|本期目录/Table of Contents|

[1]由东冉,刘剑,曹鹏,等.基于深度学习的实时热压湿变通风阻力系数时序预测研究*[J].中国安全生产科学技术,2025,21(11):150-158.[doi:10.11731/j.issn.1673-193x.2025.11.018]
　YOU Dongran,LIU Jian,CAO Peng,et al.Research on time series prediction of ventilation resistance coefficient under real-time thermal-pressure-humidity variation based on deep learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(11):150-158.[doi:10.11731/j.issn.1673-193x.2025.11.018]
点击复制

基于深度学习的实时热压湿变通风阻力系数时序预测研究*
分享到：

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
21
期数:
2025年11期
页码:
150-158
栏目:
职业安全卫生管理与技术
出版日期:
2025-11-30

文章信息/Info

Title:
Research on time series prediction of ventilation resistance coefficient under real-time thermal-pressure-humidity variation based on deep learning
文章编号:
1673-193X(2025)-11-0150-09
作者:
由东冉; 刘剑; 曹鹏; 王东; 刘丽; 杨成虎; 张鑫
（1.辽宁工程技术大学 安全科学与工程学院，辽宁 葫芦岛 125105；
2.辽宁工程技术大学 矿山热动力灾害与防治教育部重点实验室，辽宁 葫芦岛 125105；
3.山东工商学院 管理科学与工程学院，山东 烟台 264005；
4.辽宁工程技术大学 辽宁省高等学校矿产资源开发利用技术及装备研究院，辽宁 阜新 123000；
5.山西北方铜业有限公司铜矿峪矿，山西 运城 043700）
Author(s):
YOU Dongran;  LIU Jian;  CAO Peng;  WANG Dong;  LIU Li;  YANG Chenghu;  ZHANG Xin
关键词:
矿井通风; 热压湿变通风阻力系数; 深度学习; 时间序列; 时间卷积网络（TCN）
Keywords:
mine ventilation;  ventilation resistance coefficient under thermal-pressure-humidity variation;  deep learning;  time series;  temporal convolutional networks (TCN)
分类号:
X936
DOI:
10.11731/j.issn.1673-193x.2025.11.018
文献标志码:
A
摘要:
为精准预测智能通风系统中实时热压湿变通风阻力系数，克服传统机器学习方法依赖静态物理参数、忽略热力过程动态影响而无法满足实时矿井通风网络解算需求的问题，提出1种基于深度学习时间卷积网络（TCN）的实时预测方法。通过分析热压湿变通风阻力系数的时序特性，构建包含膨胀因果卷积与残差连接的TCN模型，以历史风流温度、湿度和大气压等时序特征作为输入进行实时预测。采用铜矿峪矿井实测数据进行模型训练与验证，并与6种传统算法对比。研究结果表明：深度学习时序方法预测的平均绝对百分比误差仅为1.98%，满足智能通风实际需求。7种算法的性能排序为：TCN>LSTM>GRU>Linear>Transformer>CNN>SVR，其中TCN算法表现最优，MAE为0.012，MSE为0.000 42，MAPE为1.98%，R2为0.998 1。研究结果可为实时矿井通风网络解算提供参考。
Abstract:
In order to accurately predict the real-time ventilation resistance coefficient under thermal-pressure-humidity variation in intelligent ventilation systems,and to overcome the limitations of traditional machine learning methods that rely on static physical parameters while ignoring the dynamic effects of thermal processes—thus failing to meet the requirements for real-time mine ventilation network solution—a novel real-time prediction method based on the deep learning temporal convolutional network (TCN) is proposed.By analyzing the time-series characteristics of the ventilation resistance coefficient under thermal-pressure-humidity variation,a TCN model incorporating dilated causal convolutions and residual connections is constructed.This model uses historical time-series features such as airflow temperature,humidity,and atmospheric pressure as inputs for real-time prediction.The model was trained and validated using measured data from the Tongkuangyu Mine,and compared with six traditional algorithms.The results demonstrate that the deep learning time-series method achieves a mean absolute percentage error (MAPE) of only 1.98%,meeting the practical demands of intelligent ventilation.The performance ranking of the seven algorithms is:TCN > LSTM > GRU > Linear > Transformer > CNN > SVR,with TCN exhibiting the best performance (MAE=0.012,MSE=0.000 42,MAPE=1.98%,R2=0.998 1).The findings provide a reference for real-time solution of mine-ventilation networks.

