(1. College of Energy Engineering，Xi?an University of Science and Technology，Xi?an，Shaanxi 710054，China；2. Key Laboratory of Western Mines and Hazard Prevention of China Ministry of Education，Xi?an University of Science and Technology，Xi'an，Shaanxi 710054，China；3. Key Laboratory of Coal Resources Exploration and Comprehensive Utilization，Ministry of Natural Resources，Xi'an，Shaanxi 710021，China；4. Xinjiang Energy Co.，Ltd.，State Energy Group，Urumqi，Xinjiang 830027，China)
Abstract：In the horizontal sublevel top coal caving of steeply inclined extra-thick coal seam groups，the segmented mining is prone to induce roof collapses with time delayed characteristics. The understanding of the collapse height and characteristics of the overlying strata is crucial for monitoring roof disasters，crack development，and surface collapses. To address this，a combination of similar simulation experiments，machine learning，and grey theory methods were utilized to explore the collapse characteristics of overlying strata and the laws governing surface collapses. Data indicators associated with collapse height were obtained，and the dimensionality of input data was reduced using Principal Component Analysis(PCA). A collapse height prediction model(PCA-GA-LSSVR) based on the least squares support vector regression(LSSVR) network was established through the optimization of training using genetic algorithms(GA) and leave-one-out cross-validation. Various combinations of mining parameters and monitoring indicators were employed for training and testing，facilitating the evaluation and selection of the optimal prediction model. The selected model was further validated through engineering inspections. Results indicate that with increasing mining depth in the experimental mining of steeply inclined extra-thick coal seam groups，continuous collapse of rock strata generates and expands V-shaped subsidence. There are high correlations between mining depth，single caving height，model section height，and collapse height. Additionally，the collapse of the roof in experimental mining exhibits a time delayed characteristic. Dimensionality reduction of data effectively addresses the issue of high dimensionality and information overlap，enhancing the accuracy of prediction models. The PCA-GA-LSSVR model demonstrates superior accuracy in predicting test data，with lower error rates，average absolute errors，and average absolute percentage errors(5.146，4.819，and 0.087 respectively) compared to other models. The established grey time delayed OBGM model exhibits good fitting and predictive capabilities for roof collapse height in the #45 coal seam scenario. During engineering inspection，the maximum error fluctuation range of the established collapse height prediction model is within 3.36 m. In scenarios with limited sample data availability，gradually increasing learning samples reduces model prediction errors. This research provides valuable insights for studying collapse height in steeply inclined extra-thick coal seam groups during horizontal sublevel top coal caving mining.
崔 峰1，2，3，何仕凤1，2，来兴平1，2，刘旭东4，蒋新军4，孙秉成4，贾 冲1，2，宗 程1，2，李宜霏1，2. 基于LSSVR与灰色理论的急倾斜巨厚煤层群开采冒落高度与时滞特征研究[J]. 岩石力学与工程学报, 2024, 43(4): 822-837.
CUI Feng1，2，3，HE Shifeng1，2，LAI Xingping1，2，LIU Xudong4，JIANG Xinjun4，SUN Bingcheng4，. Study on collapse height and time delayed characteristics in the mining of steeply inclined extra-thick coal seam group based on LSSVR and grey theory. , 2024, 43(4): 822-837.
