基于深度学习的煤炭井下皮带运输机异常检测方法研究  

Research on Deep Learning-based Anomaly Detection Method for Coal Underground Belt Conveyor

作　　者：张海红 Zhang Haihong(Shanxi Coking Coal Shanxi Coal International Hanjiawa Coal Co.,Ltd.,Datong Shanxi 037100,China)

机构地区：[1]山西焦煤山煤国际韩家洼煤业有限公司,山西大同037100

出　　处：《机械管理开发》2025年第2期188-190,共3页Mechanical Management and Development

摘　　要：为提高煤炭井下皮带运输机的运行可靠性,采用深度学习技术,设计并实现了皮带运输机异常检测系统。通过分析煤炭井下环境特点及皮带运输机的结构和运行数据,研究构建了多层次的数据处理与学习模型,包括数据预处理、特征提取、深度神经网络训练。结果表明,该系统能高效识别皮带运输机的异常状态,并提供前瞻性预测报告。研究成果可为煤矿安全生产提供了技术支持,全面提升井下作业的安全保障水平。In order to improve the operation reliability of the belt conveyor in coal underground,this study adopts the deep learning technology to design and realize the abnormality detection system of the belt conveyor.By analyzing the characteristics of coal underground environment and the structure and operation data of the belt conveyor,the study constructs a multi-level data processing and learning model,including data preprocessing,feature extraction,and deep neural network training.The results show that the system can efficiently identify the abnormal state of the belt conveyor and provide a forward-looking prediction report.The research results can provide technical support for the safe production of coal mines and comprehensively improve the safety and security level of underground operations.

关 键 词：深度学习 皮带运输机 异常检测 

分 类 号：TH222[机械工程—机械制造及自动化] TP18[自动化与计算机技术—控制理论与控制工程]