面向煤矿混合气体检测的神经网络算法研究进展

焦明之; 沈中丽; 周扬明; 何新建; 贺耀宜

doi:10.13272/j.issn.1671-251x.18105

面向煤矿混合气体检测的神经网络算法研究进展

doi: 10.13272/j.issn.1671-251x.18105

焦明之^{1, 2, 3, 4,},
沈中丽^{1, 2},
周扬明^{1, 2},
何新建³,
贺耀宜^4, ,

1.
矿山互联网应用技术国家地方联合工程实验室，江苏徐州　221116
2.
中国矿业大学信息与控制工程学院，江苏徐州　221116
3.
中国矿业大学安全学院，江苏徐州　221116
4.
天地（常州）自动化股份有限公司，江苏常州　213015

基金项目: 国家自然科学基金青年基金项目（62204260，52174222）；科技部长三角联合攻关项目（2022CSJGG0703）；天地科技股份有限公司科技创新创业资金专项项目（2019-TD-ZD007）。

详细信息

作者简介:
焦明之（1987—），男，安徽安庆人，副研究员，研究方向为智能传感器及煤矿安全监测监控技术，E-mail：mingzhijiao@cumt.edu.cn

通讯作者:
贺耀宜（1974—），男，陕西蓝田人，研究员，硕士，主要从事煤矿监测监控、物联网与信息化应用研究工作，E-mail： hyy@cari.com.cn。

中图分类号: TD67
计量
- 文章访问数: 998
- HTML全文浏览量: 55
- PDF下载量: 60
- 被引次数: 0
出版历程
- 收稿日期: 2023-04-10
- 修回日期: 2023-08-20
- 网络出版日期: 2023-09-27

Research progress on neural network algorithms for mixed gas detection in coal mines

JIAO Mingzhi^{1, 2, 3, 4
,},
SHEN Zhongli^{1, 2},
ZHOU Yangming^{1, 2},
HE Xinjian³,
HE Yaoyi^{4
, ,}

1.
National and Local Joint Engineering Laboratory of Internet Application Technology on Mine, China University of Mining and Technology, Xuzhou 221116, China
2.
School of Information and Control Engineering, China University of Mining and Technology, Xuzhou 221116, China
3.
School of Safety Engineering, China University of Mining and Technology, Xuzhou 221116, China
4.
Tiandi (Changzhou) Automation Co., Ltd., Changzhou 213015, China

摘要

摘要: 煤矿气体传感器用于混合气体检测时测量信号之间有交叉干扰，难以保证检测准确性。对于相同的待识别气体，传统气体识别算法的识别精度低于基于神经网络的气体识别算法，神经网络通过调整其网络层、每层神经元的数量、神经元的激活函数和各层网络之间的权重等来实现更高的气体识别精度。介绍了煤矿混合气体检测系统结构，通过构建气体传感阵列，利用其多维空间气体响应模式，并结合特定的气体识别算法，实现对混合气体的定性定量识别。重点分析了几种面向煤矿混合气体检测的神经网络算法并进行了对比分析，主要包括反向传播（BP）神经网络、卷积神经网络（CNN）、循环神经网络（RNN）、径向基函数（RBF）神经网络：BP神经网络通常可以达到较高的分类精度，然而需要训练大量的参数，训练时间长，通常为了减少时长和提高精度，可以将BP神经网络与其他算法相结合；CNN可以自动提取数据特征，精度和训练速度都优于BP神经网络，但其易于陷入局部最优；RNN可以使用更少的数据并提取更有效的特征，但容易出现梯度消失等问题；RBF神经网络具有较强的鲁棒性和在线学习能力，但其通常需要大量数据完成模型训练。神经网络算法的应用将大幅提升煤矿混合气体的检测精度，保障煤矿智能化的实现。
- 煤矿安全监控 /
- 混合气体检测 /
- 神经网络算法 /
- 传感器阵列 /
- 反向传播神经网络 /
- 卷积神经网络 /
- 循环神经网络 /
- 径向基函数神经网络
Abstract: When coal mine gas sensors are used for mixed gas detection, there is cross interference between measurement signals. It is difficult to ensure detection accuracy. For the same gas to be identified, the recognition precision of traditional gas recognition algorithms is lower than that of gas recognition algorithms based on neural networks. Neural networks achieve higher gas recognition accuracy by adjusting their network layers, the number of neurons in each layer, the activation function of neurons, and the weights between each layer of networks. This paper introduces the structure of a coal mine mixed gas detection system. By constructing a gas sensor array, utilizing its multi-dimensional gas response mode, and combining specific gas recognition algorithms, the qualitative and quantitative recognition of mixed gases is achieved. Several neural network algorithms for mixed gas detection in coal mines are analyzed and compared. The algorithms mainly include backpropagation (BP) neural network, convolutional neural network (CNN), recurrent neural network (RNN), and radial basis function (RBF) neural network. BP neural network can usually achieve high classification precision, but requires training a large number of parameters and a long training time. Usually, in order to reduce time and improve precision, BP neural networks can be combined with other algorithms. CNN can automatically extract data features, with better precision and training speed than BP neural networks. But it is prone to falling into local optima. RNN can use less data and extract more effective features, but it is prone to problems such as gradient vanishing. RBF neural networks have strong robustness and online learning capability, but they usually require a large amount of data to complete model training. The application of neural network algorithms will significantly improve the detection precision of mixed gases in coal mines, ensuring the implementation of intelligent coal mines.
- coal mine safety monitoring /
- mixed gas detection /
- neural network algorithm /
- sensor array /
- BP neural network /
- CNN /
- RNN /
- RBF neural network

