随着矿井开采强度与深度逐渐加强，采空区面积随之扩大，使得煤自燃隐患更加突出。受采空区复杂环境的影响，遗煤自燃高温点的产生与发展状况不可视，加之煤体的导热性能较差，导致采空区火源位置难以精准判断，致使矿井自燃发火问题难以有效的解决。本文通过现场测试、实验研究、数值模拟相结合的方法，开展了采空区氧浓度场、自燃温度场、含点热源采空区自燃气体CO浓度场分布规律的动态模拟。基于模拟结果以煤自燃释放的CO气体与采空区隐蔽热源位置之间的复杂非线性关系入手，开展以采空区边界CO浓度为适应度值的隐蔽火源位置反演方法研究。主要研究工作及内容如下：

首先，理论分析了井下气体流动状态和CO运移形式。得到井下CO运移主要包括：CO分子扩散、CO气体与通风分流之间的对流以及CO在多孔介质中的渗流；通过微积分分析和刘园子51507工作面现场数据检测得到CO浓度场分布存在着空间位置上的不对称性和在时间变化上的延迟性；采用程序升温实验研究了煤样氧化特性及CO生成速率，构建了煤自燃气体CO生成速率的数学模型。

其次，基于现场参数与实验结果，结合“O”型圈理论的采空区孔隙率分布特征，运用COMSOL Multiphysics软件建立以刘园子51507为研究对象的采空区煤自燃模型，模拟采空区氧浓度场分布及煤氧化高温范围动态分布特征。结果表明采空区氧气浓度场的分布规律表现为进风侧分布较深且朝回风方向逐渐向工作面收缩。采空区煤自燃高温区域主要集中在采空区内20m～50m范围内，高温范围与工作面相对位置随时间的变化逐渐趋于稳定。

再次，以氧气体积分数与煤自燃条件下温度场为判断依据的双指标划分采空区氧化升温带范围，结果表明工作面后方10m~100m范围内进入氧化区域；在采空区氧化升温带随机选取热源点A(25, 0)、B(20, 30) 、C(25, 70)，开展含点热源采空区煤自燃气体CO浓度场的动态模拟，得出CO以热源点为起点向回风方向和工作面方向运移，且采空区边界处CO浓度峰值的位置随热源点的改变而不同。

最后，采用PSO与有限元方法相结合，将采空区隐蔽热源位置反演问题转化为比较观测点CO浓度与反演输出CO浓度差异最小的问题。运用COMSOL Multiphysics与MATLAB联合反演方法，以点热源A、B、C模拟结果为反演数据支持，研究基于采空区边界CO浓度反演采空区隐蔽热源位置的可行性。点热源A、B、C在50次迭代之后，粒子群在搜索过程中从初始的随机分布不断向热源位置聚集，最终反演求得的热源位置为A(25.014, 0.0496)、B(20.101, 29.600) 、C(20, 69.831)，与点热源实际位置距离误差小于热源半径；通过附加测量误差信息对反演结果的影响研究，发现CO数据采集随机误差β在(-3%, 3%)以下时，反演所得隐蔽火源位置与真实位置之间的距离在热源半径之内。结果表明基于CO浓度场的采空区热源位置反演方法在CO数据采集随机误差β在绝对值3%以内时对于采空区隐蔽热源位置的判定是可行的。

With the gradual strengthening of mining intensity and depth, the goaf area expands, which makes the hidden danger of coal spontaneous combustion more prominent. Affected by the complex environment of the goaf, the generation and development of high temperature point of spontaneous combustion of residual coal is invisible, coupled with the poor thermal conductivity of coal, which makes it difficult to accurately judge the location of the fire source in the goaf, and the problem of spontaneous combustion in the mine is difficult to be effectively solved. In this paper, the dynamic simulation of the distribution law of the concentration field of oxygen in goaf, the temperature field of spontaneous combustion and the concentration field of CO in goaf containing point heat source is carried out through the method of field test, experimental study and numerical simulation. Based on the simulation results, based on the complex nonlinear relationship between the CO gas released by spontaneous combustion of coal and the location of concealed heat source in goaf, the inversion method of concealed fire source location with the CO concentration at the goaf boundary as the fitness value is carried out. The main research work and contents are as follows:

Firstly, the gas flow state and CO migration form are analyzed theoretically. It is concluded that the CO migration mainly includes CO molecular diffusion, convection between CO gas and ventilation shunt, and CO seepage in porous media. Through the analysis of calculus and the detection of field data of Liu Yuanzi 51507 working face, it is found that the distribution of CO concentration field has asymmetry in spatial position and delay in time change. The oxidation characteristics and CO generation rate of coal samples were studied by temperature programmed experiment, and the mathematical model of CO generation rate of coal spontaneous combustion gas was established.

