煤矿安全生产取决于“人、机、环、管”四大要素之间互相联合与紧密配合，在 煤矿行业高新技术及智能化发展的今天，“人”这一要素在整个行业生产过程中扮演着 越来越重要的角色。作为高风险职业群体，矿工上岗时身心是否处于安全状态对煤矿 安全生产工作尤为重要，因此研究煤矿从业人员不安全状态检测方法对预防事故、提 升煤矿安全管理水平具有重要意义。本文提出一种基于生理参数的煤矿从业人员不安 全状态检测方法，在探究不安全状态与生理信号之间关系的基础上，构想设计出一种 检测装置，该装置可完成生理信息数据的采集、处理、计算、显示、传输等功能，并 通过 BP 神经网络分类算法实现不安全状态的有效识别。 首先，从不安全状态的概念发展、产生因素和表现形式三个方面阐述了不安全状 态的研究基础；通过详细分析对比适用于不安全状态检测的方法，确定选用生理检测 法。通过探究生理信息与人的状态之间的关联性，建立了不安全状态与生理参数之间 的关系，根据不同生理信号的特点，确定本课题采用心率、血压、血氧饱和度、体温 作为生理检测指标。对生理信号测量原理和方法进行详细阐述，选择光电容积描记 （PPG）法检测心率血压，光学法测量血压，采用接触式体温测量。 然后，依据模块化思想将整体设计分为主控模块、生理参数采集模块、显示模块、 通信模块、警报模块以及电源电路，对硬件部分进行设计。硬件部分包括元器件选型 和连接电路设计，以 STM32C8T6 为微控中心，选用 MKB0803、MAX30102、 DS18B20 传感器组成生理参数采集模块，实现生理参数测量；使用 LCD、有源蜂鸣器 以及 USB 线路完成参数显示、预警和传输功能。该一体化式检测装置满足低功耗、小 型化、便携化的设计需求。 其次，基于 Keil μVision5 IDE 对检测装置软件部分进行设计，包括系统主程序与 其他各个模块相对应的驱动程序设计和代码编写，完成生理参数的采集与传输。将生 理信号特征转换成具体的生理数值在液晶屏上实时显示，通过有线传输发送至电脑端 动态显示并实现数据保存。 最后，将设计的装置进行功能测试和对比实验，结果表明该装置能够较为准确测 量生理数据并与电脑端进行实时传输。通过不安全状态诱发实验，采集实验对象不同 状态下的生理参数，使用 BP 神经网络分类算法对数据进行训练分类，分类准确率为 92%，最小误差为 0.0003，分类效果较为理想，证明利用生理参数能够实现不安全状 态有效分类识别。 本文研究设计的生理检测装置可以实现矿工生理信息参数的检测，并通过 BP 神经 网络实现不安全状态的分类识别。实验表明，生理参数可用于矿工不安全状态的检测 与识别。

Coal mine safety depends on the interaction and close cooperation between the four aspects of "man, machine, environment and management". In today's high-tech and intelligent development of the coal mining industry, the "human" factor plays a pivotal role in the overall development of industry. As a high-risk occupational group, whether miners are in a safe state physically and mentally when they are on duty is particularly important for coal mine safety work. Therefore, the study of miners' unsafe condition detection methods is of great significance to prevent accidents and improve coal mine safety management. This paper proposes a physiological signal-based miner unsafe condition detection method. On the basis of exploring the relationship between unsafe condition and physiological signals, a detection device is conceived and designed, which can complete the collection and processing of physiological information data, calculation, display, transmission and other functions, through the BP neural network classification algorithm to achieve effective identification of unsafe conditions. Firstly, the research basis of unsafe condition is expounded from three aspects: the conceptual development, generating factors and the manifestation of the unsafe condition; through detailed comparison and analysis of methods suitable for unsafe condition detection, the physiological detection method is determined to be selected. By exploring the correlation between physiological information and human state, the relationship between unsafe condition and physiological parameters was established. Heart rate, blood pressure, oxygen saturation, and body temperature were determined as the physiological detection indexes used in this project according to the f different eatures of physiological parameters. Detailed explanation of physiological signal measurement principles and methods, and the photoelectric volumetric tracing (PPG) method is selected to detect heart rate and blood pressure, and the optical method is used to measure blood pressure, and contact temperature is used to measure body temperature. Secondly, the overall design was divided into main control module, physiological parameter acquisition module, display module, communication module, alarm module and power supply circuit based on the modular idea for the hardware part. STM32C8T6 was used as the micro-control center. MKB0803, MAX30102 and DS18B20 sensors were used to complete the combination of physiological parameter acquisition modules to achieve physiological parameter measurement. LCD, active buzzer and USB line were used to complete data display, warning and transmission functions. The integrated detection device meets the design requirements of low power consumption, miniaturization and portability. Then, the software part of the detection device was designed based on Keil μVision5 IDE, including the design and code writing of the driver corresponding to the functions of the main system program and other parts. The physiological signal characteristics are converted into specific physiological values displayed in real time on the LCD screen, sent to the computer via wired transmission for dynamic display and data storage. Finally, the designed device is tested and compared. Results showed that the device can accurately measure physiological data and transmit it to the computer in real time. Through the unsafe condition induced experiment, the physiological parameters of the experimental subjects in different states were collected. Using BP neural network classification algorithm to train the data for classification, the accuracy was obtained as 92%, and the minimum error was 0.0003. The weights and thresholds were obtained by PSO optimization algorithm, and then input random data for verification, which proved that the use of physiological parameters can achieve effective classification and identification of unsafe conditions. The physiological detection device designed in this paper can enable the detection of miners' physiological information parameters, and achieve the classification and identification of unsafe conditions through BP neural network. Experiments show that physiological parameters can be used to detect and identify miners' unsafe conditions.  