[1] 姚 琦，冯 涛，廖 泽. 急倾斜走向分段充填倾向覆岩破坏特性及移动规律[J]. 煤炭学报，2017，42(12)：3 096–3 105.(YAO Qi，FENG Tao，LIAO Ze. Damage characteristics and movement of inclined strata with sublevel filling along the strike in the steep seam[J]. Journal of China Coal Society，2017，42(12)：3 096–3 105.(in Chinese))
[2] 崔 峰，张廷辉，来兴平，等. 冲击地压矿井不同采动强度下的开采扰动特征及其产能[J]. 煤炭学报，2021，46(12)：3 781–3 793. (CUI Feng，ZHANG Tinghui，LAI Xingping，et al. Mining disturbance characteristics and productivity of rock burst mines under different mining intensities[J]. Journal of China Coal Society，2021，46(12)：3 781–3 793.(in Chinese))
[3] 崔 峰，贾 冲，来兴平，等. 基于加卸载响应比的冲击地压矿井急倾斜巨厚煤层推进速度研究[J]. 煤炭学报，2022，47(2)：745–761.(CUI Feng，JIA Chong，LAI Xingping，et al. Study on advancing rate of steeply inclined extra-thick coal seam in rock burst mine based on loading-unloading response ratio[J]. Journal of China Coal Society，2022，47(2)：745–761.(in Chinese))
[4] 王家臣，杨胜利，李良晖. 急倾斜煤层水平分段综放顶板“倾倒–滑塌”破坏模式[J]. 中国矿业大学学报，2018，47(6)：1 175–1 184. (WANG Jiachen，YANG Shengli，LI Lianghui. Toppling-slumping failure mode in horizontal sublevel top-coal caving face in steeply-inclined seam[J]. Journal of China University of Mining and Technology，2018，47(6)：1 175–1 184.(in Chinese))
[5] 来兴平，代晶晶，李 超. 急倾斜煤层开采覆岩联动致灾特征分析[J]. 煤炭学报，2020，45(1)：122–130.(LAI Xingping，DAI Jingjing，LI Chao. Analysis on hazard characteristics of overburden structure in steeply inclined coal seam[J]. Journal of China Coal Society，2020，45(1)：122–130.(in Chinese))
[6] 鞠文君，郑建伟，魏 东，等. 急倾斜巨厚煤层多分层同采巷道冲击地压成因及控制技术研究[J]. 采矿与安全工程学报，2019，36(2)：280–289.(JU Wenjun，ZHENG Jianwei，WEI Dong，et al. Study on the causes and control technology about the coal bump in multi-layered mining roadway in steep-thick coal seams[J]. Journal of Mining and Safety Engineering，2019，36(2)：280–289.(in Chinese))
[7] 程卫民，孙路路，王 刚，等. 急倾斜巨厚煤层开采相似材料模拟试验研究[J]. 采矿与安全工程学报，2016，33(3)：387–392.(CHENG Weimin，SUN Lulu，WANG Gang，et al. Similar material simulation test of steep-inclined extra-thick coal seam[J]. Journal of Mining and Safety Engineering，2016，33(3)：387–392.(in Chinese))
[8] 孙 闯，陈东旭，程耀辉，等. 急倾斜煤层坚硬顶板塌落规律及控制研究[J]. 岩石力学与工程学报，2019，38(8)：1 647–1 658.(SUN Chuang，CHEN Dongxu，CHENG Yaohui，et al. Study on collapse rule and control of hard roofs in steeply inclined coal seams[J]. Chinese Journal of Rock Mechanics and Engineering，2019，38(8)：1 647–1 658. (in Chinese))
[9] 戴华阳，郭俊廷，易四海，等. 特厚急倾斜煤层水平分层开采岩层及地表移动机制[J]. 煤炭学报，2013，38(7)：1 109–1 115.(DAI Huayang，GUO Junting，YI Sihai，et al. The mechanism of strata and surface movements induced by extra-thick steeply inclined coal seam applied horizontal slice mining[J]. Journal of China Coal Society，2013，38(7)：1 109–1 115.(in Chinese))
[10] 杨胜利，张 鹏，李福胜，等. 急倾斜厚煤层水平分层综放工作面支架载荷确定[J]. 煤炭科学技术，2010，38(11)：37–40.(YANG Shengli，ZHANG Peng，LI Fusheng，et al. Load determination of owered supports for feully mechanized top coal caving mining face in horizontal slice of steep inclined thick seam[J]. Coal Science and Technology，2010，38(11)：37–40.(in Chinese))
[11] 付建新，朱鹏瑞，宋卫东，等. 基于时滞非线性MGM模型的崩落法开采冒落高度预测研究[J]. 采矿与安全工程学报，2020，37(4)：741–749.(FU Jianxin，ZHU Pengrui，SONG Weidong，et al. Caving height prediction of caving method based on time-delay nonlinear MGM model[J]. Journal of Mining and Safety Engineering，2020，37(4)：741–749.(in Chinese))