参考文献/References:

［1］刘剑.矿井智能通风关键科学技术问题综述［J］.煤矿安全,2020,51(10):108-111，117. LIU Jian.Overview on key scientific and technical issues of mine intelligent ventilation［J］.Safety in Coal Mines,2020,51(10):108-111，117.图8各模型预测性能对比结果
［2］冯燕,刘剑.基于BP神经网络的矿车运行时矿井摩擦阻力的预测［J］.中国安全生产科学技术,2023,19(1):54-59. FENG Yan，LIU Jian.Prediction of mine friction resistance during the operation of mining vehicles based on BP neural network ［J］.Journal of Safety Science and Technology,2023,19(1):54-59.
［3］张攀,周志强,刘剑.利用BP神经网络确定巷道摩擦阻力系数［J］.辽宁工程技术大学学报,2003(4):536-537. ZHANG Pan,ZHOU Zhiqiang,LIU Jian.Determining the friction resistance coefficient of roadway using BP neural network ［J］.Journal of Liaoning Technical University,2003(4):536-537.
［4］魏宁,孙亚胜男,邓立军,等.基于SVM的矿井通风阻力系数影响因素分析与预测［J］.中国安全生产科学技术,2018,14(4):39-44. WEI Ning,SUN Yashengnan,DENG Lijun,et al.Analysis and prediction of factors affecting mine ventilation resistance coefficient based on SVM ［J］.Journal of Safety Science and Technology,2018,14(4):39-44.
［5］GAO K,DENG L J,LIU J,et al.Study on mine ventilation resistance coefficient inversion based on genetic algorithm［J］.Archives of Mining Sciences,2018,63(4):813-826.
［6］戚志鹏,高科,刘玉姣,等.巷道通风摩擦阻力系数遗传投影寻踪回归预测［J］.安全与环境学报,2024,24(7):2648-2657. QI Zhipeng,GAO Ke,LIU Yujiao,et al.Genetic projection pursuit regression prediction of roadway ventilation friction resistance coefficient ［J］.Journal of Safety and Environment,2024,24(7):2648-2657.
［7］王思仪.基于多层前向神经网络的矿井巷道摩擦阻力系数的确定［D］.葫芦岛:辽宁工程技术大学,2014.
［8］陈可可.基于XGBoost矿井摩擦阻力系数预测及应用［D］.葫芦岛:辽宁工程技术大学,2023.
［9］陈开岩,周福宝,夏同强,等.基于空气状态参数与风量耦合迭代的风网解算方法［J］.中国矿业大学学报,2021,50(4):613-623. CHEN Kaiyan,ZHOU Fubao,XIA Tongqiang,et al.Wind network solution method based on air state parameter and air volume coupling iteration ［J］.Journal of China University of Mining & Technology,2021,50(4):613-623.
［10］赖添城,徐康康,朱成就,等.一种基于改进CNN-GRU的建筑冷负荷单步预测方法［J］.机电工程技术,2024,53(1):119-122. LAI Tiancheng,XU Kangkang,ZHU Chengjiu,et al.A one-step prediction method for building cooling load based on improved CNN-GRU ［J］.Mechanical and Electrical Engineering Technology,2024,53(1):119-122.
［11］戴璐平,沈嘉怡,张飞飞.基于时间序列算法的能源电力需求自动预测模型［J］.自动化技术与应用,2024,43(1):49-51,65. DAI Luping,SHEN Jiayi,ZHANG Feifei.An automatic prediction model for energy and power demand based on time series algorithm ［J］.Techniques of Automation and Applications,2024,43(1):49-51,65.
［12］王瑞芳,马平安,俞婷.时间序列预测模型发展趋势分析［J］.科技资讯,2024,22(19):73-78. WANG Ruifang,MA Ping’an,YU Ting.Analysis of the development trend of time series prediction models ［J］.Science & Technology Information,2024,22(19):73-78.
［13］黄元平.矿井通风［M］.徐州:中国矿业大学出版社,1986.
［14］SPEARMAN C.“General Intelligence” objectively deter-mined and measured［J］.American Journal of Psychology,1904,15(2):201-293.
［15］HAN C,GONG M,SUN J,et al.Heat load prediction for districtheating systems with temporal convolutional network and Cat-Boost［J］.Thermal Engineering,2023,70(9):719-726.

