Study on dynamic modification method of 3D model of coal seam in fully mechanized working face
-
摘要: 综采工作面的高精度煤层地理信息是实现智能无人开采的关键，但现阶段所构建的煤层三维模型垂向精度较低，无法满足智能开采的实际需求。针对该问题，提出了一种综采工作面煤层三维模型动态修正方法。将得到的初始煤层三维模型静态数据及开采过程中采煤机截割产生的动态数据融合，基于长短期记忆网络(LSTM）预测算法及其改进算法（基于空间卷积长短期记忆网络(Conv LSTM)、编码−解码长短期记忆网络(Encoder-Decoder LSTM)的预测算法），根据上一回采阶段的煤层数据，动态预测下一阶段未开采区的煤层底板曲面和煤层厚度。采用双层循环嵌套的网格搜索方法对上述3种预测算法进行参数调优，获取未开采区煤层底板曲面和煤层厚度的高精度垂向分布数据，作为煤层三维模型修正值，动态修正下一阶段未开采区的煤层三维模型；随着工作面不断开采，利用新获取的修正数据持续动态修正并更新初始煤层三维模型，从而提升初始煤层三维模型精度，使动态修正后的煤层三维模型能更准确地反映综采工作面实际煤层分布。以山西吕梁市某煤矿18201工作面煤层三维模型为例，采用提出的动态修正方法对该模型进行修正，在工作面推进方向16~23.2 m范围内，动态修正后的煤层底板平均误差为0.068 5 m，煤层顶板平均误差为0.076 m，相较于修正前的底板平均误差0.20 m、煤层厚度垂向平均误差0.40 m，动态修正后的煤层三维模型精度大大提升，证实了该修正方法的有效性。Abstract: The high-precision coal seam geographic information of fully mechanized working face is the key to realizing intelligent unmanned mining. However, the vertical precision of 3D model of coal seam constructed at this stage is low. The model cannot meet the actual needs of intelligent mining. In order to solve this problem, a dynamic modification method of 3D model of coal seam in fully mechanized working face is proposed. The static data of the initial coal seam 3D model and the dynamic data generated by the shearer cutting in the mining process are fused. The method is based on the prediction algorithms of long-short term memory (LSTM) network and its improved algorithm. The improved algorithms are based on the convolutional long-short term memory network (Conv LSTM) and encoder-decoder long-short term memory network (Encoder-Decoder LSTM). The coal seam floor curved surface and the coal seam thickness of the unmined area in the next stage are dynamically predicted according to the coal seam data of the previous mining stage. The parameters of the above three prediction algorithms are optimized by using the grid search method of double-layer loop nesting. The obtained high-precision vertical distribution data of the coal seam floor curved surface and the coal seam thickness of the unexploited area are taken as the coal seam 3D model correction value. The correction value is used to dynamically correct the coal seam 3D model of the unexploited area in the next stage. With the continuous mining of the working face, the newly obtained correction data is used to continuously and dynamically correct and update the initial coal seam 3D model, so as to improve the precision of the initial coal seam 3D model. Therefore, the dynamic modified coal seam 3D model can reflect the actual coal seam distribution of fully mechanized working face more accurately. Taking the coal seam 3D model of 18201 working face of a coal mine in Lvliang, Shanxi Province as an example, the proposed dynamic correction method is used to correct the coal seam 3D model. Within the range of 16-23.2 m in the advancing direction of the working face, the average error of the coal seam floor after the dynamic correction is 0.068 5 m. The average error of the coal seam roof is 0.076 m. Compared with the average floor error of 0.20 m and vertical average error of 0.40 m of the coal seam thickness before correction, the precision of the coal seam 3D model after dynamic correction is greatly improved. The results confirm the effectiveness of the correction method.