HTML全文

图 1 煤矿混合气体检测系统结构^[3]

Figure 1. Structure of coal mine mixed gas detection system^[3]

下载: 全尺寸图片幻灯片

图 2 经典的单隐层BP神经网络^[4]

Figure 2. Classical BP neural network with single hidden layer^[4]

下载: 全尺寸图片幻灯片

图 3 1DCNN结构^[6]

Figure 3. One dimensional convolutional neural network structure^[6]

下载: 全尺寸图片幻灯片

图 4 1D−DCNN算法的总体架构^[10]

Figure 4. The overall architecture of 1D-DCNN algorithm^[10]

下载: 全尺寸图片幻灯片

图 5 RNN模型框架^[12]

Figure 5. Framework of RNN model^[12]

下载: 全尺寸图片幻灯片

图 6 CNN−RNN结构^[13]

Figure 6. Sturcture of CNN-RNN^[13]

下载: 全尺寸图片幻灯片

图 7 RBF神经网络结构^[19]

Figure 7. RBF neural network structure^[19]

下载: 全尺寸图片幻灯片

图 8 改进RBF神经网络流程^[23]

Figure 8. Process of improved RBF neural network^[23]

下载: 全尺寸图片幻灯片

表 1 1D−DCNN与其他方法的识别准确率比较^[10]

Table 1. Comparison of recognition accuracy between 1D-DCNN and other methods^[10]

方法	准确率/%
SVM	87.45
ANN	85.85
KNN	80.45
RF	88.69
1D−DCNN	96.30

下载: 导出CSV

表 2 4种算法对混合气体浓度的预测结果对比^[13]

Table 2. Comparison of four algorithms for predicting the concentration of mixed gases^[13]

算法	均方根误差/10⁻⁶
算法	一氧化碳	乙烯
CNN−LSTM	35.11	1.33
CNN−GRU	35.64	1.55
ESN	45.76	1.78
LR	70.10	1.85

下载: 导出CSV

表 3 BP神经网络和RBF神经网络预测误差对比^[22]

Table 3. Comparison of prediction errors between BP neural network and RBF neural network^[22]

预测样本	预测误差/%
预测样本	BP神经网络	RBF神经网络
21	2.33	0.36
22	1.67	0.46
23	1.05	0.27
24	1.24	0.40
平均值/%	1.58	0.37
迭代收敛步数	3 256	1 326

下载: 导出CSV

参考文献(23)

