留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于PSO-BP神经网络的导水裂隙带高度预测

娄高中 谭毅

娄高中, 谭毅. 基于PSO-BP神经网络的导水裂隙带高度预测[J]. 煤田地质与勘探, 2021, 49(4): 198-204. doi: 10.3969/j.issn.1001-1986.2021.04.024
引用本文: 娄高中, 谭毅. 基于PSO-BP神经网络的导水裂隙带高度预测[J]. 煤田地质与勘探, 2021, 49(4): 198-204. doi: 10.3969/j.issn.1001-1986.2021.04.024
LOU Gaozhong, TAN Yi. Prediction of the height of water flowing fractured zone based on PSO-BP neural network[J]. COAL GEOLOGY & EXPLORATION, 2021, 49(4): 198-204. doi: 10.3969/j.issn.1001-1986.2021.04.024
Citation: LOU Gaozhong, TAN Yi. Prediction of the height of water flowing fractured zone based on PSO-BP neural network[J]. COAL GEOLOGY & EXPLORATION, 2021, 49(4): 198-204. doi: 10.3969/j.issn.1001-1986.2021.04.024

基于PSO-BP神经网络的导水裂隙带高度预测

doi: 10.3969/j.issn.1001-1986.2021.04.024
基金项目: 

国家自然科学基金项目 51774111

河南省科技攻关项目 212102310406

安阳工学院博士科研基金项目 BSJ2019028

详细信息
    第一作者:

    娄高中,1988年生,男,河南平顶山人,博士,讲师,从事“三下”采煤研究. E-mail:754937725@qq.com

  • 中图分类号: TD823.83

Prediction of the height of water flowing fractured zone based on PSO-BP neural network

  • 摘要: 导水裂隙带高度是西部矿区保水采煤的理论依据和关键参数。近年来,BP神经网络广泛应用于导水裂隙带高度预测,但BP神经网络存在收敛速度慢、易陷入局部极小等问题。为提高导水裂隙带高度预测的准确性,利用粒子群优化算法(PSO)对BP神经网络的权值和阈值进行优化,建立基于PSO-BP神经网络的导水裂隙带高度预测模型。选择开采厚度、开采深度、工作面倾斜长度、煤层倾角、覆岩结构特征为导水裂隙带高度主要影响因素,选取22例导水裂隙带高度实测数据对PSO-BP神经网络进行训练,将训练后的PSO-BP神经网络对2例测试样本的预测结果与实际值进行对比,并与BP神经网络预测模型及经验公式预测结果进行对比。结果表明:PSO-BP神经网络预测模型的平均相对误差为1.55%;BP神经网络预测模型的平均相对误差为4.8%,经验公式的最小相对误差为9.4%,PSO-BP神经网络预测精度明显优于BP神经网络和经验公式,且绝对误差和相对误差变化较稳定,可以有效预测导水裂隙带高度。

     

