自我国碳达峰、碳中和的开始实施，建设绿色、智能化矿山，强化矿井瓦斯抽采效果，减少矿井瓦斯抽采过程中的碳排放是落实我国碳达峰与碳中和愿景的具体行动。如何进行顺层钻孔瓦斯抽采调控是实现强化瓦斯抽采效果的重要手段，对减少碳排放具有着重要的研究意义。本文通过理论分析、物理试验、模型构建及工程验证等方法，分析了不同抽采负压下顺层钻孔瓦斯抽采特征，构建了不同负压下顺层钻孔瓦斯抽采参数预测模型和校正反馈调控模型，提出了钻孔瓦斯抽采自适应调控方案。论文的主要研究工作如下：

分析了抽采负压对煤体瓦斯赋存状态及透气性的影响，得出抽采负压影响煤体孔隙结构的改变使得大量瓦斯从吸附态转变为游离态，同时导致煤体透气性发生变化的结论；由于抽采负压的作用，顺层瓦斯抽采钻孔周围煤体裂隙增大形成漏气通道从而造成钻孔漏气；通过分析漏气类型和特征指出当前井下顺层瓦斯抽采钻孔的三种漏气类型，其中主要为具有“环形漏气圈”的漏气类型并给出了相应的漏气量公式，揭示了负压同抽采钻孔漏气间的理论关系；通过分析顺层钻孔孔周煤体瓦斯径向流动规律，结合达西定律得出了顺层钻孔瓦斯在不同煤体不同状态下的流量公式。

自主构建了顺层钻孔瓦斯抽采单孔负压调控试验系统，开展不同抽采负压下单孔瓦斯浓度、纯流量和漏气量变化特征试验，探究不同抽采负压下单孔瓦斯抽采参数变化规律，结果表明：在抽采初期封孔良好的情况下抽采负压越大，瓦斯浓度、纯流量越高，负压越大，纯流量的增长率越高于漏气量的增长；抽采后期，抽采负压更多作用于钻孔漏气，抽采负压越大，纯流量衰减速率越大，漏气量越大，瓦斯浓度变化速率越快，最终瓦斯浓度越低。

以不同负压下抽采参数时间序列变化规律为依据，构建了基于ARIMA-RBFNN组合预测模型并预测了抽采参数的未来变化，给出了抽采参数最优控制区间并通过MSE、MAE、MAPE等检验模型误差，结果表明：组合模型预测的抽采参数更为接近实际值，瓦斯浓度R²为0.88、瓦斯流量R²为0.96、漏气量R²为0.72，均大于0.7，MSE、MAE、MAPE等检验误差满足期望，模型性能良好，预测值可作为调控目标。

在得到调控目标值和最优控制区间的基础上，结合瓦斯抽采安全和效率约束准则提出了顺层钻孔瓦斯抽采负压调控自适应方案，结合调控原理确定了三大调控任务，构建了校正反馈调控模型，揭示钻孔瓦斯抽采参数校正反馈调控机制；通过瓦斯抽采双孔调控试验验证了自适应方案的可行性。

选取试验工作面顺层瓦斯抽采钻孔进行抽采参数分析，并给出负压调控建议，根据调控前后抽采参数对比确定最优抽采负压并针对问题钻孔提出优化措施。

基于以上的研究分析，形成了一套完整的负压驱动下的顺层钻孔瓦斯抽采参数预测反馈调控方法，可为煤矿瓦斯抽采过程调控提供一定的参考建议。

Since the beginning of China's carbon peaking and carbon neutral implementation, building green and intelligent mines, strengthening the effect of mine gas extraction, and reducing carbon emissions during mine gas extraction are concrete actions to implement China's vision of carbon peaking and carbon neutrality. How to regulate gas extraction in down-level boreholes is an essential means to achieve the effect of enhanced gas extraction and has essential research significance to reduce carbon emissions. In this paper, through theoretical analysis, physical tests, model construction, and engineering verification, we analyze the gas extraction characteristics of cascade boreholes under different extraction negative pressures, construct a prediction model of gas extraction parameters and a corrective feedback regulation model of cascade boreholes under different negative pressures, and propose an adaptive regulation scheme for gas extraction from boreholes. The main research works of the paper are as follows:

