矿井多人员定位轨迹的预警分类方法研究  

Study on Early Warning Classification Method of Mine Multi-personnel Location Track

作　　者：蔡安江[1] 徐海涛 程东波 刘锋伟 CAI Anjiang;XU Haitao;CHENG Dongbo;LIU Fengwei(School of Mechanical and Electrical Engineering,Xi′an University of Architecture and Technology,Xi′an 710055,China;China Minmetals Corporation Limited,Beijing 100010,China)

机构地区：[1]西安建筑科技大学机电工程学院,陕西西安710055 [2]中国五矿集团有限公司,北京100010

出　　处：《金属矿山》2025年第1期243-249,共7页Metal Mine

基　　金：工信部物联网集成创新与融合应用项目(编号:2018-4700);榆林市科技计划项目(编号:CXY-2022-172)。

摘　　要：为解决矿井综采操作区域多人员定位轨迹的预警分类问题,提出了一种基于超宽带(Ultra Wide Band,UWB)的多人员定位轨迹数据的预警分类方法。该方法首先对采集的UWB定位轨迹数据进行预处理;然后利用UWB定位轨迹数据中的人员ID、坐标、时间、求救信号等特征参数作为UWB人员定位轨迹预警分类模型的输入指标,以人员的预警行为类别作为输出指标,对预警分类模型进行拟合训练,基于人员4级违规预警机制与专家建议设置预警阈值;最后采用随机森林算法对多人员UWB定位轨迹数据进行人员行为预警识别和分类。研究表明:该方法能够对区域人员作业超员、工作超时、作业求救、定位轨迹缺失和作业越界等行为进行有效预警并准确分类,能够消除隐患,提高矿山人员管理效率和生产作业的安全性。In order to solve the problem of early warning classification of multi-personnel location track in fully mechanized mining operation area,an early warning classification method based on Ultra Wide Band(UWB)multi-personnel location track data is proposed.The method firstly preprocesses the collected UWB positioning track data.Then,characteristic parameters such as personnel ID,coordinates,time and distress signal in UWB positioning trajectory data are used as input indicators of the UWB positioning trajectory early-warning classification model,and personnel early-warning behavior categories are used as output indicators.The early-warning classification model is fitted and trained,and the early-warning threshold is set based on the four-level violation early-warning mechanism of personnel and expert suggestions.Finally,the random forest algorithm is adopted to identify and classify the multi-person UWB location track data.The research shows that this method can effectively give early warning and accurately classify the behaviors of regional personnel such as overworking,overtime,job distress,missing positioning track and operation crossing,which can eliminate hidden dangers,improve the management efficiency of mining personnel and avoid dangerous accidents.

关 键 词：矿井定位 多人员 预警分类 UWB定位轨迹数据 随机森林算法 

分 类 号：TD76[矿业工程—矿井通风与安全]