[1] 孙春升, 宋晓波, 弓海军. 煤矿智慧矿山建设策略研究[J]. 煤炭工程, 2021, 53(02): 191-196.

[2] Qiao W G, Liu Q L, Li X C, et al. Using data mining techniques to analyze the influencing factor of unsafe behaviors in Chinese underground coal mines[J]. Resources Policy, 2018(59): 210-216.

[3] 黄辉, 张雪. 煤矿员工不安全行为研究综述[J]. 煤炭工程, 2018, 50(06): 123-127.

[4] 焦彬. 基于机器视觉多信息融合的疲劳状态检测[D]. 天津大学, 2016.

[5] 陈晓勇, 施式亮, 李润求,等. 基于修正的 HFACS 与 SPA 的建筑施工安全人因分析[J]. 湖南科技大学学报(自然科学版), 2020, 35(03): 63-69.

[6] 董建梁. 基于脑电检测的矿工生理与心理疲劳监测系统[D]. 煤炭科学研究总院, 2014.

[7] 孟亦凡, 李敬兆, 张梅. 基于 LSTM 边缘计算与随机森林雾决策的矿工状态监测设备[J]. 煤矿机械, 2018, 39(11): 144-148.

[8] Chiu H Y, Wang M Y, Chang C K, et al. Early morning awakening and nonrestorative sleep areassociated with increased minor non-fatal accidents during work and leisure time[J]. Accident Analysis and Prevention, 2014, 71(10): 10-14.

[9] Lu J T, Yang N D, Ye J F, et al. Differential effects of specific negative emotions on individual risk preference behaviors under social accidents: An analysis from the perspective of affective computing theory[J]. Revista de Cercetare si Interventie Sociala, 2015, 50: 172-192.

[10]Petitta L, Probst T M, Ghezzi V, et al. Emotional contagion as a trigger for moral disengagement: Their effects on workplace injuries[J]. Safety Science, 2021, 140(9): 105317.

[11]You X D, Zhu L, Liu Z G, et al. Experimental study on the relationship between fatigue and unsafe behavior of urban rail transit drivers[J]. Transportation Research Record Journal of the Transportation Research Board, 2021(13): 036119812110148.

[12]Larsson S, Pousette A, Torner M. Psychological climate and safety in the construction industry-mediated influence on safety behaviour[J]. Safety Science, 2008, 46(3): 405-412.

[13]Boschman J S, Van D M H F, Sluiter J K, et al. Psychosocial work environment and mental health among construction workers[J]. Applied Ergonomics, 2013, 44(5): 748-755.

[14]Alavinia S M, Duivenbooden C V , Burdorf A. Influence of work-related factors and individual characteristics on work ability among Dutch construction workers[J]. Scandinavian Journal of Work, Environment & Health, 2007, 33(5): 351-357.

[15] Raman P, Sambasivan M, Kumar N. Counterproductive work behavior among frontline government employees: Role of personality, emotional intelligence, affectivity, emotional labor, and emotional exhaustion[J]. Journal of Work and Organizational Psychology, 2016, 32(1): 25-37.

[16]李乃文, 刘健, 牛莉霞. 基于扎根理论的矿工心智游移质性研究[J]. 安全与环境学报, 2018, 18(05): 1868-1871.

