Recognition of fatigue state of high-speed rail dispatchers based on EEG signal characteristics

doi:10.16265/j.cnki.issn1003-3033.2024.06.1674

Abstract

Abstract:

In order to enhance the stability and safety of railway driving and effectively identify the influence of the dispatcher's fatigue state on the driving organization, a method for identifying the fatigue state of the dispatcher was proposed based on the characteristics of EEG signals. The fatigue state of the dispatcher was divided according to the working time period, and the high-speed rail scheduling simulation experiment was designed to collect EEG data. The three types of brainwave frequency-domain amplitudes of high-speed rail dispatching subjects were extracted as the characteristic value by wavelet series expansion and Fourier transform, and the classification results of fatigue state were verified by combining the operation characteristics and EEG signal characteristics of dispatchers. The ResNet18+SoftMax model and MobileNet V2+SoftMax model were built through the Python language environment. The input features were converted into a three-dimensional rectangular model based on deep learning. The weights were optimized and adjusted to obtain the optimal model, so as to judge the fatigue state of high-speed rail dispatchers. The research results show that the fatigue state recognition accuracy of the participants in the high-speed rail scheduling experiment by ResNet18+SoftMax and MobileNet V2+SoftMax two models is 92.78% and 99.17%, respectively, compared with support vector machines(SVM) model to improve the awake state and fatigue state recognition accuracy, and reduce the model computing time. Among them, the MobileNet V2+SoftMax model can better identify the fatigue state of the dispatcher. With the principle of MobileNet V2+SoftMax model as the core, the potential fatigue risk of high-speed rail dispatchers under long-term working conditions can be identified more quickly and accurately.

Key words: electroencephalogram(EEG) signal, high-speed rail dispatcher, fatigue state recognition, MobileNet V2 network, ResNet18 network, SoftMax regression

CLC Number:

X912安全生理学

ZHANG Guangyuan, DENG Long, WANG Yawei, SUN Ziwei, LI Sha, CHEN Cheng. Recognition of fatigue state of high-speed rail dispatchers based on EEG signal characteristics[J]. China Safety Science Journal, 2024, 34(6): 235-246.

