Predicting the risk of water-sand inrushes from coal seam roofs based on a small-size sample dataset

Authors

ZHANG Wenquan, College of Energy and Mining Engineering, Shandong University of Science and Technology, Qingdao 266590, China; State Key Laboratory of Disaster Prevention and Ecology Protection in Open-Pit Coal Mines, Shandong University of Science and Technology, Qingdao 266590, ChinaFollow
LI Zixu, College of Energy and Mining Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaFollow
ZHU Xianxiang, College of Energy and Mining Engineering, Shandong University of Science and Technology, Qingdao 266590, China
QIU Wei, College of Energy and Mining Engineering, Shandong University of Science and Technology, Qingdao 266590, China
ZHANG Chengjie, College of Energy and Mining Engineering, Shandong University of Science and Technology, Qingdao 266590, China

Abstract

Objective The eastern and northern regions of China exhibit thick unconsolidated layers and thin bedrocks, with water-sand inrush disasters occurring frequently. Therefore, accurately predicting the risk of water-sand inrushes from coal seam roofs holds great significance for safe coal mining. However, the water-sand inrushes in these regions exhibit complex disaster-causing mechanisms, which involve the coupling effects of multiple factors. Accordingly, the real-time prediction of the water-sand inrush risk in the field poses challenges including high risk and cost. These issues lead to difficult data acquisition and severely insufficient samples, limiting the accuracy and performance of traditional prediction models. Therefore, there is an urgent need to explore effective prediction methods suitable for a small sample size.Methods Based on a review and analysis of the measured field data and historical cases of mining faces adjacent to unconsolidated layers, this study determined 11 factors influencing water-sand inrushes (e.g., the thickness of aquifers at the unconsolidated-layer bottom and bedrock thickness) and constructed the original sample dataset. Subsequently, the intrinsic relationships and correlations among various influencing factors were discovered using Spearman correlation. A risk prediction model for water-sand inrushes, termed CTGAN-DPSO-RF, was developed based on conditional tabular generative adversarial networks (CTGAN), detecting particle swarm optimization (DPSO) algorithm, and random forest (RF) algorithm. The quality of data synthesized using CTGAN was explored. Furthermore, the effectiveness of the CTGAN-DPSO-RF model was validated through comparison with the DPSO-SVM and DPSO-XGBoost models, along with two engineering cases.Results and Conclusions Among the 11 influencing factors, the caving zone height exhibited the strongest correlation with mining height (correlation coefficient: 0.93), while the water pressure of aquifers at the unconsolidated-layer bottom presented the weakest correlation with the height of the hydraulically conductive fracture zone. The data synthesized using CTGAN highly resembled the original data, with a comprehensive quality score reaching up to 85.03%. The DPSO algorithm yielded an optimal fitness of 0.9265 after the hyperparameter tuning, outperforming the particle swarm optimization (PSO) algorithm. The CTGAN-DPSO-RF model yielded accuracy (A_c), weighted precision (P_w), weighted recall (R_w), and weighted F1-score (F1_w) consistently reaching 1.0 on the test set, outperforming its counterparts. The risk prediction results of two mining faces derived using the proposed model were consistent with actual mining conditions. By synthesizing high-quality data to expand the sample dataset and optimizing hyperparameters, the proposed model effectively overcomes the low accuracy and poor performance of traditional models under a small sample size, providing a new method for predicting the risk of water-sand inrushes from coal seam roofs under conditions of thick unconsolidated layers and thin bedrocks.

Keywords

coal seam roof, thick unconsolidated layer and thin bedrock, water-sand inrush, small-size sample data, conditional tabular generative adversarial network (CTGAN), detecting particle swarm optimization (DPSO) algorithm, risk prediction

DOI

10.12363/issn.1001-1986.25.09.0713

Recommended Citation

ZHANG Wenquan, LI Zixu, ZHU Xianxiang, et al. (2026) "Predicting the risk of water-sand inrushes from coal seam roofs based on a small-size sample dataset," Coal Geology & Exploration: Vol. 54: Iss. 3, Article 12.
DOI: 10.12363/issn.1001-1986.25.09.0713
Available at: https://cge.researchcommons.org/journal/vol54/iss3/12

