img

QQ群聊

img

官方微信

高级检索

黄金科学技术 ›› 2022, Vol. 30 ›› Issue (2): 272-281.doi: 10.11872/j.issn.1005-2518.2022.02.052

• 采选技术与矿山管理 • 上一篇    下一篇

基于NPCA-GA-BP神经网络的采场稳定性预测方法

谢饶青(),陈建宏(),肖文丰   

  1. 中南大学资源与安全工程学院,湖南 长沙 410083
  • 收稿日期:2021-05-07 修回日期:2021-08-30 出版日期:2022-04-30 发布日期:2022-06-17
  • 通讯作者: 陈建宏 E-mail:786632011@qq.com;cjh@263.net
  • 作者简介:谢饶青(1996-),女,湖南永州人,硕士研究生,从事矿山安全与灾害防治和矿业系统工程研究工作。786632011@qq.com
  • 基金资助:
    国家自然科学基金项目“地下金属矿采掘计划可视化优化方法与技术研究”(51374242);中南大学研究生自主探索创新项目“多随机扰动下的露天矿卡智能调度优化方法研究”(1053320210291)

Prediction Method of Stope Stability Based on NPCA-GA-BP Neural Network

Raoqing XIE(),Jianhong CHEN(),Wenfeng XIAO   

  1. School of Resources and Safety Engineering,Central South University,Changsha 410083,Hunan,China
  • Received:2021-05-07 Revised:2021-08-30 Online:2022-04-30 Published:2022-06-17
  • Contact: Jianhong CHEN E-mail:786632011@qq.com;cjh@263.net

摘要:

为提高采场稳定性的预测精度,充分考虑采场稳定性高度非线性和受多因素影响的特点,提出了一种基于NPCA-GA-BP神经网络的采场稳定性预测方法。选择影响采场稳定性的10个指标,运用非线性主成分分析减少指标的维度,提取4个主成分综合指标代替原有的10个指标,简化了神经网络结构,提升了运算速度。利用GA的全局寻优特点优化BP神经网络的权值和阈值,进一步增加了神经网络预测精度。以某矿山实测数据为例,对该预测方法进行验证,对比结果显示:NPCA-GA-BP和GA-BP模型的平均相对误差比BP模型分别降低了10.5%和7.6%,表明通过遗传算法优化BP神经网络可显著提高预测精度;NPCA-GA-BP模型的平均相对误差比GA-BP模型降低了2.9%,表明通过非线性主成分分析减少了变量的维度,提高了预测准确率。研究表明:NPCA-GA-BP预测方法具有更高的采场稳定性预测精度,对实现智慧矿山有一定的指导意义。

关键词: 采场稳定性, 预测精度, 非线性主成分分析, 遗传算法, BP神经网络, 非线性相关

Abstract:

Stope stability is a geological mechanics problem that cannot be ignored in mining,and its stability directly affects the safety of mine production and engineering decision-making.Therefore,scientific prediction of stope stability plays a crucial role in mining safety.The stability of stope is a typical nonlinear problem.Since BP neural network has the virtue of tackling complex nonlinear systems,it can be applied to stope stability prediction.Nevertheless,the existing prediction methods either only focus on optimizing the weights and thresholds of the neural network or only consider that the stability of the stope is under the influence of multiple factors and the influencing indexes have a strong correlation,but do not consider the two methods in an integrated manner.Hence,the prediction accuracy of stope stability based on neural network is low,which cannot provide valid support for mine management.Due to the highly nonlinear characteristics of the mining stability system,the traditional principal component analysis will lose a large amount of information.Therefore,we propose a stope stability prediction method using nonlinear principal component analysis combined with BP neural network optimized by the genetic algorithm,which effectively improves the prediction accuracy of stope stability.The nonlinear principal component analysis method performs nonlinear dimensionality reduction on the impact indicators of stope stability,replacing the original multiple indicators with a few principal components that retain the original information,simplifying the neural network structure,and improving the operational efficiency.GA aims to optimize the initial weights and thresholds of the BP neural network to overcome the defects of unstable initial weight thresholds and further improve the accuracy of quarry stability prediction.Taking the measured data of a mine as an example,the effectiveness of the proposed method is verified.The comparison results show that the average relative errors of NPCA-GA-BP and GA-BP models are 10.5% and 7.6% lower than those of BP models,respectively,indicating that the BP neural network is optimized by the genetic algorithm can significantly improve the prediction accuracy.The average relative error of the NPCA-GA-BP model is 2.9% lower than that of the GA-BP model,indicating that the dimension of variables is reduced and the prediction accuracy is increased through nonlinear principal component analysis.It can be concluded that the NPCA-GA-BP prediction method has a higher prediction accuracy of stope stability,and has certain guiding significance for realizing intelligent mine.