[17]李付伟. 情绪对企业员工不安全状态风险感知的认知心理研究[D]. 燕山大学, 2021.

[18]李广利, 田水承, 严一知,等. 矿工关键不安全情绪识别研究[J]. 西安科技大学学报, 2021, 41(05): 793-799.

[19]刘琦, 孔琪, 张丽艳. 矿工心理与安全生产研究[J]. 中国煤炭, 2010, 36(03): 107-110.

[20]田水承, 孙璐瑶, 唐艺璇,等. 矿工不安全状态评价指标体系的构建与分析[J]. 西安科技大学学报, 2021, 41(03): 402-409.

[21]Behr C J, Kumar A, Hancke G P. A smart helmet for air quality and hazardous event detection for the mining industry[C]// IEEE International Conference on Industrial Technology. IEEE, 2016.

[22]Shabina S. Smart helmet using RF and WSN technology for underground mines safety[C]// International Conference on Intelligent Computing Applications. IEEE, 2014: 305-309.

[23]Porselvi T, Janaki B, Priyadarshini K, et al. IoT based coal mine safety and health monitoring system using LoRa WAN[C]// 2021 3rd International Conference on Signal Processing and Communication (ICPSC), 2021: 49-53.

[24]张春堂, 管利聪. 基于 SSD-MobileNet 的矿工安保穿戴设备检测系统[J]. 工矿自动化, 2019, 45(06): 96-100.

[25]郭秀才, 曹泰铭. 基于 ZigBee 的井下人员监测系统设计[J]. 电脑编程技巧与维护, 2013(08): 104-105+124.

[26]是利娜. 井下人员跟踪监测系统的设计[J]. 能源技术与管理, 2015,40(03): 144-146.

[27]金鹏, 薛哲彬, 戈垚. 具有实时瓦斯监测功能的新型智能矿工服设计[J]. 纺织学报, 2020, 41(11): 143-149.

[28]Wang M, Zhang S, Lv Y, et al. Anxiety level detection using BCI of miner’s smart

helmet[J]. Mobile Networks and Applications, 2018, 23(2): 336-343.

[29]Sarkar S, Ghosh S S. Study of cardiorespiratory and sweat monitoring wearable

architecture for coal mine workers[C]// 2020 IEEE REGION 10 CONFERENCE

(TENCON). IEEE, 2020.

[30]Sun X D, Li Y F, Chen C, et al. Multi-function fast networking rescue life escort Benan watch[C]// 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC), 2020: 1626-1630.

[31]冯禹, 刘军. 采矿工人生理状况监测系统设计[J]. 电子科技, 2012, 25(08): 96-99+103.

[32]黄剑波, 朱宗玖, 徐欢. 基于 ZigBee 的井下人员生理指标监测系统设计[J]. 工矿自动化, 2014, 40(02): 8-10.

[33]谢红卫, 孙志强, 李欣欣等. 典型人因可靠性分析方法评述[J]. 国防科技大学学报, 2007(02): 101-107.

[34]Reason J. Human error[M]// Cambridge, UK: Cambridge University Press, 1990.

[35]伍星光, 侯磊, 刘芳媛等. 安全思维和事故模型研究分析与展望[J]. 石油科学通报, 2020, 5(02): 254-268.

[36]吕春玉, 房春花. 人为因素分析与分类系统(HFACS)及事故个例分析[J]. 中国民航飞行学院学报, 2009, 20(02): 37-40.

[37]宋桂蓉, 曹宏, 徐峰. 基于人因可靠性的数控加工中心安全管理模型分析[J]. 数学的实践与认识, 2017, 47(06): 287-291.

[38]卢奎. “亚安全”状态下深化本质安全管理的途径和方法初探[J]. 北京石油管理干部学院学报, 2016, 23(03): 3-7+21.

[39]李亮, 薛剑恩. 引航员亚安全心理分析及控制[J]. 世界海运, 2020, 43(05): 29-31+51.

[40]田水承, 王雪晨, 苗彦平,等. 基于扎根理论的矿工不安全状态影响因素研究[J]. 煤矿安全, 2022, 53(02): 252-256.

[41]沈健. 基于脑电和人脸表情的双模态情绪识别系统[D]. 南京邮电大学, 2020.

[42]范如俊. 基于多生理信息的情绪识别平台建立[D]. 长春大学, 2020.

[43]林岭, 张力为. 可以用生理生化指标检测评价运动性心理疲劳吗?[J]. 中国运动医学杂志, 2005(06): 92-96.