Secondly, based on field parameters and experimental results, combined with the distribution characteristics of goaf porosity based on "O" ring theory, COMSOL Multiphysics software was used to establish the spontaneous combustion model of goaf coal with Liu Yuanzi 51507 as the research object, and simulate the distribution characteristics of oxygen concentration field in goaf and the dynamic distribution characteristics of coal oxidation high temperature range. The results show that the distribution of oxygen concentration field in goaf is deep at the inlet side and gradually shrinks towards the working face towards the return direction. The high temperature area of spontaneous combustion of coal in goaf is mainly concentrated in the range of 20m to 50m within the goaf, and the high temperature range and relative position of working face gradually tend to be stable over time.

Thirdly, oxygen volume fraction and temperature field under the condition of spontaneous combustion of coal were used as two indexes to divide the range of goaf oxidation heating zone. The results show that the oxidation zone enters within 10m~100m behind the working face. Heat source points A(25, 0), B(20, 30) and C(25, 70) were randomly selected in the oxidation heating zone of goaf to carry out dynamic simulation of CO concentration field of coal spontaneous combustion gas in goaf with point heat source. It was concluded that CO migrated from heat source point to return air direction and working face direction. The peak of CO concentration at the goaf boundary varies with the change of heat source.

Finally, by combining PSO and finite element method, the problem of location inversion of concealed heat source in goaf is transformed into the problem of minimum difference between CO concentration of observation point and output CO concentration of inversion. Based on the joint inversion method of COMSOL Multiphysics and MATLAB, with the simulation results of point heat source A, B and C as the inversion data support, the feasibility of inversion of the location of hidden heat source in goaf based on the CO concentration at the goaf boundary was studied. After 50 iterations of point heat sources A, B and C, the particle swarm continuously gathers towards the heat source position from the initial random distribution in the search process, and the heat source position obtained by the final inversion is A(25.014, 0.0496), B(20.101, 29.600) and C(20, 69.831). The distance error from the actual position of the point heat source is less than the heat source radius. Through the study of the influence of additional measurement error information on the inversion results, it is found that when the random error β of CO data acquisition is below (-3%, 3%), the distance between the inversion location of concealed fire source and the real location is within the radius of heat source. The results show that the inversion method based on CO concentration field is feasible to determine the location of hidden heat source in goaf when the random error β of CO data acquisition is within the absolute value of 3%.

[1] 樊大磊, 李富兵, 王宗礼等. 碳达峰、碳中和目标下中国能源矿产发展现状及前景展望[J]. 中国矿业, 2021, 30(06): 1-8.

[2] 袁亮, 张农, 阚甲广等. 我国绿色煤炭资源量概念、模型及预测[J]. 中国矿业大学学报, 2018, 47(01): 1-8.

[3] 邓军, 白祖锦, 肖旸等. 煤自燃灾害防治技术现状与挑战[J]. 煤矿安全, 2020, 51(10): 118-125.

[4] 朱令起. 矿井火灾预测预警及密闭启封安全性研究[D]. 北京: 中国矿业大学, 2010.

[5] 姜进军. 白芨沟煤矿渣台隐蔽火区探测技术[J]. 煤矿安全, 2014, 45(07): 53-55.

[6] 张宏. 神东矿区煤炭自燃特性及防灭火技术研究[D]. 阜新: 辽宁工程技术大学, 2013.

[7] 翟小伟. 煤氧化过程CO产生机理及安全指标研究[D]. 西安: 西安科技大学, 2012.

[8] 李波, 巨广刚, 王珂, 等. 2005—2014年我国煤矿灾害事故特征及规律研究[J]. 矿业安全与环保, 2016, 43(03): 111-114.

[9] 王德明, 邵振鲁, 朱云飞. 煤矿热动力重大灾害中的几个科学问题[J]. 煤炭学报, 2021, 46(01): 57-64.

[10] 国家矿山安全监察局.事故查询[EB/OL].[2023-2-15]. www.chinamine-safety.gov.cn

[11] 邓军, 徐精彩, 文虎. 采空区自然发火动态数学模型研究[J]. 湘潭矿业学院学报, 1998(01): 11-16.