-
-
![]()
图 2 LSTM细胞的信息流动结构
Figure 2. Information flow structure of long-short term memory (LSTM) cell
![]()
图 8 工作面推进方向16.8 m处煤层底板曲面修正值预测结果
Figure 8. Prediction results of coal seam floor curved surface correction value at location of 16.8 m in advancing direction of working face
![]()
图 9 工作面推进方向16.8 m处煤层厚度修正值预测结果
Figure 9. Prediction results of coal seam thickness correction value at location of 16.8 m in advancing direction of working face
![]()
图 10 煤层三维模型修正后的底顶板误差分布
Figure 10. Error distribution of floor and top of coal seam 3D model after correction
![]()
图 11 煤层三维模型修正前后底顶板预测误差分布占比
Figure 11. Proportion of error distribution of floor and top prediction before and after coal seam 3D model correction
表 1 煤层底板曲面预测的数据组合方式
Table 1 Prediction data combination mode of coal seam floor curved surface
类型 序号 输入序列数据 标签序列数据 训练集 1 f 1 d1 2 f 2 d2 $\vdots $ $\vdots $ $\vdots $ 20 f 20 d20 测试集 1 f 21 d21 2 f 22 d22 $\vdots $ $\vdots $ $\vdots $ 10 f 30 d30 
下载: 导出CSV
表 2 超参数候选值
Table 2 Candidate values of hyperparameters
超参数 候选值 输入输出序列步长(Steps) (4:1), (3:1), (2:1), (1:1) 隐藏层神经元个数(Units) 100, 90, 70, 50, 30, 20, 10 激活函数(Activation) ReLU, Sigmoid, Tanh, Softplus 优化器(Optimizer) SGD, Adadelta, RMSProp, Adagrad, Adam 损失函数(Loss) MAE 训练周期(Epoch) 10, 30, 50, 80, 100, 150, 200
下载: 导出CSV
表 3 煤层底板曲面修正值预测算法超参数优化结果
Table 3 Hyperparameters optimization results of prediction algorithms for coal seam floor curved surface correction value
预测算法 [Steps, Units, Activation,
Optimizer, Epoch]MAE/m LSTM [(1, 1), 20, Softplus, Adadelta, 100] 0.594 Conv LSTM [(1, 1), 10, Softplus, Adagrad, 100] 0.074 Encoder-Decoder LSTM [(1, 1), 20, Softplus, Adagrad, 50] 0.323
下载: 导出CSV
表 4 煤层厚度预测算法超参数优化结果
Table 4 Hyperparameter optimization results of coal seam thickness prediction algorithms
预测算法 [Steps, Units, Activation,
Optimizer, Epoch]RMSE/m LSTM [(1, 1), 100, ReLU, Adagrad, 100] 0.051 Conv LSTM [(1, 1), 100，Softplus，Adagrad，90] 0.051 Encoder-Decoder LSTM [(1, 1), 100，ReLU，Adagrad，90] 0.049
下载: 导出CSV
-
[1] 葛世荣. 煤矿智采工作面概念及系统架构研究[J]. 工矿自动化,2020,46(4):1-9. DOI: 10.13272/j.issn.1671-251x.2020030070 GE Shirong. Research on concept and system architecture of smart mining workface in coal mine[J]. Industry and Mine Automation,2020,46(4):1-9. DOI: 10.13272/j.issn.1671-251x.2020030070
[2] 原长锁,王峰. 综采工作面透明化开采模式及关键技术[J]. 工矿自动化,2022,48(3):11-15,31. YUAN Changsuo,WANG Feng. Transparent mining mode and key technologies of fully mechanized working face[J]. Journal of Mine Automation,2022,48(3):11-15,31.
[3] 王峰. 基于透明工作面的智能化开采概念、实现路径及关键技术[J]. 工矿自动化,2020,46(5):39-42. WANG Feng. Concept,realization path and key technologies of intelligent mining based on transparent longwall face[J]. Industry and Mine Automation,2020,46(5):39-42.
[4] 袁亮. 煤炭精准开采科学构想[J]. 煤炭学报,2017,42(1):1-7. YUAN Liang. Scientific conception of precision coal mining[J]. Journal of China Coal Society,2017,42(1):1-7.
[5] 毛善君,杨乃时,高彦清,等. 煤矿分布式协同“一张图”系统的设计和关键技术[J]. 煤炭学报,2018,43(1):280-286. MAO Shanjun,YANG Naishi,GAO Yanqing,et al. Design and key technology research of coal mine distributed cooperative "one map" system[J]. Journal of China Coal Society,2018,43(1):280-286.
