img

QQ群聊

img

官方微信

高级检索

黄金科学技术 ›› 2021, Vol. 29 ›› Issue (5): 709-718.doi: 10.11872/j.issn.1005-2518.2021.05.054

• 采选技术与矿山管理 • 上一篇    下一篇

基于GRU神经网络的巷道平均风速获取研究

邵良杉1,2(),闻爽爽1,2()   

  1. 1.辽宁工程技术大学管理科学与工程研究院,辽宁 葫芦岛 125105
    2.辽宁工程技术大学工商管理学院,辽宁 葫芦岛 125105
  • 收稿日期:2021-05-09 修回日期:2021-08-27 出版日期:2021-10-31 发布日期:2021-12-17
  • 通讯作者: 闻爽爽 E-mail:lntushao@163.com;18340312141@163.com
  • 作者简介:邵良杉(1961-),男,辽宁凌源人,教授,博士生导师,从事矿业系统工程和通风系统优化研究工作。lntushao@163.com
  • 基金资助:
    国家自然科学基金项目“基于大数据的煤与瓦斯突出的预测方法与应用研究”(71771111)

Research on Obtaining Average Wind Speed of Roadway Based on GRU Neural Network

Liangshan SHAO1,2(),Shuangshuang WEN1,2()   

  1. 1.Institute of Management Science and Engineering,Liaoning Technical University,Huludao 125105,Liaoning,China
    2.School of Business Administration,Liaoning Technical University,Huludao 125105,Liaoning,China
  • Received:2021-05-09 Revised:2021-08-27 Online:2021-10-31 Published:2021-12-17
  • Contact: Shuangshuang WEN E-mail:lntushao@163.com;18340312141@163.com

摘要:

针对矿井智能通风系统不能及时获取风速进而影响后续通风系统解算及优化的问题,利用ANSYS巷道风速分布模拟获取神经网络所需训练集,在人工测量与风速传感器监测数据的基础上,构建基于门控循环单元(Gated Recurrent Unit)神经网络的巷道平均风速预测模型。首先,提出神经网络模型,然后采用Adam优化算法对ANSYS模拟的点风速进行异常值和归一化等预处理,通过对不同形状巷道的监测点风速进行结构化处理后用于训练神经网络,找出各点风速与平均风速之间的强非线性关系,使预测风速逼近巷道实际平均风速,最后构建基于GRU神经网络的巷道平均风速预测模型。以王家岭煤矿实测数据作为测试集,将其应用于预测模型中,结果表明GRU神经网络模型具有较高精度和较强的泛化能力,能够获取巷道平均风速。矿井通风巷道平均风速预测模型在煤矿领域的成功应用,将为其他金属矿山智能通风系统及时准确获取风速参数提供新思路。

关键词: 智慧矿山, GRU神经网络, 监测监控系统, 平均风速, 智能通风系统

Abstract:

With the advent of the intelligent era, computer simulation algorithms such as machine learning and deep learning have played a role in many fields such as aerospace, medical treatment, education and communication. For the traditional industry of mining, the concept of smart mine has become a research hotspot of relevant researchers in recent years. Intelligent technologies such as machine learning have been used in pedestrian detection, gas prediction, coal rock identification has been successfully applied to practical production, but the intelligent acquisition of parameters in intelligent ventilation system is still in a blank. Therefore, under the background of smart mine, aiming at the problem that the mine intelligent ventilation system can’t obtain the wind speed in time and then complete the subsequent ventilation system solution and optimization, the training set required by the neural network is obtained by using the simulation of tunnel wind speed distribution in ANSYS. Based on manual measurement and wind speed sensor monitoring data, the prediction model of roadway average wind speed based on gated recurrent unit neural network was constructed. Firstly, the neural network model was proposed, and then the Adam optimization algorithm was used to preprocess the data such as outlier processing and normalization. After the structural processing of the wind speed at the monitoring points of the roadway with different shapes, it was used to train the neural network to find out the strong nonlinear relationship between the wind speed at each point and the average wind speed, so that the predicted wind speed is close to the actual average wind speed of the roadway. Finally, the prediction model of roadway average wind speed based on GRU neural network was constructed. Taking the measured data of Wangjialing coal mine as the model test set, the results show that the GRU neural network model has high precision and strong generalization ability, and can obtain the average wind speed of roadway, which will provide a roadway average wind speed prediction model with advanced technology, scientific process and accurate results for the mine intelligent ventilation system. Moreover, the strong prediction ability of in-depth learning will provide intelligent data for the solution and optimization of ventilation network, it can be extended to the acquisition of ventilation parameters in other metal mines to popularize the intelligent acquisition of ventilation parameters.