  • 图  PSO-BP神经网络流程

    Fig. 1  Flow chart of PSO-BP neural network

    图  适应度值变化曲线

    Fig. 2  Variation curve of the fitness value

    表  1  导水裂隙带高度实测样本数据

    Table  1  Measured sampling data of the height of water flowing fractured zone

    编号 工作面 开采厚度/m 开采深度/m 工作面倾斜长度/m 煤层倾角/(°) 覆岩结构特征 导水裂隙带高度/m
    1 北皂煤矿H2101 3.6 359 150 2.3 软弱–软弱 30.0
    2 济宁三号煤矿1301 6.3 480 170 4.0 软弱–坚硬 68.6
    3 钱家营煤矿1672东 3.0 484 143 17.0 坚硬–软弱 40.0
    4 东欢坨煤矿2186 3.7 360 70 23.0 坚硬–软弱 56.8
    5 林南仓煤矿1221 4.0 232 71 8.0 坚硬–软弱 33.0
    6 鲍店煤矿1303 8.7 435 153 8.0 坚硬–坚硬 71.0
    7 鲍店煤矿1316 8.6 357 169 6.5 坚硬–软弱 65.5
    8 南屯煤矿6310 5.8 368 125 6.0 坚硬–坚硬 70.7
    9 兴隆庄煤矿4320 8.0 450 170 8.0 坚硬–软弱 86.8
    10 兴隆庄煤矿1301 6.4 414 193 9.0 坚硬–软弱 72.9
    11 新集一矿1303 7.8 329 134 8.0 软弱–坚硬 83.9
    12 杨村煤矿301 6.4 270 120 11.5 坚硬–软弱 62.0
    13 协鑫煤矿1703-1 9.6 302 120 7.0 软弱–软弱 112.0
    14 北皂煤矿H2106 4.1 330 150 7.0 软弱–软弱 38.8
    15 下沟煤矿ZF2801 9.9 332 93 2.0 软弱–坚硬 125.8
    16 潘一煤矿2622(3) 5.8 553 180 8.0 软弱–软弱 65.3
    17 百善煤矿664 3.0 168 137 5.5 软弱–软弱 27.8
    18 南屯9301 5.3 542 175 15.0 坚硬–坚硬 67.5
    19 王庄煤矿6206 5.9 296 148 4.5 坚硬–坚硬 114.7
    20 梁家煤矿1206 4.0 350 136 9.0 软弱–软弱 35.0
    21 杨庄煤矿8煤层 1.7 320 65 6.0 坚硬–软弱 27.5
    22 某矿3煤层 6.5 263 180 4.0 坚硬–坚硬 83.9
    23 兴隆庄煤矿5306 7.1 412 160 9.5 坚硬–软弱 74.4
    24 鲍店煤矿1310 8.7 409 198 6.0 坚硬–软弱 83.0
    下载: 导出CSV

    表  2  PSO-BP神经网络、BP神经网络及经验公式预测结果

    Table  2  Predicting results of PSO-BP neural network, BP neural network and empirical formulas

    编号 导水裂隙带高度/m 绝对误差/m 相对误差/%
    PSO-BP BP 式(6) 式(7) PSO-BP BP 式(6) 式(7) PSO-BP BP 式(6) 式(7)
    23 75.3 78.6 81.4~102.2 152.0 0.9 4.2 7.0~27.8 77.6 1.2 5.6 9.4~37.4 104.3
    24 81.4 86.3 97.0~117.8 184.0 1.6 3.3 14.0~34.8 101.0 1.9 4.0 16.9~41.9 121.7
    下载: 导出CSV
  • [1] 范立民, 马雄德, 蒋泽泉, 等. 保水采煤研究30年回顾与展望[J]. 煤炭科学技术, 2019, 47(7): 1-30. https://www.cnki.com.cn/Article/CJFDTOTAL-MTKJ201907001.htm

    FAN Limin, MA Xiongde, JIANG Zequan, et al. Review and thirty years prospect of research on water-preserved coal mining[J]. Coal Science and Technology, 2019, 47(7): 1-30. https://www.cnki.com.cn/Article/CJFDTOTAL-MTKJ201907001.htm
    [2] 徐智敏, 高尚, 崔思源, 等. 哈密煤田生态脆弱区保水采煤的水文地质基础与实践[J]. 煤炭学报, 2017, 42(1): 80-87. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201701011.htm

    XU Zhimin, GAO Shang, CUI Siyuan, et al. Hydro-geological basic and practice for water-preserved mining in ecologically vulnerable area: A case study in Hami coalfield[J]. Journal of China Coal Society, 2017, 42(1): 80-87. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201701011.htm
    [3] 范立民. 保水采煤面临的科学问题[J]. 煤炭学报, 2019, 44(3): 667-674. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201903002.htm

    FAN Limin. Some scientific issues in water-preserved coal mining[J]. Journal of China Coal Society, 2019, 44(3): 667-674. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201903002.htm
    [4] 娄高中, 郭文兵, 高金龙. 基于量纲分析的非充分采动导水裂缝带高度预测[J]. 煤田地质与勘探, 2019, 47(3): 147-153.. doi: 10.3969/j.issn.1001-1986.2019.03.023

    LOU Gaozhong, GUO Wenbing, GAO Jinlong. Prediction of the height of water flowing fractured zone under subcritical mining based on dimensional analysis[J]. Coal Geology & Exploration, 2019, 47(3): 147-153.. doi: 10.3969/j.issn.1001-1986.2019.03.023
    [5] 国家安全监管总局, 国家煤矿安监局, 国家能源局, 等. 建筑物、水体、铁路及主要井巷煤柱留设与压煤开采规范[M]. 北京: 煤炭工业出版社, 2017.

