Heat stress prediction model for outdoor policeman based on machine learning

doi:10.16265/j.cnki.issn1003-3033.2024.11.0171

Abstract

Abstract:

To address the issue of predicting heat stress risks for police officers engaged in outdoor operations under high-temperature conditions, a test dataset for monitoring core temperature of police officers under different environmental working conditions, levels of physical exertion and clothing scenarios was constructed. First, features such as height, weight, age, gender, body fat percentage, physical activity ratio (PAR), clothing insulation (CI), environmental temperature and relative humidity were extracted. Then, machine learning methods, including K-nearest neighbors (KNN), random forest (RF) and gradient boosting decision trees (GBDT), were used to establish predictive models of core temperature and heat stress risk for outdoor police officers. These models were subsequently validated. The results indicate that for the predictive model of core temperature for outdoor police officers working in high-temperature environments, the goodness-of-fit measure R² exceeds 0.9 for KNN, RF and GBDT. In terms of error, the KNN model has the smallest prediction error, with a root mean square error (RMSE) of 0.053 ℃. For the heat stress prediction model for police officers engaged in outdoor operations under high-temperature conditions, the predictive performance of RF, GBDT and KNN models is significantly better than that of other models.

Key words: machine learning, outdoor operations, police officers, heat stress, core temperature, high temperature environment

CLC Number:

X968高低温防护

HU Xiaofeng, HUANG Ling. Heat stress prediction model for outdoor policeman based on machine learning[J]. China Safety Science Journal, 2024, 34(11): 220-228.