Key words: smart mine, GRU neural network, monitoring and monitoring system, average wind speed, intelligent ventilation system

中图分类号: 

  • TD76

图1

LSTM隐藏层结构示意图"

图2

GRU隐藏层结构示意图"

图3

GRU神经网络训练步骤"

图4

基于GRU神经网络的平均风速精确获取流程"

图5

不同巷道三维结构示意图"

图6

风速场分布图"

图7

平均风速曲线"

表1

误差数据"

算法遍历次数训练集损失均方误差平均绝对误差测试集损失均方误差损失平均绝对误差损失
13.66E-023.66E-021.31E-011.80E-021.80E-021.20E-01
22.50E-022.50E-021.22E-011.40E-021.40E-021.07E-01
31.91E-021.91E-021.07E-019.96E-039.96E-038.98E-02
41.41E-021.41E-029.02E-026.80E-036.80E-037.47E-02
51.01E-021.01E-027.42E-024.63E-034.63E-036.30E-02
66.99E-036.99E-036.10E-022.52E-032.52E-034.54E-02
74.78E-034.78E-034.80E-021.40E-031.40E-033.36E-02
83.27E-033.27E-033.78E-027.84E-047.84E-042.51E-02
92.30E-032.30E-033.06E-023.84E-043.84E-041.68E-02
101.70E-031.70E-032.49E-021.91E-041.91E-049.91E-03
111.33E-031.33E-032.05E-021.32E-041.32E-047.53E-03
121.12E-031.12E-031.81E-021.32E-041.32E-048.89E-03
139.92E-049.92E-041.73E-021.54E-041.54E-041.03E-02
149.19E-049.19E-041.70E-021.77E-041.77E-041.07E-02
158.71E-048.71E-041.68E-021.99E-041.99E-041.12E-02
168.42E-048.42E-041.69E-022.23E-042.23E-041.25E-02
178.21E-048.21E-041.67E-022.32E-042.32E-041.27E-02
188.03E-048.03E-041.68E-022.50E-042.50E-041.37E-02
197.90E-047.90E-041.67E-022.55E-042.55E-041.40E-02
207.78E-047.78E-041.66E-022.53E-042.53E-041.39E-02

图8

训练样本与测试样本随Epochs的损失、MSE和MAE变化曲线"

表2

拱形巷道断面九点法实测风速"

断面

断面尺寸

(高度×宽度)/m

V1V2V3V4V5V6V7V8V9实测平均风速 /(m·s-1
断面13.84×3.840.75250.77720.81420.74020.77100.81420.75250.76480.72780.7683
断面24.55×3.871.37061.37701.42041.37061.55711.12821.36441.47011.12201.3534
断面34.63×3.711.83171.66521.50481.78851.74531.52951.88101.66521.36921.6645
断面42.54×3.830.57370.61070.49350.57370.61070.52430.52430.58600.57370.5634
断面55.52×4.040.41040.33980.33390.38100.41630.42220.41040.34570.33390.3771
断面64.51×5.474.04424.46024.37703.61164.44354.42693.24563.99433.99434.0664
断面73.02×3.8110.621211.40109.21759.841311.089111.08918.531210.93318.749610.1637

表3

矩形巷道断面九点法实测风速"

断面

断面尺寸

(高度×宽度)/m

V1V2V3V4V5V6V7V8V9实测平均风速 /(m·s-1
断面14.98×3.472.16202.01802.16202.27722.19082.26282.26282.24842.26282.2052
断面24.79×4.140.45030.40100.33320.43180.38870.33320.43800.36400.35170.3880
断面33.41×4.900.43180.40100.36400.45030.46270.40100.35780.48120.40100.4168
断面43.17×4.430.70950.64720.64720.77180.71570.72820.74060.74060.70950.7123
断面52.54×3.331.32331.15571.13251.41261.42381.33451.34561.42381.44621.3331
断面62.74×4.586.09815.78625.91106.59726.56606.53486.44136.53486.25416.3026