    State Administration of Work Safety, National Coal Mine Safety Administration, National Energy Administration, et al. Specifications for coal pillar retention and compressed coal mining in buildings, water bodies, railways and main shafts[M]. Beijing: China Coal Industry Publishing House, 2017.
    [6] 许家林, 朱卫兵, 王晓振. 基于关键层位置的导水裂隙带高度预计方法[J]. 煤炭学报, 2012, 37(5): 762-769. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201205011.htm

    XU Jialin, ZHU Weibing, WANG Xiaozhen. New method to predict the height of fractured water-conducting zone by location of key strata[J]. Journal of China Coal Society, 2012, 37(5): 762-769. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201205011.htm
    [7] 施龙青, 辛恒奇, 翟培合, 等. 大采深条件下导水裂隙带高度计算研究[J]. 中国矿业大学学报, 2012, 41(1): 37-41. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGKD201201008.htm

    SHI Longqing, XIN Hengqi, ZHAI Peihe, et al. Calculating the height of water flowing fracture zone in deep mining[J]. Journal of China University of Mining & Technology, 2012, 41(1): 37-41. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGKD201201008.htm
    [8] 徐智敏, 孙亚军, 高尚, 等. 干旱矿区采动顶板导水裂隙的演化规律及保水采煤意义[J]. 煤炭学报, 2019, 44(3): 767-776. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201903013.htm

    XU Zhimin, SUN Yajun, GAO Shang, et al. Law of mining induced water conduction fissure in arid mining area and its significance in water-preserved coal mining[J]. Journal of China Coal Society, 2019, 44(3): 767-776. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201903013.htm
    [9] 李超峰, 虎维岳, 王云宏, 等. 煤层顶板导水裂缝带高度综合探查技术[J]. 煤田地质与勘探, 2018, 46(1): 101-107.. doi: 10.3969/j.issn.1001-1986.2018.01.018

    LI Chaofeng, HU Weiyue, WANG Yunhong, et al. Comprehensive detection technique for coal seam roof water flowing fractured zone height[J]. Coal Geology & Exploration, 2018, 46(1): 101-107.. doi: 10.3969/j.issn.1001-1986.2018.01.018
    [10] 郭文兵, 娄高中, 赵保才. 芦沟煤矿软硬交互覆岩放顶煤开采导水裂缝带高度研究[J]. 采矿与安全工程学报, 2019, 36(3): 519-526. https://www.cnki.com.cn/Article/CJFDTOTAL-KSYL201903012.htm

    GUO Wenbing, LOU Gaozhong, ZHAO Baocai. Study on the height of water-conductive fracture zone in alternate overburden of soft and hard with top coal caving mining in Lugou coal mine[J]. Journal of Mining & Safety Engineering, 2019, 36(3): 519-526. https://www.cnki.com.cn/Article/CJFDTOTAL-KSYL201903012.htm
    [11] 李振华, 许延春, 李龙飞, 等. 基于BP神经网络的导水裂隙带高度预测[J]. 采矿与安全工程学报, 2015, 32(6): 905-910. https://www.cnki.com.cn/Article/CJFDTOTAL-KSYL201506007.htm

    LI Zhenhua, XU Yanchun, LI Longfei, et al. Forecast of the height of water flowing fractured zone based on BP neural networks[J]. Journal of Mining & Safety Engineering, 2015, 32(6): 905-910. https://www.cnki.com.cn/Article/CJFDTOTAL-KSYL201506007.htm
    [12] 施龙青, 黄纪云, 韩进, 等. 导水裂隙带高度预测的PCA-BP模型[J]. 中国科技论文, 2019, 14(5): 471-475.. doi: 10.3969/j.issn.2095-2783.2019.05.001