相似文献/References:

[1]陈明生,陈伯辉,沈斐敏.矿井通风优化评价指标体系权重确定[J].中国安全生产科学技术,2011,7(3):22.
　CHEN Ming-sheng,CHEN Bo-hui,SHEN Fei-min.Weight determination of evaluation index system in mine ventilation optimization[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2011,7(11):22.
[2]张黔生,谢贤平,吴劭星,等.多风机多级机站通风网络的状态估计[J].中国安全生产科学技术,2010,6(3):75.
　ZHANG Qian-sheng,XIE Xian-ping,WU Shao-xing,et al.Static state estimation of ventilation system for multistage fan station[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2010,6(11):75.
[3]黄光球,孙鹏,陆秋琴.风窗对井下通风系统的影响及其调节与定位优化[J].中国安全生产科学技术,2014,10(3):160.[doi:10.11731/j.issn.1673-193x.2014.03.028]
　HUANG Guang qiu,SUN Peng,LU QIU qin.Influence of air windows on underground ventilation system and its adjusting and locating optimization[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2014,10(11):160.[doi:10.11731/j.issn.1673-193x.2014.03.028]
[4]张黔生,谢贤平,吴劭星.多风机多级机站通风系统优化的模糊群体决策法[J].中国安全生产科学技术,2009,5(5):88.
　ZHANG Qian sheng,XIE Xian ping,WU Shao xing.Study on optimization of mine ventilation system of multifan and multistage based on fuzzy population decision[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2009,5(11):88.
[5]刘剑,宋莹,刘明浩,等.基于LDA的均直巷道断面风速分布规律实验研究[J].中国安全生产科学技术,2015,11(12):65.[doi:10.11731/j.issn.1673-193x.2015.12.010]
　LIU Jian,SONG Ying,LIU Ming-hao,et al.Experimental study on distribution laws of wind velocity in straight roadway section based on LDA[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2015,11(11):65.[doi:10.11731/j.issn.1673-193x.2015.12.010]
[6]马恒,高巍,周西华.矿井气候多参数预测与通风网络自动解算算法研究[J].中国安全生产科学技术,2017,13(11):110.[doi:10.11731/j.issn.1673-193x.2017.11.018]
　MA Heng,GAO Wei,ZHOU Xihua.Study on multi-parameter prediction of mine climate and automatic solution algorithm of ventilation network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2017,13(11):110.[doi:10.11731/j.issn.1673-193x.2017.11.018]
[7]杨帅,撒占友,王相君,等.基于环境监测的矿井通风三维仿真辅助决策系统设计[J].中国安全生产科学技术,2020,16(1):80.[doi:10.11731/j.issn.1673-193x.2020.01.013]
　YANG Shuai,SA Zhanyou,WANG Xiangjun,et al.Design of 3D simulation assistant decisionmaking system for mine ventilation based on environmental monitoring[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2020,16(11):80.[doi:10.11731/j.issn.1673-193x.2020.01.013]
[8]魏宁,孙亚胜男,邓立军,等.基于SVM的矿井通风阻力系数影响因素分析与预测[J].中国安全生产科学技术,2018,14(4):39.[doi:10.11731/j.issn.1673-193x.2018.04.006]
　WEI Ning,SUN Yashengnan,DENG Lijun,et al.Influence factors analysis and prediction on mine ventilation resistance coefficient based on SVM[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(11):39.[doi:10.11731/j.issn.1673-193x.2018.04.006]
[9]赵自豪,李鹏慧,吕海建.扰动法在风网解算中的应用研究*[J].中国安全生产科学技术,2021,17(5):79.[doi:10.11731/j.issn.1673-193x.2021.05.012]
　ZHAO Zihao,LI Penghui,LYU Haijian.Study on application of perturbation method in resolution of ventilation network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(11):79.[doi:10.11731/j.issn.1673-193x.2021.05.012]
[10]张逸斌.基于机器学习的矿井通风故障智能诊断机制研究*[J].中国安全生产科学技术,2025,21(11):159.[doi:10.11731/j.issn.1673-193x.2025.11.019]
　ZHANG Yibin.Research on intelligent diagnosis mechanisms of mine ventilation faults based on machine learning[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2025,21(11):159.[doi:10.11731/j.issn.1673-193x.2025.11.019]

备注/Memo

备注/Memo:
收稿日期： 2025-07-15
* 基金项目： 辽宁省科学技术计划项目（2025-BS-0399）；国家自然科学基金项目（52404212）；辽宁省教育厅科研项目（LJ212410147072）
作者简介： 由东冉，硕士研究生，主要研究方向为矿井通风。
通信作者： 刘剑，博士，教授，主要研究方向为矿井通风及灾害防治。
更新日期/Last Update: 2025-12-03