表4

巷道断面实测风速与预测风速对比"

断面类型断面实测风速/(m·s-1预测风速/(m·s-1准确率/%断面类型断面实测风速/(m·s-1预测风速/(m·s-1准确率/%
拱形锚喷断面10.760.590.78矩形锚喷断面12.202.201
断面21.351.230.91断面20.380.340.9
断面31.661.620.98断面30.410.411
断面40.590.330.56断面40.710.490.69
断面50.370.270.73断面51.331.280.96
断面64.064.421.08断面66.306.681.06
断面710.1610.291.01

图9

巷道平均风速预测结果"

Cao Long,Yang Min,2021.Key technology of intelligent ventilation in metal mines [J]. World Nonferrous Metals, (12):100-101.
Chen Long,Wang Xiao,Yang Jianjian,al et,2021. Parallel mining operating systems: from digital twins to mining intelligence [J]. Acta Automatica Sinica,47(7):1633-1645.
Cheng Li,Liu Huanxin,Zhu Mingde, al et,2020.Current situation and prospect of research on underground goaf in metal mines [J]. Gold Science and Technology, 28(1):70-81.
Cho K,Van Merriënboer B,Bahdanau D,al et,2014.On the properties of neural machine translation:Encoder-decoder approaches[J].arXiv:1409.1259. DOI:10.3115/v1/W14-4012.
doi: 10.3115/v1/W14-4012
Ding Cui,He Xueqiu,Nie Baisheng,2015.Numerical and experimental research on the “Key Ring” of airflow distribution in mine ventilation tunnels[J].Journal of Liaoning Technical University (Natural Science Edition),34(10):1131-1136.
Hao Yuanwei,Chen Kaiyan,Jiang Zhongcheng,al et,2011.Correction processing of roadway wind speed monitoring value based on CFD simulation [J].Coal Mine Safety, 42(2):1-3,7.
Hochreiter S,Schmidhuber J,1997.Long short-term memory[J].Neural computation,9(8):1735-1780.
Hu Jianhua, Hang Long, Wang Xueliang, al et,2018.Platform structure for intelligent underground mining system its construction [J]. Mining and Metallurgical Engineering ,38(6):1-5.
Kingma D P,Ba J,2014.Adam:A method for stochastic optimization[J].arXiv:1412.6980 v8.
Li Man,Ma Huan,2017.Simulation study on air volume test method of mine main fan[J].Coal Science and Technology,45(2):151-155.
Li Xibing,Cao Zhiwei,Zhou Jian, al et,2019.Innovation of mining models and construction of intelligent green mine in hard rock mine :In Kaiyang phosphate mine as an example [J]. The Chinese Journal of Nonferrous Metals, 29(10):2364-2380.
Li Xuebing,Liu Jian,Song Ying,al et,2018.Conversion mechanism between single-point wind speed and average wind speed in the section of shaft[J].Journal of Safety and Environment, 18(1):123-128.
Liu Xiaoming,Deng Lei, Wang Liguan, al et,2020.Intelligent mine master plan for underground metal mine[J]. Gold Science and Technology, 28(2):309-316.
Lu Xinming,Yin Hong,2020.Theory and technology of mine ventilation intelligentization[J].Journal of China Coal Society,2020,45(6):2236-2247.
