ISBOA-KELM multi-sensor data fusion model for early warning method in laboratory safety

doi:10.16265/j.cnki.issn1003-3033.2026.01.1133

Abstract

Abstract:

To address the challenges of complex data environments, low accuracy of single-sensor detection, and limited precision in traditional laboratory safety systems, this study presented a multi-sensor fusion early warning model based on an ISBOA and algorithm KELM. First, the KELM framework was employed to integrate heterogeneous sensor data and construct the warning model, where a regularization term was introduced to alleviate overfitting. Then, the improved ISBOA adaptively optimized the regularization coefficient C and kernel parameterσ of the KELM, thereby enhancing parameter robustness and diagnostic accuracy. Finally, simulation and experimental analyses were conducted using both synthetic and real laboratory datasets, and the proposed ISBOA-KELM model was compared with the unimproved Secretary Bird Optimization Algorithm (SBOA), Particle Swarm Optimization (PSO), and Gray Wolf Optimization (GWO) algorithms. The experimental results show that the ISBOA-KELM model improved accuracy by 4%, 3%, and 2%, respectively, compared with the other three models. In four representative laboratory safety scenarios, including fire and gas leakage, the detection accuracy exceeds 96% with the false negative rate below 6%, which significantly improves the reliability and robustness of safety accident early warning.

Key words: laboratory safety, improved secretary bird optimization algorithm (ISBOA), kernel extreme learning machine (KELM), multi-sensor data fusion, intelligent early warning

CLC Number:

X948

GE Liang, ZHOU Nüqing, CHE Honglei, XIAO Guoqing, LAI Xi, ZENG Wen. ISBOA-KELM multi-sensor data fusion model for early warning method in laboratory safety[J]. China Safety Science Journal, 2026, 36(1): 63-71.

