基于MRAU视频分割模型的矿井涌(突)水风险识别方法  

Mine water inrush risk identification method based on MRAU video segmentation model

作　　者：武强 张帅 杜沅泽 徐华[4] 赵颖旺 WU Qiang;ZHANG Shuai;DU Yuanze;XU Hua;ZHAO Yingwang(Inner Mongolia Research Institute,University of Mining and Technology-Beijing,Ordos 017000,China;National Engineering Research Center of Coal Mine Water Hazard Controlling,China University of Mining and Technology-Beijing,Beijing 100083,China;Key Laboratory of Mine Water Control and Resources Utilization,National Mine Safety Administration,Beijing 100083,China;College of Information Engineering,Beijing Institute of Petrochemical Technology,Beijing 102617,China)

机构地区：[1]矿业大学(北京)内蒙古研究院,内蒙古鄂尔多斯017000 [2]中国矿业大学(北京)国家煤矿水害防治工程技术研究中心,北京100083 [3]矿山水防治与资源化利用国家矿山安全监察局重点实验室,北京100083 [4]北京石油化工学院信息工程学院,北京102617

出　　处：《煤炭科学技术》2024年第11期17-28,共12页Coal Science and Technology

基　　金：国家自然科学基金资助项目(42027801,42202283);中央高校基本科研业务费资助项目(2022YJSSH01)。

摘　　要：矿井涌(突)水视频识别是智能化矿井建设的关键之一,通过识别涌(突)水从无到有、从小到大的动态演变过程,有助于防止水量超出矿井排水能力并演变为水害。为此提出了一种基于多通道残差注意力机制的U^(2)Net视频分割模型(MRAU),旨在识别涌(突)水的演变过程。首先,基于卷积注意力模块(CBAM)改进U^(2)Net网络模型,以提高特征提取效果。通过多通道残差预处理,区分水流动态特征与静态背景,并将处理结果作为注意力机制输入模型,从而强化水流特征的学习。此外,使用中间帧掩码作为标签进行多帧融合学习,进一步提升网络对水流动态特征的识别能力。最终,通过学习不同场景下的水流特征,实现对未知场景中涌(突)水动态演变的有效识别。通过与Deeplab、LRASPP、FCN、U~2Net网络模型的对比试验,选用Dice和IoU作为评价指标。试验结果表明,MRAU模型的Dice和IoU分别达到92.88%和87.51%,相比U^(2)Net基础网络,识别结果分别提高了4.71%和7.41%。在未知的涌(突)水场景中测试时,MRAU的Dice和IoU得分分别达到了86.75%和80.23%。与其他模型相比,MRAU的识别精度最高,表明该模型在不同场景下对水流特征具有更强的泛化能力。此外,MRAU能够精准监测涌(突)水流量从小到大的演变过程。最后,通过在井下环境中模拟突水场景,进一步验证MRAU模型在实际生产中的实用性,为矿井水害监测提供了有效的技术手段。Mine water inrush video recognition is a key component in intelligent mine construction.By recognizing the dynamic evolution of water inrush from none to some and from small to large,it helps prevent the water volume from exceeding the mine’s drainage capacity and turning into a water hazard.Therefore,a video segmentation model based on the Multi-channel Residual Attention mechanism and U2Net(MRAU)was proposed to identify the evolution process of water inrush.First,the U^(2)Net network model was improved based on the Convolutional Block Attention Module(CBAM)to enhance feature extraction.Then,through multi-channel residual preprocessing,the dynamic features of water flow were distinguished from the static background,and the processed results were input into the model as an at-tention mechanism to reinforce the learning of water flow features.In addition,intermediate frame masks were used as labels for multiframe fusion learning,further enhancing the network’s ability to recognize the dynamic features of water flow.Finally,by learning the wa-ter flow features in different scenarios,the model effectively recognizes the dynamic changes of water inrush in unknown scenarios.Com-parative experiments with Deeplab,LRASPP,FCN,and U^(2)Net network models,using Dice and IoU as evaluation metrics,show that the Dice and IoU of the MRAU model reach 92.88% and 87.51%,respectively,which represents improvements of 4.71% and 7.41%over the baseline U2Net network.When tested in unknown water inrush scenarios,the MRAU model achieves Dice and IoU scores of 86.75% and 80.23%.Compared to other models,MRAU achieves the highest recognition accuracy,demonstrating stronger generalization capabilities in recognizing water flow features across different scenarios.Moreover,MRAU can accurately monitor the dynamic evolution of water in-rush from small to large.Finally,simulations of water inrush scenarios in underground environments further verify the practical utility of the MRAU model in real-world production,providing an effective technic

关 键 词：矿井涌(突)水 视频分割 MRAU 多通道残差预处理 注意力机制 U^(2)Net 

分 类 号：TD742[矿业工程—矿井通风与安全]