, 林红蕾, 胡海清, 等. 我国林火发生预测模型研究进展[J]. 应用生态学报, 2020, 31(9):3227-3240. GAO C, LIN H L, HU H Q, et al. A review of models of forest fire occurrence prediction in China[J]. Chinese Journal of Applied Ecology, 2020, 31(9):3227-3240. (in Chinese) [10] 吴浩, 牛风雷. 高温球床辐射传热中的机器学习模型[J]. 清华大学学报(自然科学版), 2023, 63(8):1213-1218. WU H, NIU F L. Machine learning model of radiation heat transfer in the high-temperature nuclear pebble bed[J]. Journal of Tsinghua University (Science and Technology), 2023, 63(8):1213-1218. (in Chinese) [11] DENG L, SHI C L, LI H R, et al. Prediction of energy mass loss rate for biodiesel fire via machine learning and its physical modeling of flame radiation evolution[J]. Energy, 2023, 275:127388. [12] SAEED F, PAUL A, HONG W H, et al. Machine learning based approach for multimedia surveillance during fire emergencies[J]. Multimedia Tools and Applications, 2020, 79(23-24):16201-16217. [13] 刘全义, 朱博, 邓力, 等. 基于机器学习的双参数火灾探测方法[J]. 中国安全科学学报, 2022, 32(5):90-96. LIU Q Y, ZHU B, DENG L, et al. Double parameters fire detection method based on machine learning[J]. China Safety Science Journal, 2022, 32(5):90-96. (in Chinese) [14] WU X Q, ZHANG X N, HUANG X Y, et al. A real-time forecast of tunnel fire based on numerical database and artificial intelligence[J]. Building Simulation, 2022, 15(4):511-524. [15] ZHANG X N, WU X Q, PARK Y, et al. Perspectives of big experimental database and artificial intelligence in tunnel fire research[J]. Tunnelling and Underground Space Technology, 2021, 108:103691. [16] WU X Q, PARK Y, LI A, et al. Smart detection of fire source in tunnel based on the numerical database and artificial intelligence[J]. Fire Technology, 2021, 57(2):657-682. [17] HODGES J L, LATTIMER B Y, LUXBACHER K D. Compartment fire predictions using transpose convolutional neural networks[J]. Fire Safety Journal, 2019, 108:102854. [18] MCGRATTAN K B, MCDERMOTT R, WEINSCHENK C. Fire dynamics simulator technical reference guide volume 1:Mathematical model. (2021-11-20). https://www.nist.gov/publications/fire-dynamics-simulator-technical-reference-guide-volume-1-mathematical-model. [19] 杨晓菡, 何学超, 邓玲. ISO 9705房间实体火灾试验及FDS模拟研究[J]. 消防科学与技术, 2016, 35(7):920-924. YANG X H, HE X C, DENG L. Research of ISO 9705 room fire test and FDS simulation[J]. Fire Science and Technology, 2016, 35(7):920-924. (in Chinese) [20] 朱五八. 不同通风状况下典型软垫家具火灾特性研究[D]. 合肥:中国科学技术大学, 2007. ZHU W B. The study on effect of ventilation on fire behavior of typical upholstered furniture[D]. Hefei:University of Science and Technology of China, 2007. (in Chinese) [21] DRYSDALE D. An introduction to fire dynamics[M]. 3rd ed. New York, USA:John Wiley & Sons, 2011. [22] 中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. 消防安全工程第4部分:设定火灾场景和设定火灾的选择:GB/T 31593.4-2015[S]. 北京:中国标准出版社, 2015. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Standardization Administration of the People's Republic of China. Fire safety engineering Part 4:Selection of design fire scenarios and design fires:GB/T 31593.4-2015[S]. Beijing:Standards Press of China, 2015. (in Chinese) [23] MCGRATTAN K, HOSTIKKA S, FLOYD J, et al. Fire dynamics simulator user's guide. (2021-11-20). https://pages.nist.gov/fds-smv/manuals.html. [24] BAUM H R, MCCAFFREY B J. Fire induced flow field-theory and experiment[J]. Fire Safety Science, 1989, 2:129-148. [25] YU L, LIU H. Efficient feature selection via analysis of relevance and redundancy[J]. The Journal of Machine Learning Research, 2004, 5(12):1205-1224. [26] KOHAVI R, JOHN G H. Wrappers for feature subset selection[J]. Artificial Intelligence, 1997, 97(1-2):273-324. [27] GUYON I, WESTON J, BARNHILL S, et al. Gene selection for cancer classification using support vector machines[J]. Machine Learning, 2002, 46(1-3):389-422. [28] TIBSHIRANI R. Regression shrinkage and selection via the Lasso:A retrospective[J]. Journal of the Royal Statistical Society Series B:Statistical Methodology, 2011, 73(3):273-282. [29] 李根, 邹国华, 张新雨. 高维模型选择方法综述[J]. 数理统计与管理, 2012, 31(4):640-658. LI G, ZOU G H, ZHANG X Y. Model selection for high-dimensional data:A review[J]. Journal of Applied Statistics and Management, 2012, 31(4):640-658. (in Chinese) [30] VERIKAS A, GELZINIS A, BACAUSKIENE M. Mining data with random forests:A survey and results of new tests[J]. Pattern Recognition, 2011, 44(2):330-349. [31] 张晓利, 陆化普. 非参数回归方法在短时交通流预测中的应用[J]. 清华大学学报(自然科学版), 2009, 49(9):1471-1475. ZHANG X L, LU H P. Non-parametric regression and application for short-term traffic flow forecasting[J]. Journal of Tsinghua University (Science and Technology), 2009, 49(9):1471-1475. (in Chinese) [32] 曹莹, 苗启广, 刘家辰, 等. AdaBoost算法研究进展与展望[J]. 自动化学报, 2013, 39(6):745-758. CAO Y, MIAO Q G, LIU J C, et al. Advance and prospects of AdaBoost algorithm[J]. Acta Automatica Sinica, 2013, 39(6):745-758. (in Chinese) [33] KE G L, MENG Q, FINLEY T, et al. LightGBM:A highly efficient gradient boosting decision tree[C]//31st Conference on Neural Information Processing Systems. Long Beach, USA:NIPS, 2017:3146-3154. [34] FAN J L, MA X, WU L F, et al. Light gradient boosting machine:An efficient soft computing model for estimating daily reference evapotranspiration with local and external meteorological data[J]. Agricultural Water Management, 2019, 225:105758.
2024, Vol. 64 