[44]王健琪, 薛慧君, 吕昊,等. 非接触生理信号检测技术[J]. 中国医疗设备, 2013, 28(11): 5-8+80.

[45]付莉娜, 刘淑慧, 潘婷,等. 健康指数检测与情绪管理智能设备的设计与开发[J]. 电子制作, 2019(13): 3-6.

[46]谢科, 李婷婷, 刁联旺. 多模态生理指标在指挥员工作状态监测系统中的应用设计

[C]// 第六届中国指挥控制大会论文集(上册), 2018:331-335.

[47]靳慧斌, 朱国蕾, 吕川. 基于 CART 决策树的管制员疲劳检测研究[J]. 航天医学与医学工程, 2018, 31(06): 601-606.

[48]石力涛, 杨荣, 李中宾,等. 中重度颅脑损伤患者治疗中结合有创动态颅内压监测的机制和临床价值[J]. 脑与神经疾病杂志, 2021, 29(07): 427-430.

[49]朱浩然. 有创血压模拟系统的研制[D]. 电子科技大学, 2021.

[50]邹国宝. 不同检测仪器对血糖检测方法研究现状与进展[J]. 中国医疗器械信息,2019, 25(16): 30-31.

[51]杨雷, 王志武, 姜萍萍,等. 无线健康监护系统设计及状态识别算法[J]. 中国医疗器械杂志, 2020, 44(04): 288-293.

[52]Cosimo Carriero. 生命体征监测技术: 对人体实施状态监控[J]. 电子产品世界, 2020, 27(11): 19-21.

[53]刘璐瑶, 张森, 肖文栋. 基于小波分析和自相关计算的非接触式生理信号检测[J]. 工程科学学报, 2021, 43(09): 1206-1214.

[54]杜振伟, 孙建, 秦明新,等. 基于磁感应技术的新型非接触心肺信号检测系统的研究[J]. 中国医学物理学杂志, 2014, 31(06): 5288-5294+5317.

[55]Cornacchia M, Ozcan K, Zheng Y, et al. A survey on activity detection and classification using wearable sensors[J]. IEEE Sensors Journal,, 2017, 17(2): 386-403.

[56]Shoaib M, Bosch S, Incel O D, et al. Complex human activity recognition using

smartphone and wrist-worn motion sensors[J]. Sensors, 2016, 16: 426-449.

[57]Pandit D, Zhang L, Liu C, et al. A lightweight QRS detector for single lead ECG signals

using a max-min difference algorithm [J]. Comput Methods Programs Biomed, 2017, 144: 61-75.

[58]刘艳, 刘鼎家, 韩智攀. 基于动作识别的情绪提取方法研究[J]. 计算机工程, 2015, 41(05): 300-305.

[59]张继奎, 张志公, 祝凯,等. 基于罪犯危险行为预测的视频报警机制的研究[J]. 中国新通信, 2020, 22(19): 108-109.

[60]车丹丹. 矿工警觉度对不安全行为的影响研究[D]. 西安科技大学, 2017.

[61]王娆芬, 许伟, 王行愚. 操作员功能状态的智能评估[J]. 自动化仪表, 2014, 35(02): 9-12.

[62]孟晓静, 贾开发, 张大卫. 高温作业环境下人体生理状态模糊综合评价[J]. 安全与环境学报, 2021, 21(04): 1630-1635.

[63]王宇, 王健新. 基于生理特征的应急情境下管制员安全行为能力测评[J]. 科技与创新, 2018(24): 137-138+141.

[64]王进州, 艾力·斯木吐拉. 基于驾驶员生理负荷的高原公路转角值安全风险评价[J]. 科学技术与工程, 2020, 20(07): 2939-2943.

[65]荣明, 陈英杰, 黄超等. 基于 OCSVM 的隧道人员安全检测技术的研究与应用[J]. 隧道建设(中英文), 2021, 41(12): 2122-2132.

[66]He J B, Choi W, Yang Y, et al. Detection of driver drowsiness using wearable devices: A feasibility study of the proximity sensor[J].

Applied Ergonomics, 2017, 65: 473-480.

[67]Tan A Y, Nearing B D, Rosenberg M, et al. Interlead heterogeneity of R-and T-wave morphology in standard 12-lead ECGs predicts sustained ventricular tachycardia /fibrillation and arrhythmic death in patients with cardiomyopathy[J]. Journal of Cardiovascular Electrophysiology, 2017, 28(11): 1324-1333.

[68]Healy M, Donovan R, Walsh P, el at. A machine learning emotion detection platform to support affective well being[J]. 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2018: 2694-2700.