Luo Guang,2020.Research on the Distribution Law of Wind Speed and Precise Monitoring of Air Volume in Typical Roadway Section[D].Beijing:Coal Science Research Institute.
Shao Liangshan,Yu Baocai,Chen Xiaoyong,2020.Key technology of mine intelligent ventilation[J].Coal Mine Safety,51(11):121-124.
Song Ying,Liu Jian,Li Xuebing,al et, 2016.Experimental and simulation study on the distribution law of average wind speed of mine tunnel[J].Chinese Safety Science Journal, 26(6):146-151.
Wang En,Zhang Lang,Li Wei,al et,2016.Multi-point mobile wind measurement device and key technology[J].Coal Mi-ne Safety,47(6):97-99,103.
Wang Hanfeng,2015.Simulation research on location monitoring based on average wind speed of Fluent roadway section[J].Coal Science and Technology, 43(8):92-96.
Wang Jun,Chen Kaiyan,Huang Shuai,2013.Single-point measurement of average wind speed in mine tunnel based on CFD numerical simulation[J].Coal Mine Safety,44(3):144-146.
Wu Xinzhong,Zhang Zhichao,Xu Jialin,al et,2021.Research on intelligent air volume regulation in mines [J].Industry and Mine Automation,47(4):44-50.
Yan Hang,Chen Gang,Tong Yao,al et,2021.Human rehabilitation motion recognition based on pose estimation and GRU network [J].Computer Engineering, 47(1):12-20.
Zhang Lang,2018.Optimization of the measurement position of the average wind speed of the wind speed sensor in the wind. measuring station in the roadway[J].Coal Science and Technology, 46(3):96-102.
Zhang Zhichang,Zhou Dong,Zhang Ruifang,al et,2020.Medical entity relationship recognition based on bidirectional GRU and attentional mechanism[J].Computer Engineering, 46(6):296-302.
Zhao Dan,Huang Fujun,Chen Shuai,al et, 2014.The relationship between point wind speed and average wind speed in a circular tunnel[J].Journal of Liaoning Technical University (Natural Science Edition),33(12):1654-1659.
Zhao Dan,Pan Jingtao,Liu Jian,2011.The average wind speed in the roadway is derived from the measured value of the wind speed sensor based on one-variable linear regression [J].World Science and Technology Research and Development, 33(2):229-231,241.
Zhao Wei,Li Wei,Huang Shuwei, al et,2018.Construction and practice of intelligent green mine for Sanshandao gold mine [J]. Gold Science and Technology, 26(2):219-227.
Zhou Xihua,Meng Le,Li Chengyu,al et,2012.Experimental research on wind speed. measurement and correction method for circular duct[J].Journal of Liaoning Technical University (Natural Science Edition),31(5):594.
