薄煤层采煤机螺旋滚筒截割状态识别

Cutting State Identification of Thin Seam Shearer's Spiral Drum

薄煤层采煤机螺旋滚筒作为采煤机截割部截割煤岩的关键零件,螺旋滚筒的健康状态对薄煤层采煤机的采煤效率和开采安全具有重要影响。针对采煤机螺旋滚筒实际工作信息难以采集的问题,建立采煤机螺旋滚筒的多元耦合模型;搭建煤岩测试系统,进行随机动态载荷模拟;利用小波分析将获得的感知信号转化为应力时频谱图像,并利用神经网络拓展数据集,提升时频谱图的质量和提高网络识别的准确率,为螺旋滚筒的分析提供了高精度模型,为螺旋滚筒的优化设计提供了理论和技术支撑。

As a key component for cutting coal and rock in the cutting unit of a thin seam coal mining machine,the health status of the spiral drum has a significant impact on the coal mining efficiency and mining safety of the thin seam coal mining machine.In view of the difficulty in collecting the actual working information of the spiral drum of the coal mining machine,a coal and rock testing system was built to conduct random dynamic load simulations.Wavelet analysis was utilized to transform the obtained sensing signals into stress time-frequency spectrum images,and neural networks were employed to improve the quality of the time-frequency spectrum images and the accuracy of network recognition,providing a high-precision model for the analysis of the spiral drum and theoretical and technical support for the optimal design of the spiral drum.