Figures/Tables 13

Table 1

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 2

Table 3

Fig.6

Fig.7

Fig.8

Fig.9

Table 3

References 24

[1]	王维, 马麟. 高校实验室安全事故人为因素及其影响路径分析[J]. 实验室研究与探索, 2024, 43(7):264-268.
	WANG Wei, MA Lin. Analysis of human factors of laboratory safety accidents in universities and their influence paths[J]. Research and Exploration in Laboratory, 2024, 43(7):264-268.
[2]	王红力, 杨鑫刚. 基于安全控制结构模型的高校实验室安全风险动态评估[J]. 中国安全科学学报, 2024, 34(8): 61-68. doi: 10.16265/j.cnki.issn1003-3033.2024.08.1867
	WANG Hongli, YANG Xin'gang. Safety risk dynamic assessment in university laboratories based on safety control structure model[J]. China Safety Science Journal, 2024, 34(8): 61-68. doi: 10.16265/j.cnki.issn1003-3033.2024.08.1867
[3]	艾德生. 高校实验室安全形势与任务[J]. 实验技术与管理, 2025, 42(1):1-10.
	AI Desheng. Landscape and imperatives of laboratory safety in universities[J]. Experimental Technology and Management, 2025, 42(1): 1-10.
[4]	北京市昌平区人民政府. 石油化工研究院“7.21”事故[EB/OL]. [2022-07-27]. https://www.bjchp.gov.cn/cpqzf/xxgk2671/zdlyxxgk57/aqyj/sgkb/2024050507171792165/index.html.
[5]	张亚洲, 赵小强, 惠永永, 等. 基于多传感器数据融合的 SA-DACNN 齿轮箱故障诊断方法[J]. 控制与决策, 2024, 39(11): 3699-3708.
	ZHANG Yazhou, ZHAO Xiaoqiang, HUI Yongyong, et al. SA-DACNN gearbox fault diagnosis method based on multi-sensor datafusion[J]. Control and Decision, 2024, 39(11): 3699-3708.
[6]	张兢, 路彦, 雷刚. 数据融合技术在火灾自动探测中的应用研究[J]. 计算机测量与控制, 2006, 14(5):610-613.
	ZHANG Jing, LU Yan, LEI Gang. Application research of data fusion technology in automatic fire detection system[J]. Computer Measurement & Control, 2006, 14(5): 610-613.
[7]	聂珲, 陈海峰. 基于NB-IoT环境监测的多传感器数据融合技术[J]. 传感技术学报, 2020, 33(1):144-152.
	NIE Hui, CHEN Haifeng. Multi-sensor data fusion technology based on NB-IoT environmental monitoring[J]. Chinese Journal of Sensors and Actuators, 2020, 33(1): 144-152.
[8]	杨同标, 林王兵, 成艳, 等. 基于多信息融合的高校实验室安全预警研究[J]. 科学技术创新, 2021, 25(34): 51-54.
	YANG Tongbiao, LIN Wangbing, CHENG Yan, et al. Research on safety early warning of university laboratories based on multi-information fusion[J]. Heilongjiang Science and Technology Information, 2021, 25(34): 51-54.
[9]	罗焕芝, 王骥. 面向农业温室环境的ICDO-RBFNN多传感器数据融合算法[J]. 农业工程学报, 2024, 40(21):184-191.
	LUO Huanzhi, WANG Ji. ICDO-RBFNN multi-sensor data fusion for agricultural greenhouse environment[J]. Transactions of the Chinese Society of Agricultural Engineering, 2024, 40(21): 184-191.
[10]	张龙祥, 冯全源. 基于信息融合的燃气环境监测技术研究[J]. 仪表技术与传感器, 2024, 61(7):110-115.
	ZHANG Longxiang, FENG Quanyuan. Research on information fusion-based gas environmental monitoring technology[J]. Instrument Technique and Sensor, 2024, 61(7): 110-115.
[11]	LI Xingdong, XIAO Xu, CHEN Xiandong, et al. Method of fire point real-time positioning based on multi-sensors system of UAV[J]. Fire Safety Science, 2022, 31(1): 35-45.
[12]	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. doi: 10.16265/j.cnki.issn1003-3033.2018.05.013
	HE Yongbo, DONG Yushan, LEI Jian, et al. Research on multi-sensor method for detecting aircraft cargo fire[J]. China Safety Science Journal, 2018, 28(5): 74-79. doi: 10.16265/j.cnki.issn1003-3033.2018.05.013
[13]	鲜晓东, 常超, 胡颖, 等. 基于WSNs和GSM的室内环境监测预警系统设计[J]. 传感器与微系统, 2011, 30(6):141-144.
	XIAN Xiaodong, CHANG Chao, HU Ying, et al. Design of an early-warning system based on WSNs and GSM for indoor environment monitoring[J]. Transducer and Microsystem Technologies, 2011, 30(6): 141-144.
[14]	赵地, 郭宇泰, 杜玉红, 等. 高校机械类实验室安全预警与监测智能巡检机器人设计[J]. 实验技术与管理, 2023, 40(10): 242-247.
	ZHAO Di, GUO Yutai, DU Yuhong, et al. Design of intelligent inspection robot for safety warning and monitoring of mechanical laboratories in universities[J]. Experimental Technology and Management, 2023, 40(10): 242-247.
[15]	BETHARAJOO S K, PARAMASIVAM S, CHUA Huangshen. Safety risk assessment and warning system at engineering laboratory through IoT technologies[J]. Lecture Notes in Electrical Engineering, 2021, 741: 13-23.
[16]	DING Shifei, ZHAO Han, ZHANG Yanan, et al. Extreme learning machine: algorithm, theory and applications[J]. Artificial Intelligence Review, 2015, 44(1): 103-115. doi: 10.1007/s10462-013-9405-z
[17]	SUN Shaolong, WEI Yunjie, TSUI K, et al. Forecasting tourist arrivals with machine learning and Internet search index[J]. Tourism Management, 2019, 70: 1-10. doi: 10.1016/j.tourman.2018.07.010
[18]	DING Yunfei, CHEN Zijun, ZHANG Hongwei, et al. A short-term wind power prediction model based on CEEMD and WOA-KELM[J]. Renewable Energy, 2022, 189: 188-198. doi: 10.1016/j.renene.2022.02.108
[19]	LIANG Jing, YAN Chunhua, ZHANG Ying, et al. Rapid discrimination of Salvia miltiorrhiza according to their geographical regions by laser induced breakdown spectroscopy (LIBS) and particle swarm optimization-kernel extreme learning machine (PSO-KELM)[J]. Chemometrics and Intelligent Laboratory Systems, 2020, 197: DOI:10.1016/j.chemolab.2020.103930.
[20]	SAHIL D, VIRENDRA P V. GA based KELM optimization for ECG classification[J]. Procedia Computer Science, 2020, 167: 580-588. doi: 10.1016/j.procs.2020.03.322
[21]	祁云, 薛凯隆, 李绪萍, 等. 多策略改进SSA优化KELM的边坡稳定性预测模型[J]. 中国安全科学学报, 2025, 35(3):92-98. doi: 10.16265/j.cnki.issn1003-3033.2025.03.0134
	QI Yun, XUE Kailong, LI Xuping, et al. Slope stability prediction model based on multi-strategy improved SSA for optimizing KELM[J]. China Safety Science Journal, 2025, 35(3): 92-98. doi: 10.16265/j.cnki.issn1003-3033.2025.03.0134
[22]	OSCAR D L, FERNANDO I G, CLAUDIA G, et al. Success-based optimization algorithm (SBOA): development and enhancement of a metaheuristic optimizer[J]. Computers & Chemical Engineering, 2025, 194: DOI:10.1016/j.compchemeng.2024.108987.
[23]	XIE Minghao, ZHANG Linxuan, DONG Xiaogang, et al. Prediction method of dissolved gas volume fraction in transformer oil based on OVMD-HWOA-KELM model[J]. High Voltage Engineering, 2024, 50(8): 3793-3804.
[24]	HU Qin, ZHOU Haiting, WANG Chengcheng, et al. Time-frequency fusion features-based GSWOA-KELM model for gear fault diagnosis[J]. Lubricants, 2023, 12(1): DOI:10.3390/lubricants12010010.