Key words: stope stability, prediction accuracy, nonlinear principal component analysis, genetic algorithm, BP neural network, nonlinear correlation

中图分类号: 

  • TD327

图1

NPCA-GA-BP神经网络模型流程图"

表1

采场稳定性样本数据(王杰等,2018)"

样本

编号

X1X2/mX3/MPaX4/%X5/cmX6X7/mX8/m2X9/dX10X11
13380100.5654.0522.504224001032
23380110.2250.1019.604204001433
34380130.3560.1518.514244801233
44370100.6849.0220.104183602032
54370110.2065.3319.003204001832
62340130.3527.7526.00215300534
73330100.6645.0020.402153501734
84330110.6071.3121.404174001433
94360120.8274.4751.004254503042
104340100.0067.1225.004204001642
113280120.9030.1351.902173401224
12328090.0055.0638.20325500832
132280100.0054.6024.903256002031
142290110.0084.2830.403306001131
154340120.0030.4027.104224401343
161320120.0081.5730.301305801541
174270120.4640.2121.204234401533
184270100.4063.7226.604255002333
194260110.2085.0025.00425450822
204260120.2863.4428.00422440222
213310100.4660.7527.303275401632
224300110.5585.0027.703204001433
232300120.0014.3136.002153001344
24332096.4753.0031.254183601942
254320105.8045.6030.004203002342
263320108.4556.8018.75420400732
274320125.3054.4045.454255501232
284297110.0075.0020.00418400532
29429797.4340.6518.293153502042
30328580.0040.0030.772255001542
313320120.3062.4018.523224501043
323275100.4055.0017.443203601842
33228590.0048.6027.78217340344

表2

半定量取值方法"

赋值影响因素
X1X6X10
1断层贯穿围岩水体对围岩影响大相邻采场开挖影响大
2断层切割岩体水体对围岩有影响相邻采场开挖影响较大
3断层影响小水体对围岩影响较小相邻采场开挖影响一般
4无断层、无褶皱水体对围岩无影响相邻采场开挖无影响

表3

各指标之间的Pearson相关性系数"

因素X1X2X3X4X5X6X7X8X9X10X11
X11.0000.1200.0710.163-0.1080.746-0.130-0.1680.231-0.205-0.004
X20.1201.0000.228-0.036-0.1750.223-0.151-0.1830.2270.1810.052
X30.0700.2281.000-0.0160.1760.0860.0310.002-0.169-0.2470.250
X40.160-0.036-0.0161.000-0.1120.2530.5310.4970.015-0.199-0.551
X5-0.100-0.1750.176-0.1121.000-0.1860.2090.1420.141-0.0330.047
X60.7400.2230.0860.253-0.1861.0000.0740.0170.134-0.197-0.250
X7-0.130-0.1510.0310.5310.2090.0741.0000.9070.101-0.078-0.633
X8-0.160-0.1830.0020.4970.1420.0170.9071.0000.046-0.142-0.602
X90.2300.227-0.1690.0150.1410.1340.1010.0461.0000.391-0.229
X10-0.2000.181-0.247-0.199-0.033-0.197-0.078-0.1420.3911.000-0.063
X11-0.0040.0520.250-0.5510.047-0.250-0.633-0.602-0.229-0.0631.000

图2

主成分特征值"

图3

主成分方差贡献率"

表4

前4个主成分及其累计贡献率"

因素Y1Y2Y3Y4
X10.1330.497-0.420-0.212
X20.0640.008-0.0580.072
X3-0.032-0.007-0.024-0.086
X40.0400.5440.6410.165
X5-0.002-0.2120.131-0.882
X60.1130.587-0.319-0.168
X70.0540.1190.378-0.170
X80.0350.0840.380-0.157
X90.970-0.1460.0410.015
X100.136-0.162-0.0290.234
特征值0.0670.0430.0280.016
贡献率/%38.024.716.39.5
累计贡献率/%38.063.679.989.4

表5

非线性降维后的样本数据"