    SHI Longqing, HUANG Jiyun, HAN Jin, et al. PCA-BP model for height prediction of water conducting fracture zone[J]. China Science paper, 2019, 14(5): 471-475.. doi: 10.3969/j.issn.2095-2783.2019.05.001
    [13] 邵良杉, 马寒. 煤体瓦斯渗透率的PSO-LSSVM预测模型[J]. 煤田地质与勘探, 2015, 43(4): 23-26.. doi: 10.3969/j.issn.1001-1986.2015.04.005

    SHAO Liangshan, MA Han. Model of coal gas permeability prediction based on PSO-LSSVM[J]. Coal Geology & Exploration, 2015, 43(4): 23-26.. doi: 10.3969/j.issn.1001-1986.2015.04.005
    [14] 毛志勇, 黄春娟, 路世昌, 等. 基于APSO-WLS-SVM的含瓦斯煤渗透率预测模型[J]. 煤田地质与勘探, 2019, 47(2): 66-71. doi: 10.3969/j.issn.1001-1986.2019.02.011

    MAO Zhiyong, HUANG Chunjuan, LU Shichang, et al. Model of gas-bearing permeability prediction based on APSO- WLS-SVM[J]. Coal Geology & Exploration, 2019, 47(2): 66-71. doi: 10.3969/j.issn.1001-1986.2019.02.011
    [15] 温正, 孙华克. MATLAB智能算法[M]. 北京: 清华大学出版社, 2017.

    WEN Zheng, SUN Huake. MATLAB smart algorithm[M]. Beijing: Tsinghua University Press, 2017.
    [16] 吕伟才, 黄晖, 池深深, 等. 概率积分预计参数的神经网络优化算法[J]. 测绘科学, 2019, 44(9): 35-41. https://www.cnki.com.cn/Article/CJFDTOTAL-CHKD201909006.htm

    LYU Weicai, HUANG Hui, CHI Shenshen, et al. Neural network optimization algorithm for the prediction parameters of probability integral method[J]. Science of Surveying and Mapping, 2019, 44(9): 35-41. https://www.cnki.com.cn/Article/CJFDTOTAL-CHKD201909006.htm
    [17] 钱锋. 粒子群算法及其工业应用[M]. 北京: 科学出版社, 2013.

    QIAN Feng. Particle swarm algorithm and its industrial application[M]. Beijing: Science Press, 2013.
    [18] 王雨虹, 刘璐璐, 付华, 等. 基于改进BP神经网络的煤矿冲击地压预测方法研究[J]. 煤炭科学技术, 2017, 45(10): 36-40. https://www.cnki.com.cn/Article/CJFDTOTAL-MTKJ201710006.htm

    WANG Yuhong, LIU Lulu, FU Hua, et al. Study on predicted method of mine pressure bump based on improved BP neural network[J]. Coal Science and Technology, 2017, 45(10): 36-40. https://www.cnki.com.cn/Article/CJFDTOTAL-MTKJ201710006.htm
    [19] 王晓振, 许家林, 韩红凯, 等. 顶板导水裂隙高度随采厚的台阶式发育特征[J]. 煤炭学报, 2019, 44(12): 3740-3749. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201912015.htm

    WANG Xiaozhen, XU Jialin, HAN Hongkai, et al. Stepped development characteristic of water flowing fracture height with variation of mining thickness[J]. Journal of China Coal Society, 2019, 44(12): 3740-3749. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201912015.htm
    [20] 郭文兵, 娄高中. 覆岩破坏充分采动程度定义及判别方法[J]. 煤炭学报, 2019, 44(3): 755-766. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201903012.htm

    GUO Wenbing, LOU Gaozhong. Definition and distinguishing method of critical mining degree of overburden failure[J]. Journal of China Coal Society, 2019, 44(3): 755-766. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201903012.htm
    [21] 尹尚先, 徐斌, 徐慧, 等. 综采条件下煤层顶板导水裂缝带高度计算研究[J]. 煤炭科学技术, 2013, 41(9): 138-142. https://www.cnki.com.cn/Article/CJFDTOTAL-MTKJ201309031.htm

    YIN Shangxian, XU Bin, XU Hui, et al. Study on height calculation of water conducted fractured zone caused by fully mechanized mining[J]. Coal Science and Technology, 2013, 41(9): 138-142. https://www.cnki.com.cn/Article/CJFDTOTAL-MTKJ201309031.htm
    [22] 胡炳南, 张华兴, 申宝宏. 建筑物、水体、铁路及主要井巷煤柱留设与压煤开采指南[M]. 北京: 煤炭工业出版社, 2017.