[12] 高树山. 采空区自燃发火及其预防[J]. 煤炭技术, 2006(07): 65-67.

[13] 谢振华, 金龙哲, 任宝宏, 等. 煤炭自燃特性与指标气体的优选[J]. 煤矿安全, 2004(2): 10-12.

[14] Yugou W, jianming W. Experimental study on significant gases of coal spontaneous combustion by temperature programmed [J]. Procedia Engineering, 2011(26): 120-125.

[15] 朱令起, 周心权, 朱迎春. 东欢坨煤矿自然发火标志气体的优化选择[J]. 西安科技大学学报, 2008(2): 288-292.

[16] Liu Wei, Qin Yueping. A quantitative approach to evaluate eisks of spontanceous combustion in longwall gobs based on CO emissions at upper corner[J]. Fuel, 2017, 210: 359-370.

[17] 余明高, 解俊杰, 贾海林. 机械力作用下煤结构破断CO释放规律及自燃预测指标修正方法[J]. 中国矿业大学学报, 2017, 46(4): 762-768.

[18] Itay M, Hill, et al. A study of the low temperature oxidation of coal [J]. Fuel Processing Technology, 1989, 21(2): 81-97.

[19] Yu Ming-gao, Zheng Yan-min, Lu Chang. Thermal analysis of kinetics of coal oxidation [J]. The 6th International Conference on Mining Science & Technology. Procedia Earth and Planetary Science, 2009: 341–346.

[20] P Lu, G X Liao, J H Sun. Experimental research on index gas of the coal spontaneous at low-temperature stage[J]. Journal of Loss Prevention in the Process Industries, 2004, 3(17): 243-247.

[21] Yugou W, jianming W. Experimental study on significant gases of coal spontaneous combustion by temperature programmed(TP)[J]. Procedia Engineering, 2011(26): 120-125.

[22] 邓军, 赵婧昱, 张嬿妮. 基于指标气体增长率分析法测定煤自燃特征温度[J]. 煤炭科学技术, 2014(07): 49-52．

[23] 安靖宇, 张健, 季德阁等. 基于分段拟合的煤自燃指标气体优选研究[J]. 中国安全生产科学技术, 2021, 17(10): 25-31.

[24] 肖旸, 马砺, 王振平, 等. 煤自燃指标气体的吸附与浓缩规律[J]. 煤炭学报, 2007, 32(10): 1014-1018.

[25] 马汉鹏, 陆伟, 王宝德. 煤自燃过程指标气体产生规律的系统研究[J]. 矿业安全与环保, 2007, 34(6): 4-6.

[26] 刘乔, 王德明, 仲晓星, 等. 基于程序升温的煤层自然发火指标气体测试[J]. 辽宁工程技术大学学报（自然科学版）, 2013(03): 362-366.

[27] 陈汝豪, 王德明, 曹凯. 煤炭低温氧化自燃过程指标气体影响因素关联分析[J]. 矿业工程研究, 2012(01): 32-38．

[28] 李宗翔, 吴志君, 王振祥. 采空区遗煤自燃升温过程的数值模型及其应用[J]. 安全与环境学报, 2004(06): 58-61.

[29] 李莉, 张人伟, 裴晓东, 等. 不同煤种的氧化自燃指标气体分析[J]. 矿业安全与环保, 2006(05): 69-71.

[30] 刘璐, 梅国栋. 基于灰色关联分析的煤自然发火气体预报指标研究[J]. 陕西煤炭, 2008: 33-36.

[31] 周春山, 吴玉国. CO2灭火技术在高瓦斯封闭火区的应用[J]. 中国煤炭, 2010, 37(3) : 90-92.

[32] 杨广文, 艾兴. 大雁二矿250综采工作面CO来源的分析及治理[J]. 煤矿安全, 2003(10): 41-43.

[33] Li J, Lu W, Kong B, et al. Mechanism of gas generation during low-temperature oxidation of coal and model compounds[J]. Energy & Fuels, 2019, 33(2): 1527-1539.

[34] Wang J, Zhang Y, Xue S, et al. Assessment of spontaneous combustion status of coal based on relationships between oxygen consumption and gaseous product emissions[J]. Fuel processing technology, 2018, 179: 60-71.