样本编号Y1Y2Y3Y4采场稳定性样本编号Y1Y2Y3Y4采场稳定性
11.4961.4052.218-1.6042180.7991.6152.225-1.7842
21.6371.3752.181-1.5513191.6821.3352.397-1.6952
31.5921.5082.235-1.5793201.6421.4562.341-1.6583
41.7951.3972.103-1.5692211.5160.7201.950-1.7334
51.7501.4152.249-1.5272221.7731.2972.197-1.6812
61.1160.9832.093-1.5894231.8631.3122.094-1.7042
71.6671.1362.186-1.4964241.3431.4492.202-1.5342
81.6511.4992.209-1.5793251.5901.4112.295-1.9462
92.0001.3862.362-1.9152261.2141.5902.220-1.5632
101.7201.4552.231-1.6232271.7831.2602.049-1.4882
111.4851.0122.148-1.9494281.7511.2182.456-1.6321
121.3851.3212.358-1.8182291.5101.3512.606-1.6911
131.6201.2792.035-1.7293301.6491.0982.323-1.6952
141.5680.8682.863-1.5171311.4961.3732.308-1.4943
151.6711.3762.118-1.6583321.7391.2972.233-1.4502
161.8611.4502.306-1.7113330.9331.1362.291-1.5574
171.3911.6162.354-1.7182

图4

3种预测方法对比"

表6

3种方法预测精度"

