期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
煤矿旋转机械健康指标构建及状态评估 被引量:3
1
作者 李曼 潘楠楠 +1 位作者 段雍 曹现刚 《工矿自动化》 北大核心 2022年第9期33-41,共9页
煤矿设备监测参数为时间序列数据,其时序特征对健康评估的影响较大。针对传统机械设备健康评估中存在的信号时空特性提取不完备、人为经验依赖程度高、设备早期状态变化评估难等问题,建立了基于二维数组的长短期记忆降噪卷积自编码器(2D... 煤矿设备监测参数为时间序列数据,其时序特征对健康评估的影响较大。针对传统机械设备健康评估中存在的信号时空特性提取不完备、人为经验依赖程度高、设备早期状态变化评估难等问题,建立了基于二维数组的长短期记忆降噪卷积自编码器(2D-LSTMDCAE)模型,并提出了基于2D-LSTMDCAE的煤矿旋转机械健康指标(HI)构建及状态评估方法。将一维振动数据转换为二维数组,通过二维卷积网络模型充分学习原始数据中所包含的信息,增强模型对数据特征的学习能力;将样本并行输入卷积和长短期记忆(LSTM)单元,以获取完备的信号时空特征;构建无监督学习的降噪卷积自编码器(DCAE)模型并进行样本重构,采用Bray-Curtis距离计算原始样本与重构样本间相似度,得到HI,解决设备运行过程中状态标签难以获取的问题,提升模型在强背景噪声中的适应能力。使用XJTU-SY轴承数据集验证2D-LSTMDCAE模型的特征学习能力,并采用相关性和单调性2个指标评价基于HI的状态评估方法,测试结果表明:二维输入样本构建方法及学习数据时序特征的HI构建方法对轴承的性能退化更敏感,2D-LSTMDCAE模型能够更早地检测到设备的早期故障,在测试轴承上相比于LSTMDCAE和DCAE模型构建的HI及均方根平均提前了约7 min;与LSTMDCAE和DCAE模型构建的HI、均方根相比,2D-LSTMDCAE模型构建的HI的相关性和单调性均较高,能更好地反映轴承的退化情况。采用减速器加速退化实验数据进行健康评估实验,在测试减速器上,相比于均方根指标,通过2D-LSTMDCAE模型构建的HI能够提前8 min发现早期故障,且HI相关性提高了0.007,单调性提高了0.211,能够更好地反映减速器的退化情况。 展开更多
关键词 煤矿旋转机械 状态评估 健康指标 信号时空特征 长短期记忆 降噪卷积自编码器 2D-LSTMDCAE
下载PDF
Temporal and spatial profiling of nuclei-associated proteins upon TNF-α/NF-kB signaling
2
作者 Dan-jun Ma Su-Jun Li Lian-Shui Wang Jie Dai Shi-lin Zhao Rong Zeng 《Cell Research》 SCIE CAS CSCD 2009年第5期651-664,共14页
The tumor necrosis factor (TNF)-α/NF-kB-signaling pathway plays a pivotal role in various processes including apoptosis, cellular differentiation, host defense, inflammation, autoimmunity and organogenesis. The com... The tumor necrosis factor (TNF)-α/NF-kB-signaling pathway plays a pivotal role in various processes including apoptosis, cellular differentiation, host defense, inflammation, autoimmunity and organogenesis. The complexity of the TNF-α/NF-kB signaling is in part due to the dynamic protein behaviors of key players in this pathway. In this present work, a dynamic and global view of the signaling components in the nucleus at the early stages of TNF-a/ NF-KB signaling was obtained in HEK293 cells, by a combination of subcellular fractionation and stable isotope la- beling by amino acids in cell culture (SILAC). The dynamic profile patterns of 547 TNF-α-induced nuclei-associated proteins were quantified in our studies. The functional characters of all the profiles were further analyzed using that Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation. Additionally, many previously unknown effectors of TNF-α/NF-kB signaling were identified, quantified and clustered into differential activation profiles. In- terestingly, levels of Fanconi anemia group D2 protein (FANCD2), one of the Fanconi anemia family proteins, was found to be increased in the nucleus by SILAC quantitation upon TNF-α stimulation, which was further verified by western blotting and immunofluorescence analysis. This indicates that FANCD2 might be involved in TNF-α/NF-kB signaling through its accumulation in the nucleus. In summary, the combination of subcellular proteomics with quan- titative analysis not only allowed for a dissection of the nuclear TNF-α/NF-kB-signaling pathway, but also provided a systematic strategy for monitoring temporal and spatial changes in cell signaling. 展开更多
关键词 quantitative analysis SILAC PROTEOMICS TNF-α/NF-kB NUCLEUS FANCD2 subcellular fractionation
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部