Analyze the effect of negative extraction pressure on the gas storage state and permeability of the coal body, and conclude that the negative extraction pressure affects the change of pore structure of the coal body, which causes a large amount of gas to change from adsorbed state to free state, and at the same time leads to the change of permeability of the coal body; due to the effect of negative extraction pressure, the fissures of the coal body around the cascade gas extraction borehole increase to form air leakage channels, thus causing air leakage from the borehole; by analyzing the types and characteristics of air leakage By analyzing the types and characteristics of air leakage, we pointed out three types of air leakage in current downhole layered gas extraction boreholes, among which the primary type of air leakage is that with "ring-shaped leakage circle" and gave the corresponding air leakage formula, revealing the theoretical relationship between negative pressure and air leakage in extraction boreholes; by analyzing the radial flow law of coal gas around cascade boreholes and combining with Darcy's law, we came to the conclusion that cascade By analyzing the radial flow law of gas around the perimeter of the drill hole in the cascade layer, the flow equation of gas in different coal bodies under different states was derived by combining Darcy's law.

Independent construction of a single-hole negative pressure control test system for gas extraction in cis-boreholes, carrying out experiments on the characteristics of gas concentration, pure flow rate, and air leakage in single-hole under different extraction negative pressure, exploring the law of change of single-hole gas extraction parameters under different extraction negative pressure, the results show that: the higher the extraction negative pressure at the early stage of extraction with good sealing, the higher the gas concentration and pure flow rate, the higher the negative pressure, the higher the growth rate of pure flow rate than the growth rate of air leakage; the higher the negative pressure, the higher the growth rate of pure flow rate than the growth rate of air leakage; the higher the extraction pressure, the higher the growth rate of gas concentration and pure flow rate. The larger the negative pressure, the higher the decay rate of pure flow, the larger the air leakage, the faster the change rate of gas concentration, and the lower the final gas concentration.

(3) Based on the time series variation law of extraction parameters under different negative pressures, a combined ARIMA-RBFNN based prediction model was constructed and predicted the future changes of extraction parameters, and the optimal control interval of extraction parameters was given; MSE, MAE, MAPE, etc. tested the model errors. The results showed that the extraction parameters predicted by the combined model were closer to the actual values, with gas concentration R² 0.88, gas flow R² 0.96, and air leakage R² 0.72, all greater than 0.7. The MSE, MAE, and MAPE test errors meet the expectations, the model performance is good, and the predicted values can be used as the control target.

(4) Based on the control target value and the optimal control interval, the adaptive scheme of negative pressure control of gas extraction in cis-borehole is proposed by combining the safety and efficiency constraints of gas extraction, and the three primary control tasks are determined by combining the control principles, and the corrective feedback control model is constructed to reveal the corrective feedback control mechanism of gas extraction parameters in the borehole; the gas extraction double-hole control test verifies the feasibility of the adaptive scheme. The feasibility of the adaptive scheme was verified through gas extraction dual-hole control tests.

(5) Selected gas extraction boreholes on the working face for analysis of extraction parameters and gave recommendations on negative pressure regulation, determined the optimal extraction negative pressure according to comparing extraction parameters before and after regulation, and proposed optimization measures for problematic boreholes.

Based on the above research and analysis, a complete set of predictive feedback regulation methods for gas extraction parameters of cascade boreholes driven by negative pressure is formed, which can provide some reference suggestions for regulating the coal mine gas extraction process.

[1]谢和平,任世华,谢亚辰,等. 碳中和目标下煤炭行业发展机遇[J]. 煤炭学报, 2021, 46(7): 2197-2211.

[2]朱超,史志斌,鲁金涛. 碳达峰、碳中和对我国煤炭工业发展的影响及对策[J]. 煤炭经济研究, 2021, 41(4): 59-64.

[3]袁亮. 煤矿典型动力灾害风险判识及监控预警技术研究进展[J]. 煤炭学报, 2020, 45(5): 1557-1566.

[4]贺雁鹏,黄庆享,曹健. 可视化矿山的发展现状及关键技术[J]. 煤炭科学技术, 2019, 47(4): 32-37.

[5]张长鲁. 基于数据挖掘的煤矿安全可视化管理研究[D]. 中国矿业大学(北京), 2015.

[6]程远平,董骏,李伟,等. 负压对瓦斯抽采的作用机制及在瓦斯资源化利用中的应用[J].煤炭学报, 2017, 42(6): 1466-1474.

[7]王凯. 顺层瓦斯抽采钻孔孔内负压分布规律及应用研究[D]. 煤炭科学研究总院, 2014.