[6] ZHU Liangfeng,LI Mingjiang,LI Changling,et al. Coupled modeling between geological structure fields and property parameter fields in 3D engineering geological space[J]. Engineering Geology,2013,167(24):105-116.
[7] TERCAN A,EÜNVER B,HINDISTAN M A,et al. Seam modeling and resource estimation in the coalfields of western Anatolia[J]. International Journal of Coal Geology,2013,112(2):94-106.
[8] 贾庆仁,车德福,李佳徐,等. 动态精化的煤层三维建模方法[J]. 东北大学学报(自然科学版),2018,39(5):726-730. JIA Qingren,CHE Defu,LI Jiaxu,et al. Three-dimensional modeling method of coal seam with gradual refinement[J]. Journal of Northeastern University (Natural Science),2018,39(5):726-730.
[9] 修春华,车德福,贾国兵. 含复杂地质构造的三维煤层动态建模方法[J]. 矿山测量,2015(6):52-55,59. DOI: 10.3969/j.issn.1001-358X.2015.06.16 XIU Chunhua,CHE Defu,JIA Guobing. Dynamic modeling method of 3D coal seam containing complex geological structure[J]. Mine Surveying,2015(6):52-55,59. DOI: 10.3969/j.issn.1001-358X.2015.06.16
[10] 刘万里,张学亮,王世博. 采煤工作面煤层三维模型构建及动态修正技术[J]. 煤炭学报,2020,45(6):1973-1983. DOI: 10.13225/j.cnki.jccs.zn20.0364 LIU Wanli,ZHANG Xueliang,WANG Shibo. Modeling and dynamic correction technology of 3D coal seam model for coal-mining face[J]. Journal of China Coal Society,2020,45(6):1973-1983. DOI: 10.13225/j.cnki.jccs.zn20.0364
[11] 孙振明,毛善君,祁和刚,等. 煤矿三维地质模型动态修正关键技术[J]. 煤炭学报,2014,39(5):918-924. SUN Zhenming,MAO Shanjun,QI Hegang,et al. Dynamic correction of coal mine three-dimensional geological model[J]. Journal of China Coal Society,2014,39(5):918-924.
[12] 梁耍,王世博,谢洋,等. 基于LSTM的煤层厚度动态预测方法研究[J]. 煤炭科学技术,2021,49(增刊1):150-157. LIANG Shua,WANG Shibo,XIE Yang,et al. Dynamic prediction method of coal seam thickness based on LSTM[J]. Coal Science and Technology,2021,49(S1):150-157.
[13] 李娟莉, 杜文勇, 谢嘉成, 等. 煤层数字高程模型构建与动态修正方法[J/OL]. 煤炭科学技术: 1-8[2022-05-20]. DOI: 10.13199/j.cnki.cst.2022-0147. LI Juanli, DU Wenyong, XIE Jiacheng, et al. Coal seam digital elevation model construction and dynamic correction method [J/OL]. Coal Science and Technology: 1-8[2022-05-20]. DOI: 10.13199/j.cnki.cst.2022-0147.
[14] HOCHREITER S,SCHMIDHUBER J. Long short-term memory[J]. Neural Computation,1997,9(8):1735-1780. DOI: 10.1162/neco.1997.9.8.1735
[15] GRAVES A,SCHMIDHUBER J. Framewise phoneme classification with bidirectional LSTM and other neural network architectures[J]. Neural Networks,2005,18(5/6):602-610. DOI: 10.1016/j.neunet.2005.06.042
[16] BAPPY J H,SIMONS C,NATARAJ L,et al. Hybrid LSTM and encoder–decoder architecture for detection of image forgeries[J]. IEEE Transactions on Image Processing,2019,28(7):3286-3300. DOI: 10.1109/TIP.2019.2895466
[17] 葛世荣,苏忠水,李昂,等. 基于地理信息系统(GIS) 的采煤机定位定姿技术研究[J]. 煤炭学报,2015,40(11):2503-2508. GE Shirong,SU Zhongshui,LI Ang,et al. Study on the positioning and orientation of a shearer based on geographic information system[J]. Journal of China Coal Society,2015,40(11):2503-2508.