    HU Bingnan, ZHANG Huaxing, SHEN Baohong. Guidelines for coal pillar retention and compressed coal mining in buildings, water bodies, railways and main shafts[M]. Beijing: China Coal Industry Publishing House, 2017.
    [23] 王正帅, 邓喀中, 谭志祥. 导水裂缝带高度预测的模糊支持向量机模型[J]. 地下空间与工程学报, 2011, 7(4): 723-727. https://www.cnki.com.cn/Article/CJFDTOTAL-BASE201104018.htm

    WANG Zhengshuai, DENG Kazhong, TAN Zhixiang. Height prediction of water fractured zone based on fuzzy SVM[J]. Chinese Journal of Underground Space and Engineering, 2011, 7(4): 723-727. https://www.cnki.com.cn/Article/CJFDTOTAL-BASE201104018.htm
    [24] 樊振丽. 纳林河复合水体下厚煤层安全可采性研究[D]. 北京: 中国矿业大学(北京), 2013.

    FAN Zhenli. Mining safety research of thick coal seam under aquifers and surface water in Nalinhe mine area[D]. Beijing: China University of Mining and Technology(Beijing), 2013.
    [25] 曹博, 白刚, 李辉. 基于PCA-GA-BP神经网络的瓦斯含量预测分析[J]. 中国安全生产科学技术, 2015, 11(5): 84-90. https://www.cnki.com.cn/Article/CJFDTOTAL-LDBK201505016.htm

    CAO Bo, BAI Gang, LI Hui. Prediction of gas content based on PCA-GA-BP neural network[J]. Journal of Safety Science and Technology, 2015, 11(5): 84-90. https://www.cnki.com.cn/Article/CJFDTOTAL-LDBK201505016.htm
    [26] 韩亮, 刘殿书, 李红江, 等. 基于Weibull模型的高台阶抛掷爆破爆堆形态BP神经网络预测[J]. 煤炭学报, 2013, 38(11): 1947-1952. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201311009.htm

    HAN Liang, LIU Dianshu, LI Hongjiang, et al. BP neural network forecast of blasting muck pile form of high bench cast blasting based on Weibull model[J]. Journal of China Coal Society, 2013, 38(11): 1947-1952. https://www.cnki.com.cn/Article/CJFDTOTAL-MTXB201311009.htm
    [27] 臧子婧, 吴海波, 张平松, 等. 基于ABC-BP模型的煤层含气量预测[J]. 煤田地质与勘探, 2021, 49(2): 152-158.. doi: 10.3969/j.issn.1001-1986.2021.02.019

    ZANG Zijing, WU Haibo, ZHANG Pingsong, et al. Prediction of coal seam gas content based on ABC-BP model[J]. Coal Geology & Exploration, 2021, 49(2): 152-158.. doi: 10.3969/j.issn.1001-1986.2021.02.019
    [28] 宁永香, 崔希民. 矿山边坡地表变形的PSO-ELM预测模型[J]. 煤田地质与勘探, 2020, 48(6): 201-206.. doi: 10.3969/j.issn.1001-1986.2020.06.027

    NING Yongxiang, CUI Ximin. PSO-ELM prediction model for surface deformation of mine slope[J]. Coal Geology & Exploration, 2020, 48(6): 201-206.. doi: 10.3969/j.issn.1001-1986.2020.06.027
  • 加载中
图(2) / 表(2)
计量
  • 文章访问数:  189
  • HTML全文浏览量:  11
  • PDF下载量:  28
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-02-05
  • 修回日期:  2021-06-09
  • 发布日期:  2021-08-25
  • 网络出版日期:  2021-09-10

目录

    /

    返回文章
    返回