[8]巴全斌. 瓦斯抽采钻孔漏气通道检测装置研制及应用[J]. 工矿自动化, 2021, 47(1):9-14.

[9]Liu T,Lin B,Fu X, et al. Modeling Air Leakage Around Gas Extraction Boreholes in Mining-disturbed Coal Seams[J]. Process Safety and Environmental Protection, 2020,141: 202-214.

[10]刘泽功,袁亮,戴广龙,等. 采场覆岩裂隙特征研究及在瓦斯抽放中应用[J]. 安徽理工大学学报(自然科学版), 2004(4): 10-15.

[11]蔡峰,刘泽功. 深部低透气性煤层上向穿层水力压裂强化增透技术[J]. 煤炭学报, 2016, 41(1): 113-119.

[12]吴立新,汪云甲,丁恩杰,等. 三论数字矿山——借力物联网保障矿山安全与智能采矿[J]. 煤炭学报, 2012, 37(3): 357-365.

[13]张槐森,秦玉金,苏伟伟. 负压对瓦斯抽采效果的影响机制研究[J]. 煤矿安全, 2019,50(8): 23-26, 31.

[14]马金魁. 基于大数据统计分析确定煤层瓦斯抽采合理负压[J]. 能源与环保, 2019, 41(5): 24-27, 32.

[15]李波,孙东辉,张路路. 煤矿顺层钻孔瓦斯抽采合理布孔间距研究[J]. 煤炭科学技术, 2016, 44(8): 121-126, 155.

[16]李润芝,梁冰,孙维吉,等. 顺层钻孔瓦斯抽采半径及布孔间距试验研究[J]. 中国安全科学学报, 2016, 26(10): 133-138.

[17]石占山,梁冰,王岩,等. 加载与卸载过程瓦斯抽采钻孔变形特征[J]. 煤炭学报, 2017, 42(6): 1458-1465.

[18]张天军,庞明坤,蒋兴科,等. 负压对抽采钻孔孔周煤体瓦斯渗流特性的影响[J]. 岩土力学, 2019, 40(7): 2517-2524.

[19]焦荣坤,张学博,栗翌. 钻孔不同变形失稳时抽采负压分布规律研究[J]. 工矿自动化,2019, 45(5): 40-45.

[20]张学博,王文元,沈帅帅. 钻孔变形失稳条件下抽采负压及流量分布规律试验研究[J].煤炭科学技术, 2020, 48(10): 45-51.

[21]刘军. 预抽钻孔负压沿孔长变化特性及对瓦斯抽采效果影响研究[D]. 河南理工大学,2014.

[22]张俊伟. 瓦斯抽采钻孔漏气对抽采效果的影响研究[D]. 河南理工大学, 2018.

[23]Wang X,Zhou F,Xia T, et al. A Multi-objective Optimization Model to Enhance the Comprehensive performance of Underground Gas Drainage System[J]. Journal of Natural Gas Science and Engineering, 2016, 36: 852-864.

[24]张超,杨朴超,李树刚,等. 瓦斯抽采钻孔密封段防塌支护材料试验研究及现场应用[J].采矿与安全工程学报, 2021, 38(1): 199-205.

[25]杨宏民,任发科,王兆丰,等. 瓦斯抽采钻孔封孔质量检测及定量评价方法[J]. 煤炭学报, 2019, 44(S1): 164-170.

[26]杨真,郭昌放,王静宜,等. 由数据驱动的智慧矿山建设研究[J]. 中国煤炭, 2019, 45(11): 41-48.

[27]袁亮. 煤矿典型动力灾害风险判识及监控预警技术研究进展[J]. 煤炭学报, 2020, 45(5): 1557-1566.

[28]潘玉民,邓永红,张全柱. 小波神经网络模型的确定性预测及应用[J]. 计算机应用, 2013, 33(4): 1001-1005.

[29]王小龙,张军. 回转钻进钻孔数据处理与三维显示软件设计[J]. 工矿自动化, 2021,47(6): 84-90.

[30]Zhang C,Tu S,Bai Q, et al. Evaluating Pressure-relief Mining Performances Based on Surface Gas Venthole Extraction Data in Longwall Coal Mines[J]. Journal of Natural Gas Science and Engineering, 2015, 24: 431-440.