样本编号BP算法GA-BP算法NPCA-GA-BP算法
EEˉEEˉEEˉ
300.100.1300.080.0541.930.025
310.160.093.09
320.140.032.06
330.140.013.96
Bi Aorui, Luo Zhengshan, Qiao Wei,et al,2018.Prediction of pipeline inner-corrosion based on principal component analysis and particle swarm optimization-support vector machine[J].Surface Technology,47(9):133-140.
Ding S, Zhang Y, Chen J,et al,2013.Research on using genetic algorithms to optimize Elman neural networks [J].Neural Computing and Applications,23(2):293-297.
Erten G E, Keser S B, Yavuz M,2021.Grid search pptimised artificial neural network for open stope stability prediction[J].International Journal of Mining, Reclamation and Environment,35(8):600-617.
Gong Jian, Hu Nailian, Wang Xiaodong,et al,2015.Stability analysis and rock movement prediction of stope roof below the subsidence area [J].Journal of Mining & Safety Engineering,32(2):337-342.
Guan Ziqi, Zhu Yulong, Liu Xiaoguang,et al,2019.Modeling of weld pool illumination based on GA optimizing BP neural network [J].Hot Working Technology,48(7):216-220.
Guo Chao, Song Weihua, Wei Wei,2014.Stope roof stability prediction based on both SVM and grid-search method [J].China Safety Science Journal,24(8):31-36.
Hu Hongwang, Ye Yicheng, Geng Hongbo,et al,2018.Study on stability evalution of layered deposit goaf based on BP neural network[J].Industrial Minerals & Processing, 47(3):60-63,69.
Ilonen J, Kamarainen J K, Lampinen J,2003.Differential evolution training algorithm for feed-forward neural networks[J].Neural Processing Letters,17(1):93-105.
Ji Long, Shen Hongfang, Xu Chunchun,et al,2019.Comprehensive evaluation of green super rice varieties based on nonlinear principal component analysis[J].Acta Agronomica Sinica,45(7):982-992.
Kang Yu, Wang Xinmin, Zhang Qinli,et al,2015.Application of VW-UM model in stope stability evaluation[J].China Safety Science Journal,25(7):128-134.
Lei Dingyou, Ma Qiang, Xu Xinping,et al,2018.Forecasting method of expressway traffic volume based on NPCA and GA-RBF[J].Journal of Traffic and Transportation Engineering,18(3):210-217.
Li Qihang, Li Xiaoshuang, Geng Jiabo,et al,2021.FLAC3D numerical simulation of open-pit transformation to underground slope and stability[J].Nonferrous Metals(Mining Section), 73(2):5-10.
Li Xiaobei, Dai Xingguo, Wang Xinmin,et al,2015.Prediction of stope roof displacement based on CT-GRNN[J].Mining and Metallurgical Engineering,35(6):30-34.
Lima D C L, Limao D O R, Roisenberg M,2016.Optimization of neural networks through grammatical evolution and a genetic algorithm[J].Expert Systems with Applications,56(2):368-384.
Ling Biaocan, Peng Suping, Zhang Shenghe,et al,2003.Dynamic engineering classification of stope roof stability [J].Chinese Journal of Rock Mechanics and Engineering,(9):1474-1477.
Ning Yulin, Jiang Fanjun, Lin Weixing,2019.Study on stability evaluation and treatment scheme of goaf in Huanggang iron mine[J].Mining Research and Development,39(8):74-77.
Oo C T, Sasaoka T, Shimada H,et al,2021.Evaluation of stope stability in underground mine;Hermyingyi(Sn-W deposit)mine in Myanmar [J].Journal of Geoscience and Environment Protection,9(1):107-120.
Qiao Shoule, Li Qing, Liang Zhiqiang,et al,2018.Mining field stability analysis based on FLAC3D and Monte Carlo method[J].Mining and Metallurgy,27(4):45-49.
Santos J, Ferreira A, Flintsch G,2019.An adaptive hybrid genetic algorithm for pavement management[J].International Journal of Pavement Engineering,20(3):226-286.
Shao Liangshan, Xu Bo,2015.KPCA-SVM model for predicting Karst collapse tendency level[J].China Safety Science Journal,25(3):60-65.
Wang J, Fang J D, Zhao Y D,2019.Visual prediction of gas diffusion concentration based on regression analysis and BP neural network[J].The Journal of Engineering,(13):19-23.
Wang Jie, Luo Zhouquan, Qin Yaguang,et al,2018.Prediction of stope stability based on random forest[J].China Safety Science Journal,28(3):155-160.
Wang Zhenhua, Gong Dianyao, Li Guangtao,et al,2018.Bending force prediction model in hot strip rolling based on artificial neural network optimize by genetic algorithm[J].Journal of Northeastern University( Natural Science),39(12):1717-1722.
