为满足煤矿安全生产的需求，针对煤层顶板高地应力区域易诱发煤矿冲击地压等动力灾害问题，利用微震台站实时传输数据，采用快速三维层析反演算法对煤岩层波速进行反演，从而实现对高地应力区域的实时监测和快速预警。层析反演结果的精度直接决定了高地应力区域的判断，因此，通过建立三维地质模型，分析微震事件空间分布的不同以及高地应力区域与煤层顶板的间距对层析反演结果的影响，并将该方法在某矿区进行了试验。试验结果表明：(1) 由于地震波在煤层和围岩中的传播规律不同，微震事件在煤岩层空间分布不同会降低探测高地应力区域的精确度；(2) 煤层顶板与高地应力区域间距过小时，高波速区域附近所产生的波速梯度会影响低速区域，导致反演结果中煤层位置不清晰；(3) 通过筛选微震事件使其均匀分布以及合理的布置微震地面观测系统可以有效的提高数据完整度和反演精度。研究结果为基于微震的快速三维层析反演技术探测煤层顶板高地应力区域提供理论依据。

High in-situ stress areas in the coal seam roofs are prone to induce mine dynamic hazard such as rock bursts. To meet the demand for safe coal mining, this study achieved real-time monitoring and swift early warning of high in situ stress areas. Specifically, based on data transmitted from microseismic stations in real time, this study conducted the inversion of the wave velocities in coal seams and rocks using a fast three-dimensional (3D) tomographic inversion algorithm. The accuracy of tomographic inversion results directly determines the judgment on the locations of high in situ stress areas. Therefore, by building a 3D geological model, this study analyzed the influence of different spatial distributions of microseismic events and the spacing between high in situ stress areas and the coal seam roofs on the tomographic inversion results. Furthermore, the method used in this study was tested in a certain mining area. The test results are as follows: (1) Due to different propagation patterns of seismic waves in coal seams and surrounding rocks, the different spatial distributions of microseismic events in coal seams and rocks can reduce the detection accuracy of high in situ stress areas; (2) In the case of too small spacing between high in situ stress areas and the coal seam roofs, the wave velocity gradients near areas with high wave velocities will affect areas with low wave velocities, leading to unclear coal seam locations in the inversion results. (3) The data integrity and inversion accuracy can be effectively improved using evenly distributed microseismic events and rationally arranged microseismic surface observation system. The results of this study will provide a theoretical basis for detecting high in situ stress areas through microseism-based fast 3D tomographic inversion.