[12] 张 军，王建鹏，杨文光. 综采工作面冒落高度模糊综合预测模型研究[J]. 中国矿业大学学报，2014，43(3)：426–431.(ZHANG Jun，WANG Jianpeng，YANG Wenguang. Research on the fuzzy comprehensive predicting model for caved zone height of fully mechanized working face[J]. Journal of China University of Mining and Technology，2014，43(3)：426–431.(in Chinese))
[13] 陈尚波，张耀平，夏弋江，等. 基于残差型G-LSSVM的岩石冒落高度及碴石厚度预测[J]. 金属矿山，2013，(8)：8–10.(CHEN Shangbo，ZHANG Yaoping，XIA Yijiang，et al. Prediction of rock caving height and ballast stone thickness based on the residuals G-LSSVM[J]. Metal Mine，2013，(8)：8–10.(in Chinese))
[14] 赵毅鑫，杨志良，马斌杰，等. 基于深度学习的大采高工作面矿压预测分析及模型泛化[J]. 煤炭学报，2020，45(1)：54–65.(ZHAO Yixin，YANG Zhiliang，MA Binjie，et al. Deep learning prediction and model generalization of ground pressure for deep longwall face with large mining height[J]. Journal of China Coal Society，2020，45(1)：54–65.(in Chinese))
[15] 薛建坤，王 皓，赵春虎，等. 鄂尔多斯盆地侏罗系煤田导水裂隙带高度预测及顶板充水模式[J]. 采矿与安全工程学报，2020，37(6)：1 222–1 230.(XUE Jiankun，WANG Hao，ZHAO Chunhu，et al. Prediction of the height of water-conducting fracture zone and water-filling model of roof aquifer in Jurassic coalfield in Ordos Basin[J]. Journal of Mining and Safety Engineering，2020，37(6)：1 222– 1 230.(in Chinese))
[16] 崔 峰，何仕凤，来兴平，等. 基于相空间重构与深度学习的冲击地压矿井时间序列b值趋势[J]. 煤炭学报，2023，48(5)：2 022–2 034. (CUI Feng，HE Shifeng，LAI Xingping，et al. Trend of time sequence b value of rock burst mine based on phase space reconstruction and deep learning[J]. Journal of China Coal Society，2023，48(5)：2 022–2 034. (in Chinese))
[17] 马 砺，张鹏宇，郭睿智，等. 巷道火灾密闭过程烟气温度预测的GA-SVM模型[J]. 中国矿业大学学报，2021，50(4)：641–648.(MA Li，ZHANG Pengyu，GUO Ruizhi，et al. GA-SVM model for prediction flue gas temperature of roadway fire under sealing process[J]. Journal of China University of Mining and Technology，2021，50(4)：641–648.(in Chinese))
[18] SUYKENS J A K，VANDEWALLE J. Least squares support vector machine classifiers[J]. Neural Processing Letters，1999，9(3)：293–300.
[19] 谷远利，张 源，芮小平，等. 基于免疫算法优化LSSVM的短时交通流预测[J]. 吉林大学学报：工学版，2019，49(6)：1 852–1 857. (GU Yuanli，ZHANG Yuan，RUI Xiaoping，et al. Short-term traffic flow prediction based on LSSVM optimized by immune algorithm[J]. Journal of Jilin University：Engineering and Technology，2019，49(6)：1 852–1 857.(in Chinese))
[20] 杨建新，兰小平，冯亚东，等. 基于改进樽海鞘群和最小二乘支持向量机算法的新型弹药质量评估方法[J]. 兵工学报，2022，43(5)：1 012–1 022.(YANG Jianxin，LAN Xiaoping，FENG Yadong，et al. An ammunition quality evaluation method based on least squares support vector machine[J]. Acta Armamentarii，2022，43(5)：1 012–1 022.(in Chinese))
[21] HUANG P H，YANG Z Y，WANG X Y，et al. Research on Piper-PCA-Bayes-LOOCV discrimination model of water inrush source in mines[J]. Arabian Journal of Geosciences，2019，12(11)：1–14.
[22] MIRI M，GHASSOUN Y，DOVLATABADI A，et al. Estimate annual and scasonal PM1，PM2.5 and PM10 concentrations using land use regression model[J]. Ecotoxicology and Environmental Safety，2019，174：137–145.
[23] 张施令，姚 强. 联合NDIR技术与灰色系统OBGM(1，N)模型的高压组合电器SF_6分解产物时间序列测量与预测[J]. 电网技术，2020，44(7)：2 770–2 777.(ZHANG Shiling，YAO Qiang. Time series measurement and prediction of SF_6 decomposition products of high voltage switch-gear combined NDIR technology and grey system OBGM(1，N) model[J]. Power System Technology，2020，44(7)： 2 770–2 777.(in Chinese))
[24] JAHIRUL M I，RASUL M G，BROWN R J，et al. Investigation of correlation between chemical composition and properties of biodiesel using principal component analysis(PCA) and artificial neural network(ANN)[J]. Renewable energy，2021，168：632–646.