Reference

[1] 张贵彬,王荣强,马俊鹏,等. 浅埋薄基岩顶板采动突水溃砂固流耦合相似模拟试验研究[J]. 煤炭科学技术,2024,52(6):165−175

ZHANG Guibin,WANG Rongqiang,MA Junpeng,et al. Study on solid–fluid coupling similarity simulation test of water–sand inrush during mining of shallow buried thin bedrock roof[J]. Coal Science and Technology,2024,52(6):165−175

[2] 隋旺华，费芳草. 松散含水层下采煤水砂突涌防治研究现状与展望[C]//第二届全国岩土与工程学术大会论文集. 北京：科学出版社，2006：300–304.

[3] 隋旺华. 基于结构水文地质学的采掘诱发高势能突水溃砂主动防控[J]. 工程地质学报,2022,30(1):101−109

SUI Wanghua. Active prevention and control of water–sand mixture inrush with high potential energy due to mining based on structural hydrogeology[J]. Journal of Engineering Geology,2022,30(1):101−109

[4] 隋旺华,梁艳坤,张改玲,等. 采掘中突水溃砂机理研究现状及展望[J]. 煤炭科学技术,2011,39(11):5−9

SUI Wanghua,LIANG Yankun,ZHANG Gailing,et al. Study status and outlook of risk evaluation on water inrush and sand inrush mechanism of excavation and mining[J]. Coal Science and Technology,2011,39(11):5−9

[5] 孙文斌，戴宪政，马诚. 新近系松散层薄基岩下采动突水溃砂机制试验研究[J/OL]. 中国矿业，2025：1–8 (2025-11-25) [2025-06-25]. https://link.cnki.net/urlid/11.3033.TD.20250625.1522.006.

SUN Wenbin，DAI Xianzheng，MA Cheng. Experimental study on the mechanisms of water–sand inrushes induced by mining under Neogene unconsolidated strata[J/OL]. China Mining Magazine，2025：1–8 (2025-11-25) [2025-06-25]. https://link.cnki.net/urlid/11.3033.TD.20250625.1522.006.

[6] 李波,张文泉,马兰. 厚松散层薄基岩条件下矿井顶板涌水致灾因素分析及预测研究[J]. 山东科技大学学报(自然科学版),2017,36(6):39−46

LI Bo,ZHANG Wenquan,MA Lan. Influencing factors and prediction of mine water inrush disaster under thick unconsolidated layers and thin bedrock[J]. Journal of Shandong University of Science and Technology (Natural Science),2017,36(6):39−46

[7] 王忠昶，唐小航，张文泉，等. 基于SSA–RF模型的导水裂隙带发育高度的智能预测[J/OL]. 煤炭科学技术，2025：1–12 (2025-11-25) [2025-02-25]. https://link.cnki.net/urlid/11.2402.td.20250225.1226.004.

WANG Zhongchang，TANG Xiaohang，ZHANG Wenquan，et al. Intelligent prediction of the development height of the water–conducting fracture zone based on the SSA–RF model[J/OL]. Coal Science and Technology，2025：1–12 (2025-11-25) [2025-02-25]. https://link.cnki.net/urlid/11.2402.td.20250225.1226.004.

[8] 武强,黄晓玲,董东林,等. 评价煤层顶板涌(突)水条件的“三图–双预测法”[J]. 煤炭学报,2000,25(1):60−65

WU Qiang,HUANG Xiaoling,DONG Donglin,et al. “Three maps–two predictions” method to evaluate water bursting conditions on roof coal[J]. Journal of China Coal Society,2000,25(1):60−65

[9] 武强,许珂,张维. 再论煤层顶板涌(突)水危险性预测评价的“三图–双预测法”[J]. 煤炭学报,2016,41(6):1341−1347

WU Qiang,XU Ke,ZHANG Wei. Further research on “three maps–two predictions” method for prediction on coal seam roof water bursting risk[J]. Journal of China Coal Society,2016,41(6):1341−1347

[10] 庞凯,武强,曾一凡. 含水层与多烧变岩含水体互层下煤层顶板突(涌)水风险性预测[J]. 煤炭工程,2022,54(12):135−141

PANG Kai,WU Qiang,ZENG Yifan. Prediction of roof water inrush risk under the complex condition of interbedded aquifer and multi burnt rock aquifers[J]. Coal Engineering,2022,54(12):135−141