Wei Pengyu, Pan Fucheng, Li Shuai,2018.Study on classification of improved artificial bee colony algorithm to optimi-zation of BP neural network[J].Computer Engineering and Applications,54(10):158-163.
Wu Shuliang, Chen Jianhong, Yang Shan,2012.Optimization of bolting scheme based on combination of principal component analysis and BP neural network [J].Chinese Journal of Engineering Design,19(2):150-155.
Zhang Fei, Yang Tianhong, Hu Gaojian,2018.Stability evaluation of surrounding rock and parameter optimization of stope under complex stress disturbance[J].Journal of Nor-theastern University(Natural Science),39(5):699-704.
Zhang Qinli, Li Xieping, Yang Wei,2013.Optimization of filling slurry ratio in a mine based on back-propagation neural network [J].Journal of Central South University (Science and Technology),44(7):2867-2874.
Zhang Siyuan, Bao Yanping, Zhang Chaojie,et al,2017.Prediction model of aluminum consumption with BP neural networks in IF steel production[J].Chinese Journal of Engineering,39(4):511-519.
Zhao Guangyuan, Ma Fei,2018.Prediction of dust concentration based on particle swarm optimization BP neural network [J].Measurement & Control Technology,37(6):20-23.
Zhao Xingdong, Li Huaibin, Zhang Shujing,et al,2020.Analyzing and controlling the stope stability from open-pit to underground mining in Qinglonggou gold mine[J].Metal Mine,49(4):10-14.
毕傲睿,骆正山,乔伟,等,2018.基于主成分和粒子群优化支持向量机的管道内腐蚀[J].表面技术,47(9):133-140.
龚剑,胡乃联,王孝东,等,2015.塌陷区下部采场顶板稳定性分析及岩移预测[J].采矿与安全工程学报,32(2):337-342.
关子奇,朱玉龙,刘晓光,等,2019.基于GA优化BP神经网络的焊接熔池照度建模[J].热加工工艺,48(7):216-220.
郭超,宋卫华,魏威,2014.基于网格搜索—支持向量机的采场顶板稳定性预测[J].中国安全科学学报,24(8):31-36.
胡洪旺,叶义成,耿宏波,等,2018.基于BP神经网络的层状矿床采空区稳定性评价研究[J].化工矿物与加工,47(3):60-63,69.
纪龙,申红芳,徐春春,等,2019.基于非线性主成分分析的绿色超级稻品种综合评价[J].作物学报,45(7):982-992.
康虞,王新民,张钦礼,等,2015.VW-UM模型在采场稳定性评价中得应用[J].中国安全科学学报,25(7):128-134.
雷定猷,马强,徐新平,等,2018.基于非线性主成分分析和GA-RBF的高速公路交通量预测方法[J].交通运输工程学报,18(3):210-217.
李启航,李小双,耿加波,等,2021.FLAC3D数值模拟露天转地下边坡及采场稳定性研究[J].有色金属(矿山部分),73(2):5-10.
李小贝,戴兴国,王新民,等,2015.基于CT-GRNN模型的采场顶板位移预测[J].矿冶工程,35(6):30-34.
凌标灿,彭苏萍,张慎河,等,2003.采场顶板稳定性动态工程分类[J].岩石力学与工程学报,(9):1474-1477.
甯瑜琳,姜凡均,林卫星,2019.黄岗铁矿采空区稳定性评价及治理方案研究[J].矿业研究与开发,39(8):74-77.
乔守乐,卿黎,梁志强,等,2018.基于FLAC3D和Monte Carlo法采场稳定性分析[J].矿冶,27(4):45-49.
邵良杉,徐波,2015.岩溶塌陷倾向性等级的KPCA-SVM预测模型[J].中国安全科学学报,25(3):60-65.
王杰,罗周全,秦亚光,等,2018.基于随机森林理论的采场稳定性预测研究[J].中国安全科学学报, 28(3):155-160.
王振华,龚殿尧,李广焘,等,2018.遗传算法优化神经网络的热轧带钢弯辊力预报模型[J].东北大学学报(自然科学版),39(12):1717-1722.
韦鹏宇,潘福成,李帅,2018.改进人工蜂群优化BP神经网络的分类研究[J].计算机工程与应用,54(10):158-163.
邬书良,陈建宏,杨珊,2012.基于主成分分析与BP网络的锚杆支护方案优选[J].工程设计学报,19(2):150-155.
张飞,杨天鸿,胡高建,2018.复杂应力扰动下围岩稳定性评价与采场参数优化[J].东北大学学报(自然科学版),39(5):699-704.
张钦礼,李谢平,杨伟,2013.基于BP网络的某矿山充填料浆配比优化[J].中南大学学报(自然科学版),44(7):2867-2874.
张思源,包燕平,张超杰,等,2017.BP神经网络IF钢铝耗的预测模型[J].工程科学学报,39(4):511-519.
赵广元,马霏,2018.粒子群算法优化BP神经网络的粉尘浓度预测[J].测控技术,37(6):20-23.
赵兴东,李怀宾,张姝婧,等,2020.青龙沟金矿露天转地下采场稳定性分析及控制[J].金属矿山,49(4):10-14.
[1] 公凡波,毕林. 露天矿电铲铲装移动轨迹规划研究[J]. 黄金科学技术, 2021, 29(1): 43-52.
[2] 许瑞, 侯奎奎, 王玺, 刘兴全, 李夕兵. 基于核主成分分析与SVM的岩爆烈度组合预测模型[J]. 黄金科学技术, 2020, 28(4): 575-584.
[3] 王梅,陈建宏,杨珊. 基于等维动态马尔科夫SCGM(1,1)C模型的黄金价格预测[J]. 黄金科学技术, 2020, 28(1): 158-166.
[4] 卜斤革,陈建宏. 基于粒子群算法优化BP神经网络的溶浸开采浸出率预测[J]. 黄金科学技术, 2020, 28(1): 82-89.
[5] 肖文丰,陈建宏,陈毅,王喜梅. 基于神经网络与遗传算法的多目标充填料浆配比优化[J]. 黄金科学技术, 2019, 27(4): 581-588.
[6] 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.
[7] 胡建华,任启帆,亓中华,张纪伟. 卧虎山铁矿采场极限暴露面积回归优化模型[J]. 黄金科学技术, 2018, 26(4): 503-510.
[8] 邓高,杨珊. 基于组合预测与变精度粗糙模糊集的采空区稳定性评价[J]. 黄金科学技术, 2017, 25(3): 98-107.
[9] 刘志祥,刘奕然,兰明. 矿井涌水量预测的PCA-GA-ELM模型及应用[J]. 黄金科学技术, 2017, 25(1): 61-67.
[10] 史秀志,范玉乾,尚雪义. 基于PCA-BP神经网络模型的充填体强度预测[J]. 黄金科学技术, 2016, 24(3): 64-69.
[11] 李夕兵,朱玮,刘伟军,张德明. 基于主成分分析法与RBF神经网络的岩体可爆性研究[J]. 黄金科学技术, 2015, 23(6): 58-63.
[12] 过江,张碧肖. 基于PCA与BP神经网络的充填管道失效风险评估[J]. 黄金科学技术, 2015, 23(5): 66-71.
[13] 张玲玲,李国清*,姜光成,李威,胡乃联. 基于BP神经网络的地质缺失数据处理方法[J]. 黄金科学技术, 2015, 23(5): 53-59.
[14] 陈建宏,朱鼎耀,陈轶俊,叶阿明,邱文. 基于PCA-BP神经网络的尾矿库坝体稳定性分析[J]. 黄金科学技术, 2015, 23(5): 47-52.
[15] 叶培. 一种新的量子遗传算法在阳山金矿GPS卫星信号去噪处理中的应用探讨[J]. 黄金科学技术, 2015, 23(2): 83-87.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!