事故类型	主要成因	造成后果
火灾	易燃化学品靠近热源; 操作不当引发明火	设备毁损、人员伤亡、火势蔓延
爆炸	可燃气体泄漏;反应剧烈未控温控压;器皿密封不当	爆炸冲击、实验室破坏、重大伤亡
气体泄漏	管道老化、密封失效、阀门操作错误	中毒、环境污染、人员健康受损
电气故障	超负荷运行、私拉乱接、线路老化	短路、触电、引发火灾
漏水	水管破裂、接头松动、无人值守操作	设备受潮、数据丢失、电气隐患

检测模型	参数类型	准确率	总体精度
SVM	火灾	72.5	84
	可燃气体泄漏	87.5
	电气故障	77.5
	渗水漏水	100
	正常状态	82.5
CNN	火灾	75	83.5
	可燃气体泄漏	92.5
	电气故障	80
	渗水漏水	95
	正常状态	75
ELM	火灾	72.5	81.5
	可燃气体泄漏	85
	电气故障	72.5
	渗水漏水	100
	正常状态	77.5
KELM	火灾	75	85
	可燃气体泄漏	87.5
	电气故障	82.5
	渗水漏水	100
	正常状态	80

事故类型	样本数	特征维数	状态标签
正常状态	1~200	10	1
火灾	201~400	10	2
可燃气体泄漏	401~600	10	3
电气故障	601~800	10	4
渗水漏水	801~1 000	10	5

序号	试验类型	试验次数	报警次数	准确率/%	未报警次数	漏报率/%	平均报警时间/ s
1	火灾	50	47	94	3	6	13.58
2	可燃气体泄漏	50	49	98	1	2	17.33
3	电气事故	50	48	96	2	4	11.6
4	渗水漏水	50	50	100	0	0	3.5