基于安全监控系统CO数据的综采工作面作业工序模糊聚类模式智能识别研究  

Study on Intelligent Recognition of Fuzzy Clustering Pattern of Working Procedure in Fully Mechanized Mining Face Based on CO Data of Safety Monitoring System

作　　者：邢震 Xing Zhen(Changzhou Research Institute,China Coal Technology and Engineering Group,Changzhou 213015,China;Tiandi(Changzhou)Autumation Co.,Ltd.,Changzhou 213015,China)

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

出　　处：《煤矿机械》2023年第1期66-69,共4页Coal Mine Machinery

基　　金：煤矿生产调度协同管控应用研究(智能生产调度协同与班组精细化管理)(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。通过收集煤矿安全监控系统KJ95X现场采集的上隅角CO浓度时间数据,采用小波包阈值去噪的方法对原始数据进行数据处理,利用小波包分解提取不同采煤工序数据的能量谱作为特征向量,模糊c-均值聚类和加权模糊c-均值聚类对比聚类识别准确率和计算耗时发现,加权模糊c-均值聚类耗时长,但是聚类准确度大大提升。According to the actual situation of the regular working process and the law of CO emission at different stages of a coal mine fully mechanized mining face,the mining procedures are reasonably divided.By collecting the CO concentration time data at the upper corner collected on site by the coal mine safety monitoring system KJ95X,the wavelet packet threshold de-noising method was used to process the original data,and the energy spectrum of different coal mining process data was extracted by using wavelet packet decomposition as the feature vector.Fuzzy c-mean clustering and weighted fuzzy c-mean clustering were compared to identify the accuracy of clustering and calculate the time consumption,found that weighted fuzzy c-means clustering takes a long time,but the clustering accuracy is greatly improved.

关 键 词：安全监控系统 作业工序 模糊聚类分析 小波包 能量谱 模式识别 

分 类 号：TD655[矿业工程—矿山机电]