[69]刘梅颜. 精神压力和心血管损伤[J]. 中华内科杂志, 2021, 60(09): 846-848.

[70]于汶, 陶淑慧. 精神压力相关高血压[J]. 中华内科杂志, 2021, 60(08): 769-772.

[71]Shao D, Liu C, Tsow F, et al. Noncontact monitoring of blood oxygen saturation using camera and dual-wavelength imaging system[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(6): 1091-1098.

[72]陈天启, 张玉乾, 宗宝超,等. 基于心率及呼吸生命特征监测技术发展和应用[J]. 中国医疗器械杂志, 2021, 45(02): 188-193.

[73]李万磊. 面向智能穿戴的人体生理参数监测系统研发[D]. 西南交通大学, 2019.

[74]田泽懿, 张磊, 单新治,等. 基于脉搏波传导时间的血压检测研究进展[J]. 光学仪器, 2020, 42(01): 88-94.

[75]梁永波, 陈真诚, 朱健铭,等. 基于容积脉搏波的无创连续血压测量系统[J]. 航天医学与医学工程, 2013, 26(01): 47-50.

[76]谢运泉, 李宏, 邬杨波等. 一种基于血氧饱和度与鼾声检测的呼吸暂停监测系统设计[J]. 宁波大学学报(理工版), 2021, 34(02): 9-16.

[77]廖国杰, 张志强, 高博等. 反射式血氧饱和度检测方法的研究[J]. 光散射学报, 2009, 21(03): 274-278.

[78]CSDN. MAX30102 血氧传感器的实际标定公式[EB/OL]. (2018-4-25). https://blog.csdn.net/u011419265/article/details/80078913.

[79]王大任. 人体多生理参数监测系统研制[D]. 吉林大学, 2019..

[80]刘永平, 谢彩洪, 刘凌昕. 电子体温计在临床使用的现状分析[J]. 中国医疗设备, 2018, 33(04): 111-114..

[81]能明凯, 周广明, 赵伟,等. 老年人健康检测系统的设计[J]. 现代计算机, 2020(16): 33-36.

[82]白鹏飞, 刘强, 段飞波,等. 基于 MAX30102 的穿戴式血氧饱和度检测系统[J]. 激光与红外, 2017, 47(10): 1276-1280.

[83]张天骐. 浅析 DS18B20 数字温度计[J]. 科学技术创新, 2021(22): 165-166.

[84]赵华峰. LCD1602 模块的汉字显示研究[J]. 现代信息科技, 2020, 4(17): 35-37.

[85]张云柯. 远程智能火灾监控报警及控制系统的设计与实现[D]. 西安建筑科技大学,

2018.

[86]肖艳军, 毛哲, 温博,等. 基于 STM32 的综合实验平台设计[J]. 实验技术与管理, 2019, 36(12): 72-76.

[87]张文宇. 基于 STC 单片机的 8X8X8 光立方设计[J]. 电子世界, 2020(22): 206-207.

[88]黄洁波. 基于 UWB 的人员定位及生理参数监测的研究[D]. 广东工业大学, 2017.

[89]赵亮. 液晶显示模块 LCD1602 应用[J]. 电子制作, 2007(03): 58-59.

[90]杨娱琦, 朱晟. 基于 BP 神经网络的膨胀土判别分级方法研究[J]. 水力发电, 2022, 48(03): 24-29+93.

[91]闫鹏程, 尚松行, 张超银,等. 改进 BP 神经网络算法对煤矿水源的分类研究[J]. 光谱学与光谱分析, 2021, 41(07): 2288-2293.

[92]王莉, 张紫烨, 郭晓东,等. 基于粒子群优化 BP 神经网络的心电信号分类方法[J].

自动化与仪表, 2019, 34(09): 84-87+93.

[93]仇志凡. 大型飞机操纵品质等级评估方法研究[D]. 南京航空航天大学 2019.

[94]葛均波, 徐永健, 王辰. 内科学(第九版)[M]// 北京: 人民卫生出版社, 2018.

[95]金刚. 矿工入井前生理状态快速评价方法研究[D]. 西安科技大学, 2021.

[96]沈志远, 魏义涛, 闫永刚,等. 空中交通管制员疲劳检测与管理综述[J]. 航空工程进展, 2021, 12(06): 26-38.

[97]Xing X J, Zhong B T, Luo H B, et al. Effects of physical fatigue on the induction of mental fatigue of construction workers: A pilot study based on a neurophysiological approach[J]. Automation in Construction, 2020, 120(103381).