[18] 郝尚清,李昂,王世博,等. 采煤机惯性导航安装偏差对定位误差的影响[J]. 煤炭学报,2015, 40(8):1963-1968. DOI: 10.13225/j.cnki.jccs.2014.1561 HAO Shangqing,LI Ang,WANG Shibo,et al. Effects of shearer inertial navigation installation noncoincidence on shearer positioning error[J]. Journal of China Coal Society,2015, 40(8):1963-1968. DOI: 10.13225/j.cnki.jccs.2014.1561
[19] 王玉璟. 空间插值算法的研究及其在空气质量监测中的应用[D]. 开封: 河南大学, 2010. WANG Yujing. Research of spatial interpolation algorithm and its application in air quality monitoring[D]. Kaifeng: Henan University, 2010.
-
期刊类型引用(18)
1. 李富强. 基于深度卷积神经网络与多源信号的煤岩识别研究. 煤炭技术. 2025(03): 233-238 . 
百度学术
2. 高如新，杜亚博，常嘉浩. 基于改进YOLOX-S的轻量化煤矸石检测方法研究. 河南理工大学学报(自然科学版). 2024(04): 133-140 . 百度学术
3. 陈晓杰，王亮，赵美玲，刘光伟，涂俊雄. 基于ECA-YOLOv5s的煤矿带式输送机异物检测网络模型. 采矿技术. 2024(04): 316-324 . 百度学术
4. 陈世涛，张敏，栗超. 基于YOLOv5的带式输送机煤堆异物检测. 洁净煤技术. 2024(S2): 12-18 . 百度学术
5. 涂灿. VCS智能干选机的试验研究. 煤炭加工与综合利用. 2023(06): 37-41 . 百度学术
6. 高如新，常嘉浩，杜亚博，刘群坡. 基于改进YOLOv5s的煤矸石目标检测算法. 电子测量技术. 2023(13): 95-101 . 百度学术
7. 汪岩，李自强. 基于AI图像处理的煤矸石特征提取及分类方法. 煤炭技术. 2023(11): 231-233 . 百度学术
8. 曹现刚，刘思颖，王鹏，许罡，吴旭东. 面向煤矸分拣机器人的煤矸识别定位系统研究. 煤炭科学技术. 2022(01): 237-246 . 百度学术
9. 倪云峰，封子杰，郭苹，王静. 基于卷积神经网络的煤矸石识别算法研究. 现代电子技术. 2022(10): 57-62 . 百度学术
10. 张红，李晨阳. 基于光学图像的煤矸石识别方法综述. 煤炭工程. 2022(07): 159-163 . 百度学术
11. 申利飞，田子建，白林绪. 改进纹理模糊筛选下煤矸石X射线图像处理. 激光与红外. 2022(07): 1090-1097 . 百度学术
12. 陈彪，卢兆林，代伟，邵明，于大伟，董良. 基于轻量化HPG-YOLOX-S模型的煤矸石图像精准识别. 工矿自动化. 2022(11): 33-38 . 本站查看
13. 郑新涛，苏道玉. 基于小波矩的智能手绘草图识别系统设计. 现代电子技术. 2021(12): 177-181 . 百度学术
14. 胡璟皓，高妍，张红娟，靳宝全. 基于深度学习的带式输送机非煤异物识别方法. 工矿自动化. 2021(06): 57-62+90 . 本站查看
15. 王冠军，苏婷婷，刘文博，钱智平，李佳泽. 基于EAIDK的智能煤矸分拣系统设计. 工矿自动化. 2020(01): 105-108 . 本站查看
16. 柴炳升，胡峰. 石槽村选煤厂重介质浅槽分选工艺探究. 煤炭加工与综合利用. 2020(08): 33-34+38 . 百度学术
17. 章振原，秦训鹏，李轶峰. 基于机器视觉的废旧有色金属碎料识别方法. 激光与光电子学进展. 2020(16): 194-201 . 百度学术
18. 潘卫东，李新源，员明涛，袁永康，杨克虎. 基于顶煤运移跟踪仪的自动化放煤技术原理及应用. 煤炭学报. 2020(S1): 23-30 . 百度学术
其他类型引用(31)
计量
- 文章访问数: 227
- HTML全文浏览量: 52
- PDF下载量: 36
- 被引次数: 49