[11] 秋兴国,李靖. 基于IWOA–SVM的煤矿突水预测模型[J]. 工矿自动化,2022,48(1):71−77

QIU Xingguo,LI Jing. Prediction model of water inrush in coal mine based on IWOA–SVM[J]. Industry and Mine Automation,2022,48(1):71−77

[12] 尹会永,周鑫龙,郎宁,等. 基于SSA优化的GA–BP神经网络煤层底板突水预测模型与应用[J]. 煤田地质与勘探,2021,49(6):175−185

YIN Huiyong,ZHOU Xinlong,LANG Ning,et al. Prediction model of water inrush from coal floor based on GA–BP neural network optimized by SSA and its application[J]. Coal Geology & Exploration,2021,49(6):175−185

[13] 高卫东,王正帅. 基于粒子群优化支持向量机的煤层底板突水量等级预测[J]. 煤田地质与勘探,2012,40(6):44−47

GAO Weidong,WANG Zhengshuai. Forecast of inrushed water volume grade from coal floor based on support vector machine with particle swarm optimization[J]. Coal Geology & Exploration,2012,40(6):44−47

[14] 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

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

[15] 赵凯琳,靳小龙,王元卓. 小样本学习研究综述[J]. 软件学报,2021,32(2):349−369

ZHAO Kailin,JIN Xiaolong,WANG Yuanzhuo. Survey on few–shot learning[J]. Journal of Software,2021,32(2):349−369

[16] 陈良臣,傅德印. 面向小样本数据的机器学习方法研究综述[J]. 计算机工程,2022,48(11):1−13

CHEN Liangchen,FU Deyin. Survey on machine learning methods for small sample data[J]. Computer Engineering,2022,48(11):1−13

[17] 王文,何剑锋,朱文松,等. 小样本下基于改进ACGAN数据增强的X射线矿石图像分类方法[J]. 有色金属工程,2024,14(3):122−132

WANG Wen,HE Jianfeng,ZHU Wensong,et al. X–ray ore image classification method based on improved ACGAN data augmentation in small samples[J]. Nonferrous Metals Engineering,2024,14(3):122−132

[18] 刘奇,梁智昊,訾建潇. SMOGN过采样下导水裂隙带高度的MPSO–BP预测模型[J]. 煤田地质与勘探,2024,52(11):72−85

LIU Qi,LIANG Zhihao,ZI Jianxiao. A SMOGN–based MPSO–BP model to predict the height of a hydraulically conductive fracture zone[J]. Coal Geology & Exploration,2024,52(11):72−85

[19] 李晨曦,鲁海峰. 基于小样本数据机器学习的煤层底板突水预测[J]. 煤矿安全,2025,56(1):171−179

LI Chenxi,LU Haifeng. Prediction of water inrush from coal seam floor based on machine learning with small sample data[J]. Safety in Coal Mines,2025,56(1):171−179

[20] GOODFELLOW I J，POUGET–ABADIE J，MIRZA M，et al. Generative adversarial nets[C]//The 28^th International Conference on Neural Information Processing Systems. USA Cambridge：MIT Press，2014：2672–2680.

[21] 王坤峰,苟超,段艳杰,等. 生成式对抗网络GAN的研究进展与展望[J]. 自动化学报,2017,43(3):321−332

WANG Kunfeng,GOU Chao,DUAN Yanjie,et al. Generative adversarial networks:The state of the art and beyond[J]. Acta Automatica Sinica,2017,43(3):321−332

[22] XU Lei，VEERAMACHANENI K. Synthesizing tabular data using generative adversarial networks[J]. Computer Science，2018，1811：11264.

[23] GULRAJANI I，AHMED F，ARJOVSKY M，et al. Improved training of wasserstein GANs[C]//The 31^st International Conference on Neural Information Processing Systems (NIPS 2017). Long Beach：NIPS，2017：5769–5779.