曹龙,杨敏,2021.金属矿山矿井智能通风关键技术[J].世界有色金属, (12):100-101.
陈龙,王晓,杨健健,等,2021.平行矿山:从数字孪生到矿山智能[J].自动化学报 ,47(7):1633-1645.
程力,刘焕新,朱明德,等,2020.金属矿山地下采空区问题研究现状与展望[J].黄金科学技术, 28(1):70-81.
丁翠,何学秋,聂百胜,2015.矿井通风巷道风流分布“关键环”数值与实验研究[J].辽宁工程技术大学学报(自然科学版),34(10):1131-1136.
郝元伟,陈开岩,蒋中承,等,2011.基于CFD模拟的巷道风速监测值修正处理[J].煤矿安全, 42(2):1-3,7.
胡建华,张龙,王学梁,等,2018.井下矿山智能采矿体系的平台架构研究与实现[J].矿冶工程 ,38(6):1-5.
李曼,马欢,2017.矿井主通风机风量测试方法的模拟研究[J].煤炭科学技术,45(2):151-155.
李夕兵,曹芝维,周健,等,2019.硬岩矿山开采方式变革与智能化绿色矿山构建——以开阳磷矿为例[J].中国有色金属学报, 29(10):2364-2380.
李雪冰,刘剑,宋莹,等,2018.井巷断面内单点风速与平均风速转换机制[J].安全与环境学报,18(1):123-128.
刘晓明,邓磊,王李管,等,2020.地下金属矿智能矿山总体规划[J].黄金科学技术,28(2):309-316.
卢新明,尹红,2020.矿井通风智能化理论与技术[J].煤炭学报,45(6):2236-2247.
罗广,2020.典型巷道断面风速分布规律及风量精准监测研究[D].北京:煤炭科学研究总院.
邵良杉,于保才,陈晓永,2020.矿井智能通风关键技术[J].煤矿安全,51(11):121-124.
宋莹,刘剑,李雪冰,等,2016.矿井巷道风流平均风速分布规律的试验与模拟研究[J].中国安全科学学报, 26(6):146-151.
王恩,张浪,李伟,等,2016.多点移动式测风装置及关键技术[J].煤矿安全, 47(6):97-99,103.
王翰锋,2015.基于Fluent巷道断面平均风速点定位监测模拟研究[J].煤炭科学技术,43(8):92-96.
王军,陈开岩,黄帅,2013.基于CFD数值模拟的矿井巷道平均风速单点测法[J].煤矿安全,44(3):144-146.
吴新忠,张芝超,许嘉琳,等,2021.矿井智能风量调节研究[J].工矿自动化,47(4):44-50.
闫航,陈刚,佟瑶,等,2021.基于姿态估计与GRU网络的人体康复动作识别[J].计算机工程,47(1):12-20.
张浪,2018.巷道测风站风速传感器平均风速测定位置优化研究[J].煤炭科学技术,46(3):96-102.
张志昌,周侗,张瑞芳,等,2020.融合双向GRU与注意力机制的医疗实体关系识别[J].计算机工程,46(6):296-302.
赵丹,黄福军,陈帅,等,2014.点风速与平均风速在圆形巷道中的关系[J].辽宁工程技术大学学报(自然科学版),33(12):1654-1659.
赵丹,潘竞涛,刘剑,2011.基于一元线性回归由风速传感器测量值推导巷道平均风速[J].世界科技研究与发展,33(2):229-231,241.
赵威,李威,黄树巍,等,2018.三山岛金矿智能绿色矿山建设实践[J].黄金科学技术, 26(2):219-227.
周西华,孟乐,李诚玉,等,2012.圆形管道风速测定与校正方法实验研究[J].辽宁工程技术大学学报(自然科学版),31(5):594.
[1] 毕林,周超,姚鑫. 基于视频序列的矿卡司机不安全行为识别[J]. 黄金科学技术, 2021, 29(1): 14-24.
[2] 王牧帆,罗周全,于琦. 基于 Stacking 模型的采空区稳定性预测[J]. 黄金科学技术, 2020, 28(6): 894-901.
[3] 廖智勤, 王李管, 何正祥. 基于EEMD和关联维数的矿山微震信号特征提取和分类[J]. 黄金科学技术, 2020, 28(4): 585-594.
[4] 党明智,张君,贾明涛. 黄土坡铜锌矿微震监测技术应用与灾害预警方法研究[J]. 黄金科学技术, 2020, 28(2): 246-254.
[5] 随晓丹,罗周全,秦亚光,王玉乐,彭东. 基于小波分解的尾矿坝浸润线预测方法研究[J]. 黄金科学技术, 2019, 27(1): 137-143.
[6] 张二洋,陈建宏. 基于Surpac矿山设计软件及虚幻引擎实现的矿山虚拟现实漫游系统[J]. 黄金科学技术, 2017, 25(4): 93-98.
[7] 刘晓明,赵君杰,彭平安,毕林,代碧波. 有效微震信号自动识别技术研究[J]. 黄金科学技术, 2017, 25(3): 84-91.
[8] 聂兴信,张国丹. 基于熵值法—突变理论的地下矿山紧急避险系统可靠性研究[J]. 黄金科学技术, 2016, 24(6): 72-77.
[9] 王婷玉,罗周全,秦亚光,孙杨. 主溜井垮塌三维探测及可视化分析与计算[J]. 黄金科学技术, 2016, 24(1): 97-101.
[10] 尹土兵,王品,张鸣鲁. 基于AHP及模糊综合评判的地下金属矿山安全分析与评价[J]. 黄金科学技术, 2015, 23(3): 60-66.
[11] 秦亚光,罗周全,周吉明,汪伟,孙杨. 采空区可视化集成系统信息管理研究与应用[J]. 黄金科学技术, 2015, 23(2): 57-62.
[12] 赵永未,王京海,吕春堂,杨力纲,朱会凯. Dini03数字水准仪在望儿山采空区沉降监测及地质灾害预测分析中的应用[J]. 黄金科学技术, 2013, 21(6): 68-72.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!