[31]路正雄,郭卫,张帆,等. 基于数据驱动的综采装备协同控制系统架构及关键技术[J].煤炭科学技术, 2020, 48(7): 195-205.

[32]Ziyuan tong, MinMing-Tong. Research on Gas Emission Distribution in Coalface Based on Trend Surface Polynomial Analysis[J]. IEEE SENSORS JOURNAL, 2015, 15(1): 164-171.

[33]Jia P,Liu H,Wang S, et al. Research on a Mine Gas Concentration Forecasting Model Based on a Gru Network[J]. IEEE Access, 2020, 8: 38023-38031.

[34]Jin K,Cheng Y,Wang W, et al. Evaluation of the Remote Lower Protective Seam Mining for Coal Mine Gas Control a Typical Case Study From the Zhuxianzhuang Coal Mine, Huaibei Coalfield, China[J]. Journal of Natural Gas Science and Engineering, 2016, 33:44-55.

[35]蒲阳. 高产高效矿井煤与瓦斯突出远近场协同预测技术[J]. 煤矿安全, 2020, 51(7):70-74, 80.

[36]蒲阳,宋志强,宁小亮. 煤与瓦斯突出预测多元数据可视化系统构建与应用[J]. 工矿自动化, 2020,46(7): 64-69.

[37]任文永,刘引军,丰界泽. 基于瓦斯浓度预测的采煤机与刮板输送机协调控制研究[J].煤炭技术, 2021, 40(3): 141-145.

[38]袁春花. 一种基于改进DBSCAN算法的煤矿瓦斯数据挖掘方法[J]. 煤炭技术, 2012, 31(11): 177-179.

[39]王丹丹,付华,徐耀松. 基于DBSCAN算法的煤矿瓦斯监测信息聚类分析方法研究[J]. 工矿自动化, 2010, 36(8): 45-48.

[40]石记斌,石记红. 关联分析数据挖掘在煤矿瓦斯安全监测预警中的应用[J]. 能源与环保, 2017, 39(8): 1-5, 11.

[41]潘红宇,赵云红,张卫东,等. 基于Adaboost的改进BP神经网络地表沉陷预测[J]. 煤炭科学技术, 2019, 47(2): 161-167.

[42]张震,朱权洁,李青松,等. 基于Python的瓦斯浓度ARIMA预测模型构建及其应用[J]. 华北科技学院学报, 2020, 17(2): 23-28, 49.

[43]侯东毅,董晓雷. 基于RS-GA-ELM的矿井瓦斯涌出量预测[J]. 世界科技研究与发展, 2014, 36(6): 638-642.

[44]黄凯峰. 煤矿安全监测监控系统瓦斯浓度异常信号辨识方法研究[D]. 安徽理工大学, 2016.

[45]青梅. 基于回归分析的矿井监控数据挖掘分析[J]. 煤炭技术, 2018, 37(7): 260-262.

[46]王春光. 煤层气智能抽采与联动装置[J]. 煤矿安全, 2015, 46(4): 94-96.

[47]王春光. 瓦斯智能抽采发展探讨[J]. 煤矿安全, 2015, 46(12): 179-183.

[48]赵旭生,马国龙. 煤矿瓦斯智能抽采关键技术研究进展及展望[J]. 煤炭科学技术, 2021, 49(5): 27-34.

[49]周福宝,刘春,夏同强,等. 煤矿瓦斯智能抽采理论与调控策略[J]. 煤炭学报, 2019, 44(8): 2377-2387.

[50]周福宝,王鑫鑫,夏同强. 瓦斯安全抽采及其建模[J]. 煤炭学报, 2014,39(8):1659-1666.

[51]黄瑞峰,张志刚,程波. 井下抽采钻孔瓦斯浓度控制机理及方法研究[J]. 煤炭科学技术, 2017, 45(5): 128-135.

[52]程健维,赵刚,刘雨涛,等. 瓦斯抽采钻孔非凝固材料性能与智能封孔装置研究[J]. 煤炭科学技术, 2020, 48(2): 131-135.

[53]蒋志刚,陶云奇,王峰. 瓦斯抽采浓度自动化调控系统的研究与应用[J]. 煤炭技术, 2014, 33(10): 28-30.

[54]宋爽,张天军,李树刚,等. 采空区卸压瓦斯抽采评判系统设计与实现[J]. 矿业安全与环保, 2019, 46(5): 85-89.

[55]宋爽. 基于深度学习的采空区卸压瓦斯抽采智能评价方法研究[D]. 西安科技大学, 2019.