[25] 古 创，姚军强，吴志跃，等. 基于主成分分析优化BP神经网络模型的厌氧膜生物反应器膜污染预测[J]. 新能源进展，2022，10(2)：95–102.(GU Chuang，YAO Junqiang，WU Zhiyue，et al. Membrane fouling prediction of anaerobic membrane bioreactor based on BP neural network model optimized by principal component analysis[J]. Advances in New and Renewable Energy，2022，10(2)：95–102.(in Chinese))
[26] HUO H B，YI J，ZHU X J，et al. Control-oriented dynamic identification modeling of a planar SOFC stack based on genetic algorithm-least squares support vector regression[J]. Journal of Zhejiang University-Science A：Applied Physics and Engineering，2014，15(10)：829–839.
[27] 成 珂，孙琦琦，马晓瑶. 基于主成分回归分析的气象因子对光伏发电量的影响[J]. 太阳能学报，2021，42(2)：403–409.(CHENG Ke，SUN Qiqi，MA Xiaoyao. Influence of meteorological factors on photovoltaic power generation based on principal component regression analysis[J]. Acta Energiae Solaris Sinica，2021，42(2)：403–409.(in Chinese))
[28] 周小麟，童晓阳. 基于CEEMD-SBO-LSSVR的超短期风电功率组合预测[J]. 电网技术，2021，45(3)：855–864.(ZHOU Xiaolin，TONG Xiaoyang. Ultra-short-term wind power combined prediction based on CEEMD-SBO-LSSVR[J]. Power System Technology，2021，45(3)：855–864.(in Chinese))
[29] KIM W，PARK J，YOO J，et al. Target localization using ensemble support vector regression in wireless sensor networks[J]. IEEE Transactions on Cybernetics，2013，43(4)：1 189–1 198.
[30] 陈小波，刘 祥，韦中杰，等. 基于GA-LSSVR模型的路网短时交通流预测研究[J]. 交通运输系统工程与信息，2017，17(1)：60–66.(CHEN Xiaobo，LIU Xiang，WEI Zhongjie，et al. Short-term traffic flow forecasting of road network based on GA-LSSVR model[J]. Journal of transportation systems engineering and information technology，2017，17(1)：60–66.(in Chinese))
[31] TAKENS F. Detecting strange attractors in turbulence[M]. Heidelberg：Springer Berlin Heidelberg，1981，898：366–381.
[32] 王启阳，吴文福，兰天忆. 东北地区稻谷储藏期间脂肪酸含量的预测模型[J]. 农业工程学报，2020，36(6)：269–275.(WANG Qiyang，WU Wenfu，LAN Tianyi. Models for predicting the fatty acid contents of rice during storage in the northeast China[J]. Transactions of the Chinese Society of Agricultural Engineering，2020，36(6)：269–275.(in Chinese))
[33] 陈 勇，李 鹏，张忠军，等. 基于PCA-GA-LSSVM的输电线路覆冰负荷在线预测模型[J]. 电力系统保护与控制，2019，47(10)：110–119.(CHEN Yong，LI Peng，ZHANG Zhongjun，et al. Online prediction model for power transmission line icing load based on PCA-GA-LSSVM[J]. Power System Protection And Control，2019，47(10)：110–119.(in Chinese))
[34] 杜子学，文孝霞，杨 震，等. 跨座式单轨车辆走行轮偏磨损因子模型研究[J]. 西南交通大学学报，2019，54(2)：373–380.(DU Zixue，WEN Xiaoxia，YANG Zhen，et al. Study on partial-wear-factor model monorail vehicle running wheel[J]. Journal of Southwest Jiaotong University，2019，54(2)：373–380.(in Chinese))
[35] ZENG B，LI C. Improved multi-variable grey forecasting model with a dynamic background-value coefficient and its application[J]. Computers and Industrial Engineering，2018，(118)：278–290.
[36] 曾 欣，刘 军，柳福祥，等. 基于灰色时滞OBGM-BP神经网络模型的滑坡变形预测[J]. 数学的实践与认识，2022，52(5)：112–122.(ZENG Xin，LIU Jun，LIU Fuxiang，et al. Prediction of landslide deformation based on grey time-delay OBGM-BP neural network model[J]. Mathematics in Practice and Theory，2022，52(5)：112–122.(in Chinese))
2024, Vol. 43 
