综放工作面放煤工艺中的煤矸智能识别是智能化综放开采的瓶颈，经过十几年的不断探索研究，基于煤层的沉积环境和岩性赋存特征，提出了煤岩层低水平自然γ射线辐射高精度实时探测方法和煤矸自动识别原理，为综放工作面放煤工序的煤矸智能识别控制奠定了基础。根据特厚煤层含多层夹矸的复杂结构特征，以提高煤矸识别的精准性为目的，分析了特厚煤层夹矸的赋存特点及自然辐射特性；开展了煤-矸(夹矸)-岩(直接顶岩石)放落流动的时序规律研究，得出顶煤中夹矸的层位、间距、层数等因素会影响放落流动时序，且呈现不同的时序状态，据此进行了放落流动时序分区；相应地不同时序区的混矸率会呈现不同的特征：当夹矸无时序放出时，混矸率在一固定的范围内波动，当煤-夹矸渐次放出时，混矸率呈现上升—稳定—连续台阶式上升的特征，直接顶矸石的混入，使得混矸率呈线性增加；基于煤-矸-岩放落流动的时序特征，得出了不同时序条件下自然射线的辐射变化特征，确定了自然射线辐射强度台阶式变化与夹矸赋存参数的影响关系，区分了放煤流中含有夹矸和直接顶矸石混入的辐射特征，确定了不同煤-矸-岩放落时序区的辐射强度范围，提出了以自然射线辐射强度作为煤-矸-岩自动识别的主要参数和2种不同关窗方式的识别方法；在同煤集团塔山煤矿进行了现场测试，测试结果与研究结果得到了很好的印证，达到了预期效果，为智能放煤的现场应用奠定了基础。

The intelligent coal-gangue identification is the bottleneck of top-coal-caving mining operation. After more than ten years of continuous exploration and research, based on the deposition environment and lithological characteristics of coal seams, the authors have proposed a high-precision and real time detection method based onlow level natural ray radiation in coal and rock seams and the automatic identification principle of coal and gangue. It lays the foundation for the intelligent identification and control of coal and gangue in top-coal-caving mining working face. According to the complex structural characteristics of extra thick coal seam containing multi-layer gangue, in order to improve the reliability of coal gangue identification, the characteristics of occurrence and natural radiation of coal and gangue  in extra thick coal seam are analyzed. The research of coal-gangue(parting)rock (immediate roof rock) timing law in the caving flow is carried out, and it is concluded that the layer position, spacing and number of gangue layer in top coal affect the cavingflow sequential rules, which can present some different timing states according to which the release flow timing zoning is carried out. Accordingly, the mixed gangue rate will present different characteristics. That is, when the partingis drawn outwith no timing order, the mixing rate fluctuates within a fixed range, and when the coal and gangue are caved gradually, the mixing rate shows the characteristics of rising-stable-continuous step increase, and the mixing of immediate roof rock makes the mixing rate increase linearly. According to the timing characteristics of coal-gangue-rock fall flow, the radiation change characteristics of natural rays under different timing conditions are studied, and the relationship between the step change of natural ray radiation intensity and the storage parameters of partingis determined. Based on the relationship above the main parameters of coal-gangue-rock automatic identification and the identification methods of two different window closing methods are proposed. Finally, the field test has been carried out in the Tashan Coal Mine of Tongmei Group, China, and the test and research results have been well verified. The expected results have been achieved, which lays a foundation for the field application of intelligent top caving.

1 特厚煤层夹矸的赋存特点及自然辐射特性

   1.1 典型矿区特厚煤层夹矸的赋存分布特征

   1.2 夹矸及直接顶放射性辐射特征

2 特厚煤层煤-矸-岩放落流动的时序规律

   2.1 正常放煤期间煤-矸-岩放落流动的时序规律

   2.2 夹矸赋存参数对煤-矸-岩放落流动时序的影响

   2.3 正常放煤期间煤-矸-岩放落流动的混矸规律

3 基于煤-矸-岩放落流动时序特征的自然射线辐射变化规律

4 煤-矸-岩识别方法及现场试验

   4.1 复杂结构特厚煤层综放煤-矸-岩识别方法

   4.2 现场试验

5 结论