[56]李延辉,魏引尚,宋国良,等. 基于层次分析法的瓦斯抽采系统可靠性综合评价[J]. 煤矿安全, 2012, 43(6): 141-144.

[57]申健,李雪冰. 瓦斯抽采达标评价AHP-FCE模型及应用[J]. 辽宁工程技术大学学报 (自然科学版), 2015, 34(5): 578-583.

[58]王振锋. 瓦斯抽采浓度控制机理及管路浓度自动调控预警系统研究[D]. 河南理工大学, 2014.

[59]翟成,林柏泉,叶青. 图论理论在矿井瓦斯抽放系统优化中的应用[J]. 采矿与安全工程学报, 2008, 84(1): 119-122.

[60]陈勇. 负压动态调控下矿井瓦斯抽采工况参数变化对比与分析[J]. 矿业安全与环保,2020, 47(5): 108-112.

[61]王维忠,刘东,许江,等. 瓦斯抽采过程中钻孔位置对煤层参数演化影响的试验研究[J]. 煤炭学报, 2016, 41(2): 414-423.

[62]安赛,张辉,李宏艳,等. 自适应瓦斯抽采管路阀门自动调控系统设计与应用[J]. 煤炭工程, 2016, 48(3): 24-26.

[63]李健威,万勇. 基于WinCC的无人值守瓦斯抽采智能控制系统研究[J]. 自动化与仪表,2019, 34(5): 49-52, 94.

[64]龚晓燕,雷可凡,吴群英,等. 数字孪生驱动的掘进工作面出风口风流智能调控系统[J]. 煤炭学报, 2021, 46(4): 1331-1340.

[65]陈月霞.瓦斯抽采过程中参数演化及钻孔优化数值模拟研究[D]. 重庆大学,2018.

[66]Valliappan S, Zhang Wohua. Numerical modelling of methanegas migration in dry coal seams[J]. International Journal for Nu-merical and Analytical Methods in Geomechanics,1996, 20(8) :571-593.

[67]屈先朝. 瓦斯抽采孔孔距及煤层透气性的测定方法[J]. 煤炭学报, 2009, 34(11):1470-1474.

[68]煤层透气性系数测定方法—径向流量法:MT/T 1173-2019[S]. 2019.

[69]王刚,程卫民,郭恒,等.瓦斯压力变化过程中煤体渗透率特性的研究[J].采矿与安全工程学报, 2012,29(5):735-739.

[70]程远平,刘洪永,郭品坤,等.深部含瓦斯煤体渗透率演化及卸荷增透理论模型[J].煤炭学报, 2014,39(8):1650-1658.

[71]周厚权,申凯,陈宾.瓦斯抽采钻孔漏气类型划分与高效封孔技术应用研究[J].矿业安全与环保,2019,46(01):33-36+42.

[72]孙玉宁. 本煤层瓦斯抽采成孔与封孔技术研究[D]. 焦作:河南理工大学, 2008:47-56.

[73]林柏泉,张建国. 矿井瓦斯抽放理论与技术[M].徐州:中国矿业大学出版,1996:66-123

[74]Amit M S A S. ARIMA Modelling for Forecasting of Rice Production: A Case Study of India[J]. Agricultural Science Digest, 2020,40(4):404-407.

[75]王鹏,伍永平,王栓林,宋超,吴学明.矿井瓦斯浓度Lagrange-ARIMA实时预测模型研究[J].煤炭科学技术,2019,47(04):141-146.

[76]Selvakumar S, Kasthuri V. Forecasting Wheat Production in India Using ARIMA and Radial Basis Function[J]. Journal on Data Science and Technology. Vol, 2022, 8(4): 61-66.

[77]Amorim E A B E. Multidimensional Projection with Radial Basis Function and Control Points Selection: IEEE Pacific Visualization Symposium, Yokohama(JP), 2014[C].

[78]Yang W H A Y. Water Quality Sensor Model Based on An Optimization Method of RBF Neural Network[J]. Water and Environmental Engineering, 2020(1):1-11.

[79]Schilling R J, Carroll J J, Al-Ajlouni A F. Approximation of Nonlinear Systems with Radial Basis Function Neural Networks.[J]. IEEE Trans. Neural Networks, 2001,12(1).

[80]Fiesler E, Beale R. Handbook of Neural Computation[M]. CRC Press.