Figures/Tables 16

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 1

Fig.6

Table 2

Table 3

Table 4

Fig.7

Fig.8

Fig.9

Fig.10

Fig.11

Table 5

References 30

[1]	中国铁路成都局集团有限公司. 高速铁路行车组织细则[M]. 北京: 中国铁道出版社, 2018:73-75.
[2]	田起利. 列车调度员的疲劳研究[D]. 成都: 西南交通大学, 2011.
	TIAN Qili. Study on fatigue of train controller[D]. Chengdu: Southwest Jiaotong University, 2011.
[3]	王列妮, 牟瑞芳, 张磊. 我国某铁路局2005—2014年行车事故分布规律分析[J]. 安全与环境学报, 2018, 18(1):406-412.
	WANG Lieni, MU Ruifang, ZHANG Lei. Analysis of the distribution law of traffic accidents in a railway bureau in my country from 2005 to 2014[J]. Journal of Safety and Environment, 2018, 18(1):406-412.
[4]	HOCKEY G R. The psychology of fatigue: work, effort and control[M]. Cambridge: Cambridge University Press, 2013:123-126.
[5]	杨奎. 铁路列车调度员疲劳机理与发展规律研究[D]. 成都: 西南交通大学, 2017.
	YANG Kui. Mechanism and developing law of railway dispatcher's fatigue[D]. Chengdu: Southwest Jiaotong University, 2017.
[6]	LAL S, CRAIG A. A critical review of the psychophysiology of driver fatigue[J]. Biological Psychology, 2001, 55(3):173-194. doi: 10.1016/s0301-0511(00)00085-5 pmid: 11240213
[7]	JAP B T, LAL S, FISCHER P, et al. Using EEG spectral components to assess algorithms for detecting fatigue[J]. Expert Systems with Applications, 2009, 36:2352-2359.
[8]	WANG Qingjun, LI Yibo, LIU Xueping. Analysis of feature fatigue EEG signals based on wavelet entropy[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2018, 32(8):DOI: 10.1142/S021800141854023X.
[9]	张光远, 胡晋, 文原劲, 等. 高速铁路调度员疲劳程度分级及预测方法研究[J]. 铁道学报, 2021, 43(5):18-26.
	ZHANG Guangyuan, HU Jin, WEN Yuanjin, et al. Research on the classification and prediction method of fatigue degree of high-speed railway dispatchers[J]. Journal of the China Railway Society, 2021, 43(5):18-26.
[10]	TREJO L J, KNUTH K, PRADO R, et al. EEG-Based estimation of mental fatigue: convergent evidence for a three-state model[J]. Foundations of Augmented Cognition, Proceedings, 2007, 4565:201-211.
[11]	CHUA B L, DAI Z, THAKOR N, et al. Connectome pattern alterations with increment of mental fatigue in one-hour driving simulation[C]. 2017 39^th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2017:4355-4358.
[12]	张朋, 周前祥, 于洪强, 等. 基于EEG信号特征的脑力疲劳快速检测方法研究[J]. 北京航空航天大学学报, 2023, 49(1):145-154.
	ZHANG Peng, ZHOU Qianxiang, YU Hongqiang, et al. Fast detection method of mental fatigue based on EEG signal characteristics[J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49(1):145-154.
[13]	郭孜政, 牛琳博, 吴志敏, 等. 基于EEG的驾驶疲劳识别算法及其有效性验证[J]. 北京工业大学学报, 2017, 43(6):929-934.
	GUO Zizheng, NIU Linbo, WU Zhimin, et al. Driver's fatigue recognition algorithm based on EEG and its validity verification[J]. Journal of Beijing University of Technology, 2017, 43(6):929-934.
[14]	GRANDJEAN E. Fatigue in industry[J]. British Journal of Industrial Medicine, 1979, 36(3):175-186. doi: 10.1136/oem.36.3.175 pmid: 40999
[15]	郭孜政, 谭永刚, 马国忠, 等. 基于BP神经网络的驾驶精神疲劳识别方法[J]. 哈尔滨工业大学学报, 2014, 46(8):118-121.
	GUO Zizheng, TAN Yonggang, MA Guozhong, et al. Recognition method of driving mental fatigue based on BP neural network[J]. Journal of Harbin Institude of Technology, 2014, 46(8):118-121.
[16]	KAIDA K, TAKAHASHI M, AKERSTEDT T, et al. Validation of the Karolinska sleepiness scale against performance and EEG variables[J]. Clinical Neurophysiology, 2006, 117(7):1574-1581. doi: 10.1016/j.clinph.2006.03.011 pmid: 16679057
[17]	JAP B T, LAL S, FISCHER P, et al. Using EEG spectral components to assess algorithms for detecting fatigue[J]. Expert Systems with Applications An International Journal, 2009, 36(2):2352-2359.
[18]	SEN B, PEKER M, CAVUSOGLU A, et al. A comparative study on classification of sleep stage based on EEG signals using feature selection and classification algorithms[J]. Journal of Medical Systems, 2014, 38(3):DOI: 10.1007/s10916-014-0018-0.
[19]	程佳. 基于脑电信号的睡眠分期研究[D]. 北京: 北京理工大学, 2015.
	CHENG Jia. Research of sleep staging based on EEG signals[D]. Beijing: Beijing Institute of Technology, 2015.
[20]	封常生. 小波分析在信号处理中的应用[D]. 上海: 上海交通大学, 2007.
	FENG Changsheng. Application of wavelet analysis on signal processing[D]. Shanghai: Shanghai Jiaotong University, 2007.
[21]	HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]. IEEE Conference on Computer Vision & Pattern Recognition. IEEE Computer Society, 2016:770-778.
[22]	BOUGUEZZI S, FAIEDH H, SOUANI C. Slim MobileNet: an enhanced deep convolutional neural network[C]. 18^th International Multi-Conference on Systems, Signals & Devices (SSD). IEEE, 2021:12-16.
[23]	苏畅. 基于RFID的预制装配式住宅构件追踪管理研究[D]. 哈尔滨: 哈尔滨工业大学, 2012.
	SU Chang. RFID-Based tracking management system of prefabricated housing components[D]. Harbin: Harbin Institute of Technology, 2012.
[24]	STERN J M. Atlas of EEG patterns[M]. Baltimore MD: Lippincott Williams & Wilkins, 2005:377-379.
[25]	胥川, 王雪松, 陈小鸿. 无侵入测量指标的驾驶疲劳检测性能评估[J]. 西南交通大学学报, 2014, 49(4):720-726.
	XU Chuan, WANG Xuesong, CHEN Xiaohong. Evaluating performance of non-intrusive indicators on drowsy driving detection[J]. Journal of Southwest Jiaotong University, 2014, 49(4):720-726.
[26]	黄诗童, 张威强, 张朋柱. 基于HRV分析的可穿戴心电仪精神疲劳检测[J]. 计算机应用研究, 2019, 36(7):2093-2097, 2103.
	HUANG Shitong, ZHANG Weiqiang, ZHANG Pengzhu. Wearable ECG for mental fatigue detection based on HRV analysis[J]. Computer Application Research, 2019, 36(7):2093-2097, 2103.
[27]	张蕾. 管制员工作负荷评估系统及扇区容量问题研究[D]. 南京: 南京航空航天大学, 2006.
	ZHANG Lei. The research on the evaluation system of controller's workload and sector capacity[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2006.
[28]	张明岛, 陈兴时. 脑诱发电位学:第2版[M]. 上海: 上海科技教育出版社, 2000:387-390.
[29]	TING P, HWANG J, DOONG J, et al. Driver fatigue and highway driving: a simulator study[J]. Physiology & Behavior, 2008, 94(3):448-453.
[30]	RULI E, VENTURA L. Accurate likelihood inference for the volume under the ROC surface[J]. Cancer Reports, 2020, 3(4): DOI: 10.1002/cnr2.1206.

作业类型	具体工作内容
监视作业	查看列车运行计划、查看列车实绩运行信息、查看列车运行状态、查看施工维修计划
通话作业	发布电话调度命令、各调度工种间联系、发布口头指示
操作记录作业	调度命令、交班事宜、安监报

指标	P 值	T	F 值
α	1.16×10^-40	16.16	7.42
β	9.00×10^-6	4.54	36.45
θ	5.01×10^-81	-28.73	4.46

指标	综合疲劳程度值	α波	β波	θ波
综合疲劳程度值	1.000	-0.720^**	-0.391^**	0.700^**
α波	-0.720^**	1.000	0.037	-0.318^**
β波	-0.391^**	0.037	1.000	-0.055
θ波	0.700^**	-0.318^**	-0.055	1.000

模型	各状态准确率/%				AUC	运行环境及运行速度/ms
模型	清醒	轻度疲劳	疲劳	总准确率	AUC	运行环境及运行速度/ms
ResNet18+ SoftMax	95.83	85.83	96.67	92.78	0.93	GPU 2.724
MobileNet V2+SoftMax	100	97.5	100	99.17	0.945	GPU,5.585
SVM	93.3	87.5	92.3	91.6	0.91	CPU,速度慢