In mining the left-over coal above the gob,stope wall rock of mining area have hard limestone.through field observation,the face-contacted block structure was found in rocks between coal seams to mine the left-over co...In mining the left-over coal above the gob,stope wall rock of mining area have hard limestone.through field observation,the face-contacted block structure was found in rocks between coal seams to mine the left-over coal above the gob.In order to probe into the movement law of rock strata and strata control measures,it is very important to identify the mobile block in face-contacted block structure of rocks between coal seams.This paper relies on the thought of block theory to establish appropriate parameter matrix and figure out its discrimination matrix in view of the fact that the block in face-contacted block structure has high intensity and stiffness,the展开更多
针对短时间主动热激励作用下煤岩介质表征差异不明显,不易快速、准确识别煤岩界面的难题,提出一种基于改进金字塔场景解析网络(pyramid scene parsing network,简称PSPnet)模型-MobileNetV2的煤岩界面快速精准识别方法。通过搭建煤岩主...针对短时间主动热激励作用下煤岩介质表征差异不明显,不易快速、准确识别煤岩界面的难题,提出一种基于改进金字塔场景解析网络(pyramid scene parsing network,简称PSPnet)模型-MobileNetV2的煤岩界面快速精准识别方法。通过搭建煤岩主动红外试验平台,采集并获取短时主动热激励作用下的煤岩界面红外热图像,构建了煤岩红外图像数据集;对传统PSPnet模型进行改进,采用轻量级网络模型MobileNetV2作为主干网络提取特征,大幅降低了网络模型所占内存和训练时间,同时将注意力机制模块(convolutional block attention module,简称CBAM)与金字塔场景解析(pyramid scene parsing,简称PSP)模块的上采样特征层和PSPnet网络模型的浅层特征层进行融合,有效提升模型对特征的细化能力。试验结果表明:基于改进的PSPnet-MobileNetV2网络模型所占内存仅为9.12 MB,较原始PSPnet模型减少了94.88%;煤和岩的交并比为96.52%和96.87%,分别提升了8.29%和7.7%;像素准确度分别为97.25%和99.15%,较原始网络模型分别提升了7.32%和1.64%;测试时间降低了53.70%。该方法为煤岩界面的快速和预先精准识别提供了一种有效技术手段。展开更多
文摘In mining the left-over coal above the gob,stope wall rock of mining area have hard limestone.through field observation,the face-contacted block structure was found in rocks between coal seams to mine the left-over coal above the gob.In order to probe into the movement law of rock strata and strata control measures,it is very important to identify the mobile block in face-contacted block structure of rocks between coal seams.This paper relies on the thought of block theory to establish appropriate parameter matrix and figure out its discrimination matrix in view of the fact that the block in face-contacted block structure has high intensity and stiffness,the
文摘针对短时间主动热激励作用下煤岩介质表征差异不明显,不易快速、准确识别煤岩界面的难题,提出一种基于改进金字塔场景解析网络(pyramid scene parsing network,简称PSPnet)模型-MobileNetV2的煤岩界面快速精准识别方法。通过搭建煤岩主动红外试验平台,采集并获取短时主动热激励作用下的煤岩界面红外热图像,构建了煤岩红外图像数据集;对传统PSPnet模型进行改进,采用轻量级网络模型MobileNetV2作为主干网络提取特征,大幅降低了网络模型所占内存和训练时间,同时将注意力机制模块(convolutional block attention module,简称CBAM)与金字塔场景解析(pyramid scene parsing,简称PSP)模块的上采样特征层和PSPnet网络模型的浅层特征层进行融合,有效提升模型对特征的细化能力。试验结果表明:基于改进的PSPnet-MobileNetV2网络模型所占内存仅为9.12 MB,较原始PSPnet模型减少了94.88%;煤和岩的交并比为96.52%和96.87%,分别提升了8.29%和7.7%;像素准确度分别为97.25%和99.15%,较原始网络模型分别提升了7.32%和1.64%;测试时间降低了53.70%。该方法为煤岩界面的快速和预先精准识别提供了一种有效技术手段。
