不同采煤工序下CO涌出特征提取及测试数据研究  

Research on the feature extraction and test data of CO emission under different coal mining processes

作　　者：邢震 XING Zhen(CCTEG Changzhou Research Institute Co.,Ltd.,Changzhou,Jiangsu 213015,China;Tiandi(Changzhou)Autumation Co.,Ltd.,Changzhou,Jiangsu 213015,China)

机构地区：[1]中煤科工集团常州研究院有限公司,江苏省常州市213015 [2]天地(常州)自动化股份有限公司,江苏省常州市213015

出　　处：《中国煤炭》2023年第2期44-51,共8页China Coal

基　　金：天地(常州)自动化股份有限公司科研项目(2022TY2004);天地科技股份有限公司科技创新创业资金专项项目(2021-TD-ZD004)。

摘　　要：根据某煤矿综采工作面正规作业流程实际情况以及不同阶段CO涌出规律,合理划分采煤工序。在实验室通过上隅角CO浓度时间序列的小波阈值去噪,消除了原始序列中大部分噪声的影响,同时保持了原序列的特征和趋势,为后续研究CO涌出特征提取奠定了良好的数据基础。阐述了综采工作面根据能量谱特征提取和CO涌出特征提取的技术方法,并基于RBF神经网络对CO涌出模式进行了有效识别。结果证明,采用小波包能量谱作为特征向量对综采工作面不同工序进行模式识别的方法是可行的。According to the actual situation of the formal operation process of a fully mechanized mining face in a coal mine and the rule of CO emission at different stages,the mining process is reasonably divided.In the laboratory,the wavelet threshold de-noising of the corner CO concentration time series eliminates the influence of most of the noise in the original sequence,while maintaining the characteristics and trends of the original sequence,which lays a good data foundation for the subsequent study of CO emission feature extraction.This paper expounds the extraction technical methods of the features of energy spectrum and CO emission in the fully mechanized mining face,and effectively identifies the CO emission pattern based on RBF neural network.The results show that it is feasible to use the wavelet packet energy spectrum as the feature vector for pattern recognition of different processes in fully mechanized mining face.

关 键 词：安全监控系统 CO涌出 小波包 能量谱 模式识别 

分 类 号：TD655[矿业工程—矿山机电]