[24] 韩晟,韩坚舟,赵璇,等. 距离权重改进的Pearson相关系数及应用[J]. 石油地球物理勘探,2019,54(6):1363−1370

HAN Sheng,HAN Jianzhou,ZHAO Xuan,et al. A Pearson correlation coefficient improved by spatial weight[J]. Oil Geophysical Prospecting,2019,54(6):1363−1370

[25] SANTURKAR S，TSIPRAS D，ILYAS A，et al. How does batch normalization help optimization?[C]//The 32^nd International Conference on Neural Information Processing Systems. Montréal：Curran Associates Inc. ，2018：2488–2498.

[26] HUIJBEN I A M,KOOL W,PAULUS M B,et al. A review of the Gumbel–max trick and its extensions for discrete stochasticity in machine learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(2):1353−1371.

[27] CHENG Gaofeng，PEDDINTI V，POVEY D，et al. An exploration of Dropout with LSTMs[C]//Interspeech 2017. Stockholm：ISCA，2017：1586–1590.

[28] BREIMAN L. Random forests[J]. Machine Learning,2001,45(1):5−32.

[29] 田怡帆，梁皓，陈亮，等. 基于机器学习的崩滑灾害易发性评价模型研究：以白龙江流域为例[J/OL]. 工程地质学报，2025：1–10(2025-11-25) [2025-07-21]. https://link.cnki.net/urlid/11.3249.P.20250721.1036.004.

TIAN Yifan，LIANG Hao，CHEN Liang，et al. Research on susceptibility assessment of landslide and collapse disasters based on machine learning models：A case study of Bailong River Basin[J/OL]. Journal of Engineering Geology，2025：1–10 (2025-11-25) [2025-07-21]. https://link.cnki.net/urlid/11.3249.P.20250721.1036.004.

[30] 陈晓青,陆慧娟,关伟,等. 基于探测粒子群的小波核极限学习机算法[J]. 计算机科学,2016,43(6A):77−80

CHEN Xiaoqing,LU Huijuan,GUAN Wei,et al. Wavelet kernel extreme learning machine algorithm based on detecting particle swarm optimization[J]. Computer Science,2016,43(6A):77−80

[31] 鞠红梅，周菲，孟凡. 基于模糊隶属度与粒子群算法协同优化的双子支持向量机鲁棒性分类方法[J/OL]. 系统科学与数学，2025：1–22(2025-11-20)[2025-09-17]. https://link.cnki.net/urlid/11.2019.o1.20250916.1042.002.

JU Hongmei，ZHOU Fei，MENG Fan. Twin support vector machine based on fuzzy membership and particle swarm algorithm collaborative optimization for robust classification[J/OL]. Journal of Systems Science and Mathematical Sciences，2025：1–22 (2025-11-20) [2025-09-17]. https://link.cnki.net/urlid/11.2019.o1.20250916.1042.002.

[32] 张英男. 改进的粒子群优化算法(APSO和DPSO)研究[D]. 大连：大连理工大学，2008.

ZHANG Yingnan. The improved particle swarm optimization algorithms：APSO and DPSO[D]. Dalian：Dalian University of Technology，2008.

[33] 郝显福,贺严,孙嘉,等. 基于可参考经验公式的防水矿柱留设研究[J]. 黄金,2016,37(3):37−40

HAO Xianfu,HE Yan,SUN Jia,et al. Research on waterproof pillars setting based on referable empirical formula[J]. Gold,2016,37(3):37−40

[34] 武宇亮,闫增会,谢春雷,等. 薄基岩区特厚煤层不同开采工艺下防砂安全煤岩柱留设研究[J]. 煤炭技术,2024,43(12):44−48

WU Yuliang,YAN Zenghui,XIE Chunlei,et al. Study on safety coal and rock pillar for sand control under different mining technology of extra thick coal seam in thin bedrock area[J]. Coal Technology,2024,43(12):44−48

[35] 吕玉广,李宏杰,夏宇君,等. 基于多类型四双法的煤层顶板突水预测评价研究[J]. 煤炭科学技术,2019,47(9):219−228

LYU Yuguang,LI Hongjie,XIA Yujun,et al. Prediction and evaluation study on coal seam roof water inrush based on multi–type four–double method[J]. Coal Science and Technology,2019,47(9):219−228