Figures/Tables 10

Fig.1

Table 1

Table 2

Table 3

Table 4

Fig.2

Table 5

Table 6

Table 7

Table 8

References 33

[1]	POTTER A W, GONZALEZ J A, KARIS A J, et al. Biophysical assessment and predicted thermophysiologic effects of body armor[J]. Plos One, 2015, 10(7): DOI: 10.1371/journal.pone.0132698.
[2]	DU Chenqiu, LI Baizhan, LI Yongqiang, et al. Modification of the predicted heat strain (PHS) model in predicting human thermal responses for Chinese workers in hot environments[J]. Building and Environment, 2019, 165: DOI:10.1016/j.buildenv.2019.106349.
[3]	MALCHAIRE J, PIETTE A, KAMPMANN B, et al. Development and validation of the predicted heat strain model[J]. The Annals of Occupational Hygiene, 2001, 45(2):123-135.
[4]	安启启, 徐刚, 杨杰, 等. 高温高湿环境下人体热生理模型的检验及应用[J]. 西安科技大学学报, 2021, 41(2):253-261.
	AN Qiqi, XU Gang, YANG Jie, et al. Examination and application of human thermo physiological model in high temperature and humidity environment[J]. Journal of Xi'an University of Science and Technology, 2021, 41(2):253-261.
[5]	李春艺, 李川, 闫肖岳, 等. 户外高温环境建筑工人热应激预测分析评价[J]. 劳动保护, 2022(6):78-82.
[6]	吴建松, 李乐天, 韩新阳, 等. 高温户外环境电网作业人员热应激预测评价[J]. 中国安全生产科学技术, 2021, 17(12):169-175.
	WU Jiansong, LI Letian, HAN Xinyang, et al. Prediction and evaluation on heat strain of power grid workers in high-temperature outdoor environment[J]. Journal of Safety Science and Technology, 2021, 17(12):169-175.
[7]	聂兴信, 王廷宇, 孙锋刚, 等. 高温矿井热湿环境对人体机能的影响[J]. 金属矿山, 2020(4):186-193.
	NIE Xingxin, WANG Tingyu, SUN Fenggang, et al. Influence of heat and humidity environment on function of human body in high temperature mine[J]. Metal Mine, 2020(4):186-193.
[8]	柯莹, 周文. 个体降温服降温效应评价指标及方法[J]. 服装学报, 2021, 6(1):1-7.
	KE Ying, ZHOU Wen. Evaluation indicators and methods of cooling effects for personal cooling clothing[J]. Journal of Clothing Research, 2021, 6(1):1-7.
[9]	HAN Xinge, HU Zhuqiang, LI Chuan, et al. Prediction of human thermal comfort preference based on supervised learning[J]. Journal of Thermal Biology, 2023,112: DOI:10.1016/j.jtherbio.2023.103484.
[10]	CHAUDHURI T, SOH Y C, LI H, et al. Machine learning based prediction of thermal comfort in buildings of equatorial Singapore[C]. 2017 IEEE International Conference on Smart Grid and Smart Cities (ICSGSC). IEEE, 2017:72-77.
[11]	MORRESI N, CASACCIA S, SORCINELLI M, et al. Sensing physiological and environmental quantities to measure human thermal comfort through machine learning techniques[J]. IEEE Sensors Journal, 2021, 21(10):12322-12 337.
[12]	LIU Kuixing, NIE Ting, LIU Wei, et al. A machine learning approach to predict outdoor thermal comfort using local skin temperatures[J]. Sustainable Cities and Society, 2020,59: DOI:10.1016/j.scs.2020.102216.
[13]	MORISHIMA S, XU Yingjie, URASHIMA A, et al. Human body skin temperature prediction based on machine learning[J]. Artificial Life and Robotics, 2021, 26:103-108.
[14]	YANG Jie, WANG Lijuan, YIN Haiguo, et al. Thermal responses of young males of three thermal preference groups in summer indoor environments[J]. Building and Environment, 2021, 194(11): DOI:10.1016/j.buildenv.2021.107705.
[15]	KARJALAINEN S. Gender differences in thermal comfort and use of thermostats in everyday thermal environments[J]. Building and Environment, 2007, 42(4):1 594-1 603.
[16]	KARJALAINEN S. Thermal comfort and gender: a literature review[J]. Indoor Air, 2012, 22(2):96-109. doi: 10.1111/j.1600-0668.2011.00747.x pmid: 21955322
[17]	FERRARO S, IAVICOLI S, RUSSO S, et al. A field study on thermal comfort in an Italian hospital considering differences in gender and age[J]. Applied Ergonomics, 2015, 50:177-184. doi: 10.1016/j.apergo.2015.03.014 pmid: 25959333
[18]	HAVENITH G. Individualized model of human thermoregulation for the simulation of heat stress response[J]. Journal of Applied Physiology, 2001, 90(5):1 943-1 954.
[19]	HUIZENGA C, ZHANG H, THOMAS D. An improved multimode model of human physiology and thermal comfort[J]. Proceedings of Building Simulation, 1999, 35(1):353-359.
[20]	TAKAHASHI Y, NOMOTO A, YODA S, et al. Thermoregulation model JOS-3 with new open source code[J]. Energy and Buildings, 2021,231: DOI:10.1016/j.enbuild.2020.110575.
[21]	ISO 8996-2021,Ergonomics of the thermal environment:determination of metabolic rate[S].
[22]	范淼. 机器学习及实践:从零开始通往Kaggle竞赛之路[M]. 北京: 清华大学出版社, 2016:183.
[23]	HEARST M A, DUMAIS S T, OSUNA E, et al. Support vector machines[J]. IEEE Intelligent Systems and Their Applications, 1998, 13(4): 18-28.
[24]	FRIEDL M A, BRODLEY C E. Decision tree classification of land cover from remotely sensed data[J]. Remote Sensing of Environment, 1997, 61(3):399-409.
[25]	KE Guolin, MENG Qi, FINLEY T, et al. LightGBM: a highly efficient gradient boosting decision tree[C]. 31^st International Conference on Neural Information Processing Systems, 2017:3 149-3 157.
[26]	RIGATTI S J. Random forest[J]. Journal of Insurance Medicine, 2017, 47(1):31-39. doi: 10.17849/insm-47-01-31-39.1 pmid: 28836909
[27]	BUZA K, NANOPOULOS A, NAGY G. Nearest neighbor regression in the presence of bad hubs[J]. Knowledge-Based Systems, 2015, 86:250-260.
[28]	MCDONALD G C. Ridge regression[J]. Wiley Interdisciplinary Reviews: Computational Statistics, 2009, 1(1):93-100.
[29]	AWAD M, KHANNA R. Support vector regression[J]. Neural Information Processing Letters & Reviews, 2007, 11(10):203-224.
[30]	FAWCETT T. An introduction to ROC analysis[J]. Pattern Recognition Letters, 2006, 27(8):861-874.
[31]	STOLWIJK J A. A mathematical model of physiological temperature regulation in man[R]. Nasa Langley, 1971.
[32]	WOLD S, ESBENSEN K, GELADI P. Principal component analysis[J]. Chemometrics and Intelligent Laboratory Systems, 1987, 2(1/2/3):37-52.
[33]	LOUPPE G, WEHENKEL L, SUTERA A, et al. Understanding variable importances in Forests of randomized trees[J]. Advances in Neural Information Processing Systems, 2013, 26:431-439.

