非均温度场采空区CO运移动力学机制及时空分布预测模型研究

CO is the main iconic gas for spontaneous combustion of coal. Accurate prediction the migration law and the trend of temporal and spatial distribution of CO in goaf is the key to successful early warning of coal spontaneous combustion hazard. CO in gobs with heterogeneous temperature field is the research object of this subject. The reconstruction and prediction method of CO concentration field under the condition of sparse monitoring points will be the target of this research. The following research is mainly carried out in this topic: (1) It is prepared to study the CO dispersion characteristics ( such as pressure difference, adsorption/resolution, oxidation reaction, etc.) in loose coal under multi-factors through the seepage and diffusion experiments of porous media. A dispersion kinetics model of CO in loose coal well be established through the above research. (2) The CO migration process in goaf under the influence of different high temperature conditions (such as position, temperature values, etc.) is studied by using simulation experiments and physical similar experiments. Then, Master the CO change and migration trajectory under the dual dimensions of time and space. The law of temporal and spatial distribution of CO under the influence of abnormal temperature is revealed. (3) It is proposed to use a multi-objective particle swarm optimization algorithm to select sparse monitoring points in the goaf. Then the CO sample database of the monitoring point will be created by us. A STKriging model will be establish for reconstructing the concentration field of CO. In ELM, time and space delay operators are introduced to create a STELM model to predict CO concentration. Finally, a prediction method of spatiotemporal distribution trend of CO in gobs with heterogeneous temperature field will be proposed. The research results will have important theoretical and practical significance for the arly warning of spontaneous combustion of coal in goaf and the analysis of the source of CO anomaly.

CO是煤自燃的主要标志性气体。采空区CO运移特征和时空分布的准确预测是煤自燃危险性成功预警的关键。本课题以温度异常采空区CO为研究对象，以稀疏测点条件下CO浓度场重构和预测为目标，开展以下研究：①利用多孔介质渗流和扩散实验，研究多因素（压差、吸附/解析、氧化反应等）叠加作用下松散煤体内CO渗流和扩散特征，建立CO运移动力学模型。②采用物理相似模拟和数值模拟实验，研究不同高温点（位置和温度值等）时的采空区CO运移过程，获取时空双重维度下CO变化量和运移轨迹，揭示非均温度场采空区CO时空分布规律。③采用多目标粒子群算法优选稀疏测点，创建CO样本数据集；建立时空克里金重构模型，重建采空区CO浓度场；在ELM中引入时间和空间延迟算子，构建时空极限学习机模型，预测CO分布态势；提出非均温度场采空区CO时空分布预测方法。研究成果对采空区自燃预警和CO异常涌出来源判定具有重要理论和实际意义。

煤自燃的精准预测与判定是世界性难题，气体是自燃预测预警的关键指标。CO是煤自燃的主要标志性气体。采空区CO运移特征和时空分布的准确预测是煤自燃危险性成功预警的关键。CO渗流与扩散过程中受干扰因素较多，影响对采空区CO时空分布规律的分析与判定；工作面CO传感器数量少，难以实时了解未布测点区域CO信息，无法对采空区CO时空分布态势进行有效预测。针对上述关键问题，项目以温度异常采空区CO为研究对象，以稀疏测点条件下CO浓度场重构和预测为目标，开展研究并取得相关结果。设计并搭建了松散煤体温度-渗流综合实验平台，测试分析了温度、粒径及变质程度等条件下松散煤体内气体渗流特性，掌握了煤温与渗透率的映射关系，建立了松散煤体内气体渗流数学模型，明确了不同高温条件下采空区气体运移路径演变规律；设计并搭建了松散煤体吸附/解吸实验装置，研究了温度、压差、粒度和水分对煤样吸附解吸CO气体的影响规律，掌握多因素叠加下松散煤体内CO吸附解吸特性；通过Materials Studio软件构建煤分子结构模型，运用蒙特卡罗方法模拟CO在煤分子模型中的吸附过程，剖析了CO的吸附等温线、吸附热和相互作用能，从微观角度探究了CO在煤分子中的吸附特性；根据松散煤体渗透率及CO吸附/解吸规律，结合达西定律、菲克定律及守恒方程，构建了CO运移扩散方程；研究了基于ELM模型的CO浓度时空关联性，提出了基于时空克里金模型的CO浓度场重构方法；以某矿工作面CO超限为研究对象，完成了重构预测模型精度的验证。研究成果为采空区自燃预警和CO异常涌出来源判定提供了理论与技术支撑。