[1]	金智新,王宏伟,付翔. HCPS理论体系下新一代智能煤矿发展路径[J]. 工矿自动化,2022,48(10):1-12. JIN Zhixin,WANG Hongwei,FU Xiang. Development path of new generation intelligent coal mine under HCPS theory system[J]. Journal of Mine Automation,2022,48(10):1-12.
[2]	王国法,张良,李首滨,等. 煤矿无人化智能开采系统理论与技术研发进展[J]. 煤炭学报,2023,48(1):34-53. WANG Guofa,ZHANG Liang,LI Shoubin,et al. Progresses in theory and technological development of unmanned smart mining system[J]. Journal of China Coal Society,2023,48(1):34-53.
[3]	许刚. 基于GA-RBF的煤矿机器人井下混合气体检测系统的研究[J]. 计算技术与自动化,2018,37(3):66-68. XU Gang. Research on underground gas mixture detection system for coal mine robot based on GA-RBF[J]. Computing Technology and Automation,2018,37(3):66-68.
[4]	XU Xuebin,QIN Hu,ZHOU Jie. Cyber intrusion detection based on a mutative scale chaotic bat algorithm with backpropagation neural network[J]. Security and Communication Networks,2022. DOI: 10.1155/2022/5605404.
[5]	LECUN Y,BOSER B,DENKER J S,et al. Backpropagation applied to handwritten zip code recognition[J]. Neural Computation,1989,1(4):541-551. doi: 10.1162/neco.1989.1.4.541
[6]	李鹏,徐永凯,杨佳康,等. 基于一维卷积神经网络的气体识别方法研究[J]. 电子器件,2022,45(3):645-650. LI Peng,XU Yongkai,YANG Jiakang,et al. Study on gas recognition method based on one-dimensional convolutional neural network[J]. Chinese Journal of Electron Devices,2022,45(3):645-650.
[7]	PENG Pai,ZHAO Xiaojin,PAN Xiaofang,et al. Gas classification using deep convolutional neural networks[J]. Sensors 2018,18(1). DOI: 10.3390/s18010157.
[8]	谭光韬,张文文,王磊. 气体传感器阵列混合气体检测算法研究[J]. 电子测量与仪器学报,2020,34(7):95-102. TAN Guangtao,ZHANG Wenwen,WANG Lei. Research on mixed gas detection algorithm of gas sensor array[J]. Journal of Electronic Measurement and Instrumentation,2020,34(7):95-102.
[9]	SHARMA M,MAITY T. Multisensor data-fusion-based gas hazard prediction using DSET and 1DCNN for underground longwall coal mine[J]. IEEE Internet of Things Journal,2022,9(21):21064-21072. doi: 10.1109/JIOT.2022.3175724
[10]	ZHAO Xiaojin,WEN Zhihuang,PAN Xiaofang,et al. Mixture gases classification based on multi-label one-dimensional deep convolutional neural network[J]. IEEE Access,2019,7:12630-12637. doi: 10.1109/ACCESS.2019.2892754
[11]	LI Xiulei,GUO Jiayi,XU Wangping,et al. Optimization of the mixed gas detection method based on neural network algorithm[J]. ACS Sensors,2023,8(2):822-828. doi: 10.1021/acssensors.2c02450
[12]	罗敏,黄小美,吕山. 基于PCA−LSTM的城市燃气日负荷预测[C]. 中国燃气运营与安全研讨会(第十届)暨中国土木工程学会燃气分会2019年学术年会,上海,2019:120-132. LUO Min,HUANG Xiaomei,LYU Shan. Daily load forecasting of urban gas based on PCA-LSTM[C]. China Gas Operation and Safety Symposium (10th) and 2019 Academic Annual Meeting of the Gas Branch of the Chinese Civil Engineering Society,Shanghai,2019:120-132.
[13]	温志煌. 用于智能电子鼻系统的新型混合气体识别算法研究[D]. 深圳:深圳大学,2019. WEN Zhihuang. Research on the novel mixture gas recognition algorithms for smart electronic nose system[D]. Shenzhen:Shenzhen University,2019.
[14]	LYU Pingyang,CHEN Ning,MAO Shanjun,et al. LSTM based encoder-decoder for short-term predictions of gas concentration using multi-sensor fusion[J]. Process Safety and Environmental Protection,2020,137:93-105. doi: 10.1016/j.psep.2020.02.021
[15]	ZHANG Wenwen,WANG Lei,CHEN Jia,et al. A novel gas recognition and concentration detection algorithm for artificial olfaction[J]. IEEE Transactions on Instrumentation and Measurement,2021,70. DOI: 10.1109/TIM.2021.3071313.
[16]	BAKILER H,GUNEY S. Estimation of concentration values of different gases based on long short-term memory by using electronic nose[J]. Biomedical Signal Processing and Control,2021,69. DOI: 10.1016/j.bspc.2021.102908.
[17]	张海庆. 基于LSTM循环神经网络的矿用甲烷传感器自校准研究[J]. 煤矿机械,2022,43(6):168-171. ZHANG Haiqing. Research on self-calibration of mine methane sensor based on LSTM recurrent neural network[J]. Coal Mine Machinery,2022,43(6):168-171.
[18]	WANG Jianjun,XU Zongben. New study on neural networks:the essential order of approximation[J]. Neural Networks,2010,23(5):618-624. doi: 10.1016/j.neunet.2010.01.004
[19]	WANG Xi,ZHOU Yangming,ZHAO Zhikai,et al. Advanced algorithms for low dimensional metal oxides-based electronic nose application:a review[J]. Crystals,2023,13(4). DOI: 10.3390/cryst13040615.
[20]	YU Hao,XIE Tiantian,PASZCZYNSKI S,et al. Advantages of radial basis function networks for dynamic system design[J]. IEEE Transactions on Industrial Electronics,2011,58(12):5438-5450. doi: 10.1109/TIE.2011.2164773
[21]	赵金宪,于光华. 瓦斯浓度预测的混沌时序RBF神经网络模型[J]. 黑龙江科技学院学报,2010,20(2):131-134. ZHAO Jinxian,YU Guanghua. Model of chaotic sequence and RBF neural network on gas concentration forecast[J]. Journal of Heilongjiang Institute of Science and Technology,2010,20(2):131-134.
[22]	李万庆,裴志全,孟文清. AHP−RBF神经网络在煤矿安全风险评价中的应用[J]. 河北工程大学学报(自然科学版),2014,31(2):101-105. LI Wanqing,PEI Zhiquan,MENG Wenqing. The application of AHP-RBF neural network in coal mine safety risk evaluation[J]. Journal of Hebei University of Engineering(Natural Science Edition),2014,31(2):101-105.
[23]	西安科技大学. 煤矿井下多气体浓度采集传输装置:2014206299489[P]. 2014-10-28. Xi'an University of Science and Technology. Multi gas concentration collection and transmission device in coal mine underground:2014206299489[P]. 2014-10-28.