[35] 薛冰, 李再峰, 陈兴权等. 低阶煤在干燥氧气下低温氧化过程的机理研究[J]. 煤炭转化, 2006(02): 12-15.

[36] Zhang D, Yang X, Deng J, et al. Research on coal spontaneous combustion period based on pure oxygen adiabatic oxidation experiment[J]. Fuel, 2021, 288: 119651.

[37] 王涌宇, 邬剑明, 王俊峰, 等. 煤氧化过程中CO生成机理的原位红外实验研究[J]. 煤炭学报, 2016, 41(02): 451-457.

[38] 张哲, 魏晨慧, 刘书源等. 煤粒尺寸对气体扩散过程影响的数值模拟研究[J]. 矿业研究与开发, 2021, 41(07): 85-92.

[39] 李宗翔,衣刚,武建国等.基于“O”型冒落及耗氧非均匀采空区自燃分布特征[J].煤炭学报,2012,37(03):484-489.DOI:10.13225/j.cnki.jccs.2012.03.031.

[40] 高光超, 李宗翔, 张春, 等. 基于三维“O”型圈的采空区多场分布特征数值模拟[J]. 安全与环境学报, 2017, 17(03): 931-936.

[41] 夏海斌. 灵武矿区易自燃煤层综放面采空区CO运移规律研究[D]. 西安: 西安科技大学, 2011.

[42] 赵勇. 烧结余热回收竖罐内气体流动数值计算[D]. 沈阳: 东北大学, 2012.

[43] Brown D L, Popov P, Efendiev Y. Effective equations for fluid-structure interaction with applications to poroelasticity[J]. Applicable Analysis, 2014, 93(4): 771-790.

[44] Lei C, Jiang L, Bao R, et al. Study on Multifield Migration and Evolution Law of the Oxidation Heating Process of Coal Spontaneous Combustion in Dynamic Goaf[J]. ACS Omega, 2023.

[45] Gui X, Xue H, Zhan X, et al. Measurement and numerical simulation of coal spontaneous combustion in goaf under Y-type ventilation mode[J]. ACS omega, 2022, 7(11): 9406-9421.

[46] Xie S, Wang G, Wang E, et al. Determination of Hazardous Zone of Coal Spontaneous Combustion in Ultra-Long Working Face Based on the Gob Porosity Evolution and Flow Field Distribution[J]. Applied Sciences, 2023, 13(7): 4574.

[47] 李宗翔, 吴志君, 马友发. 采空区场域自燃CO向工作面涌出的数值模拟[J]. 燃烧科学与技术, 2006(6): 563-568.

[48] 李宗翔. 综放工作面采空区瓦斯涌出规律的数值模拟研究[J]. 煤炭学报, 2002 (2): 173-178.

[49] 李宗翔, 纪书丽, 题正义. 采空区瓦斯与大气两相混溶扩散模型及其求[J]. 岩石力学与工程学报, 2005, 24(16): 971-976.

[50] Zhu X, Wen H. Numerical simulation study on the influence of air leakage on oxygen concentration in goafs of fully mechanized caving mining with shallow buried and large mining height[J]. Frontiers in Earth Science, 2023, 11: 1138925.

[51] 翟小伟. 煤氧化过程CO产生机理及安全指标研究[D]. 西安: 西安科技大学, 2012.

[52] 张玉波, 张健, 冯建生, 等. 察哈素煤矿回风隅角CO积聚影响因素关联度分析[J]. 煤矿安全, 2020, 51(12): 224-227.

[53] 王新宇. 易自燃煤层综采面超大采空区CO运移规律及控制技术研究[D]. 太原: 太原理工大学, 2013.

[54] Liu Wei, Qin Yueping. A quantitative approach to evaluate risks of spontaneous combustion in longwall gobs based on CO emissions at upper corner[J]. Fuel, 2017, 210: 359-370.

[55] 赵罗飞. 补连塔煤矿工作面低氧原因及防控方法研究[D]. 徐州: 中国矿业大学, 2020.

[56] Chen Xiangjun, Feng Shuailong, Wang Lin, et al. Distribution and Prevention of CO in a Goaf of a Working Face with Y-Type Ventilation[J]. ACS omega, 2021, 6(3): 1787-1796.

[57] 王锐. 火区下近距离煤层开采有害气体入侵灾害防控技术研究[D]. 青岛: 山东科技大学, 2018.