组别	温度/℃	相对湿度/%	作业强度	服装
1	34	60	轻度	警服
2	36	60	轻度	警服
3	38	60	轻度	警服
4	40	60	轻度	警服
5	34	60	中度	警服
6	36	60	中度	警服
7	38	60	中度	警服
8	40	60	中度	警服
9	34	60	重度	警服
10	36	60	重度	警服
11	38	60	重度	警服
12	40	60	重度	警服
13	38	60	轻度	便服
14	34	60	中度	便服
15	36	60	中度	便服
16	38	60	中度	便服
17	40	60	中度	便服
组别	温度/℃	相对湿度/%	作业强度	服装
18	38	60	重度	便服
19	36	30	轻度	警服
20	38	30	轻度	警服

特征	性别	身高/m		体质量/kg		体脂率		年龄/ 岁	相对湿度/%		温度/ ℃	PAR		CI/ (m²·K·W^-1)
特征	男	1.8		82		25.3		20	60		34	1.4		0.077 5
标签	核心温度/℃													热应激风险
	N₁	N₂	N₃	N₄	N₅	N₆	N₇	N₈	N₉	N₁₀	N₁₁	N₁₂	N₁₃	热应激风险
	37.59	37.59	37.54	37.50	37.44	37.37	37.33	37.29	37.28	37.27	37.27	37.30	37.32	0

模型	MSE	RMSE	MAE	R²
KNN	0.003	0.053	0.010	0.986
RR	0.134	0.402	0.304	0.573
SVM	0.081	0.238	0.152	0.847
DT	0.025	0.145	0.034	0.918
GBDT	0.023	0.143	0.088	0.925
RF	0.013	0.105	0.042	0.954

因变量		MSE	RMSE	MAE	R²
核心温度	节点1	0.003	0.058	0.010	0.999
	节点2	0.004	0.065	0.013	0.995
	节点3	0.000	0.014	0.003	1.000
	节点4	0.002	0.039	0.008	0.997
	节点5	0.002	0.045	0.009	0.996
	节点6	0.009	0.093	0.013	0.977
	节点7	0.003	0.055	0.011	0.981
	节点8	0.003	0.054	0.010	0.978
	节点9	0.002	0.048	0.009	0.978
	节点10	0.002	0.045	0.009	0.979
	节点11	0.003	0.053	0.009	0.975
	节点12	0.004	0.063	0.013	0.975
	节点13	0.003	0.056	0.012	0.982

核心温度预测模型						热应激风险预测模型
RF			GBDT			RF			GBDT
特征		重要性得分	特征		重要性得分	特征		重要性得分	特征		重要性得分
环境温度		0.29	环境温度		0.3	环境温度		0.27	环境温度		0.31
PAR		0.23	PAR		0.22	PAR		0.24	PAR		0.27
相对湿度		0.18	年龄		0.19	CI		0.11	体脂率		0.12
年龄		0.13	相对湿度		0.17	体脂率		0.1	体质量		0.098
体脂率		0.074	体脂率		0.058	体质量		0.097	CI		0.083
CI	0.039		CI	0.026		身高	0.069		身高	0.067
体质量	0.03		体质量	0.016		年龄	0.052		年龄	0.036
身高	0.027		身高	0.012		相对湿度	0.036		相对湿度	0.02
性别	0.008 6		性别	0.008 1		性别	0.029		性别	0.002 9