施引文献

资源附件(0)

访问统计

点击查看大图

图(8) / 表(3)

计量

文章访问数: 998
HTML全文浏览量: 55
PDF下载量: 60
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

留言板

面向煤矿混合气体检测的神经网络算法研究进展

doi: 10.13272/j.issn.1671-251x.18105

作者简介:
焦明之（1987—），男，安徽安庆人，副研究员，研究方向为智能传感器及煤矿安全监测监控技术，E-mail：mingzhijiao@cumt.edu.cn

通讯作者:
贺耀宜（1974—），男，陕西蓝田人，研究员，硕士，主要从事煤矿监测监控、物联网与信息化应用研究工作，E-mail： hyy@cari.com.cn。

计量

Research progress on neural network algorithms for mixed gas detection in coal mines

计量

目录

留言板

面向煤矿混合气体检测的神经网络算法研究进展

doi: 10.13272/j.issn.1671-251x.18105

作者简介: 焦明之（1987—），男，安徽安庆人，副研究员，研究方向为智能传感器及煤矿安全监测监控技术，E-mail：mingzhijiao@cumt.edu.cn

通讯作者: 贺耀宜（1974—），男，陕西蓝田人，研究员，硕士，主要从事煤矿监测监控、物联网与信息化应用研究工作，E-mail： hyy@cari.com.cn。

计量

出版历程

Research progress on neural network algorithms for mixed gas detection in coal mines

计量

出版历程

目录

作者简介:
焦明之（1987—），男，安徽安庆人，副研究员，研究方向为智能传感器及煤矿安全监测监控技术，E-mail：mingzhijiao@cumt.edu.cn

通讯作者:
贺耀宜（1974—），男，陕西蓝田人，研究员，硕士，主要从事煤矿监测监控、物联网与信息化应用研究工作，E-mail： hyy@cari.com.cn。