Results 51 to 60 of about 1,715 (102)

基于深度学习的雷电活动预测方法及其输电线路防雷应用

open access: yesDianci bileiqi
为提升输电线路雷击灾害主动防御能力,提出了一种基于深度学习的雷电活动预测方法,提前时间为72 h,时间和空间精度分别为3 h和5 km。基于统一的时空网格完成了预报区域雷电数据的归一化,通过卡方统一性检验提取了与雷电活动强关联的气象参量;建立了雷电发生概率预测深度神经网络模型,采用贝叶斯算法优化了模型超参数组合;建立了落雷次数与雷电流强度的分类预测卷积神经网络模型。算例验证表明,雷电发生概率预测的平均命中率和虚警率分别为69.10%和71.18%,落雷次数和雷电流强度预测命中率的平均值分别为39.03 ...
张永刚   +4 more
doaj  

[POEMS syndrome misdiagnosed as systemic sclerosis: A case report]. [PDF]

open access: yesBeijing Da Xue Xue Bao Yi Xue Ban
Liang J, Zhang X, Yao H.
europepmc   +1 more source

[Effects of cold atmosphere plasma treatment on the biological behavior of human gingival fibroblasts]. [PDF]

open access: yesBeijing Da Xue Xue Bao Yi Xue Ban
Zheng M   +7 more
europepmc   +1 more source

[Diagnostic value of intestinal tissue metagenomic next-generation sequencing in severe diarrhea following haploidentical hematopoietic stem cell transplantation]. [PDF]

open access: yesZhonghua Xue Ye Xue Za Zhi
Lin QX   +10 more
europepmc   +1 more source

Home - About - Disclaimer - Privacy