[58] Wen Hu, Yu Zhijin, Fan Shixing, et al. Prediction of spontaneous combustion potential of coal in the gob area using CO extreme concentration: A case study[J]. Combustion Science and Technology, 2017, 189(10): 1713-1727.

[59] 贾玉泉. 综采工作面上隅角CO积聚特性分析[D]. 西安: 西安科技大学, 2019.

[60] 常明然. 高产高效矿井CO产生机理及控制方法[D]. 北京: 中国矿业大学, 2017.

[61] Taraba B, Michalec Z. Effect of longwall face advance rate on spontaneous heating process in the gob area–CFD modelling[J]. Fuel, 2011, 90(8): 2790-2797.

[62] Brodny J, Tutak M. Applying computational fluid dynamics in research on ventilation safety during underground hard coal mining: A systematic literature review[J]. Process Safety and Environmental Protection, 2021, 151: 373-400.

[63] 胡汉华. 矿井通风系统设计[M]. 北京: 化学工业出版社, 2010.

[64] 梁栋. 通风过程瓦斯运移规律和数值模拟[M]. 北京: 煤炭工业出版社, 1999: 5.

[65] 陈开岩, 张占国, 林柏泉, 等. 综放工作面抽放条件下瓦斯涌出及分布特征[J]. 采矿与安全工程学报, 2009, 26(4): 418-422.

[66] Cui M, Li N, Liu Y, et al. Robust inverse approach for two-dimensional transient nonlinear heat conduction problems[J]. Journal of Thermophysics and Heat Transfer, 2015, 29(2): 253-262.

[67] 韦善阳. 瓦斯异常涌出气体运移规律及影响范围研究[D]. 北京: 中国矿业大学, 2013.

[68] 王恩元, 梁栋, 柏发松. 巷道瓦斯运移机理及运移过程的研究[J]. 山西矿业学院学报, 1996, (02): 24-29.

[69] 刘伟, 秦跃平, 乔珽等. 煤耗氧速率与CO生成速率的计算及实验论证[J]. 中国矿业大学学报, 2016, 45(06): 1141-1147.

[70] 秦跃平, 宋怀涛, 刘伟等. 煤粒分散度对遗煤自燃影响的试验研究[J]. 煤矿安全, 2015, 46(1): 21-2.

[71] 朱建芳, 蔡卫, 秦跃平. 基于移动坐标的采空区自然发火模型研究[J]. 煤炭学报, 2009, 34(8): 1095-109.

[72] 徐永圻. 采矿学[M]. 徐州: 中国矿业大学出版社, 2003: 531-533.

[73] 钱鸣高. 矿山压力与岩层控制[M]. 徐州: 中国矿业大学出社, 2003: 119-180.

[74] Xia T, Wang X, Zhou F, et al. Evolution of coal self-heating processes in longwall gob areas[J]. International Journal of Heat and Mass Transfer, 2015, 86: 861-868.

[75] Sidiropoulou M G, Moutsopoulos K N, Tsihrintzis V A. Determination of Forchheimer equation coefficients a and b[J]. Hydrological Processes, 2007, 21(4): 534-554.

[76] Xia T, Zhou F, Wang X, et al. Controlling factors of symbiotic disaster between coal gas and spontaneous combustion in longwall mining gobs[J]. Fuel, 2016, 182: 886- 896.

[77] Ejlali A, Mee D J, Hooman K, et al. Numerical modelling of the self-heating process of a wet porous medium[J]. International Journal of Heat and Mass Transfer, 2011, 54: 5200-5206.

[78] Alazmi B, Vafai K. Analysis of variants within the porous media transport models[J]. Journal of Heat Transfer-Transactions of the ASME, 2000, 122(2): 303-326.

[79] 李宗翔,刘宇,吴邦大等.基于封闭耗氧实验的窒熄带氧临界体积分数研究[J].煤炭学报,2017,42(07):1776-1781.

[80] 衡益, 罗玖, 杨青青等. 池沸腾强化传热中的三维瞬态热传导反问题[J]. 科学通报, 2020, 65(18): 1857-1874.

[81] Cui M, Li N, Liu Y, et al. Robust inverse approach for two-dimensional transient nonlinear heat conduction problems[J]. Journal of Thermophysics and Heat Transfer, 2015, 29(2): 253-262.

[82] 高明荣. 基于粒子群优化的神经网络算法与天线设计[D]. 上海: 东华大学, 2022.