首页 | 本学科首页   官方微博 | 高级检索  
文章检索
  按 检索   检索词:      
出版年份:   被引次数:   他引次数: 提示:输入*表示无穷大
  收费全文   406篇
  免费   38篇
  国内免费   33篇
电工技术   19篇
综合类   11篇
化学工业   69篇
金属工艺   48篇
机械仪表   7篇
建筑科学   26篇
矿业工程   1篇
能源动力   25篇
轻工业   49篇
水利工程   7篇
石油天然气   36篇
武器工业   2篇
无线电   61篇
一般工业技术   47篇
冶金工业   26篇
原子能技术   8篇
自动化技术   35篇
  2024年   2篇
  2023年   6篇
  2022年   12篇
  2021年   26篇
  2020年   11篇
  2019年   11篇
  2018年   9篇
  2017年   15篇
  2016年   7篇
  2015年   20篇
  2014年   19篇
  2013年   23篇
  2012年   24篇
  2011年   37篇
  2010年   28篇
  2009年   24篇
  2008年   27篇
  2007年   35篇
  2006年   12篇
  2005年   13篇
  2004年   12篇
  2003年   6篇
  2002年   13篇
  2001年   8篇
  2000年   8篇
  1999年   12篇
  1998年   14篇
  1997年   7篇
  1996年   5篇
  1995年   5篇
  1994年   3篇
  1993年   2篇
  1992年   4篇
  1991年   2篇
  1990年   3篇
  1989年   1篇
  1986年   5篇
  1985年   1篇
  1983年   2篇
  1977年   1篇
  1966年   1篇
  1961年   1篇
排序方式: 共有477条查询结果,搜索用时 0 毫秒
471.
基于链接路径预测的聚焦Web实体搜索   总被引:1,自引:1,他引:0  
实体搜索是一个有前景的研究领域,因为它能够为用户提供更为详细的Web信息.快速、完全地收集特定领域实体所在的网页是实体搜索中的一个关键问题.为了解决这个问题,将Web网站建模为一组互连的状态构成的图,提出一种链接路径预测学习算法LPC,该模型能够学习大型网站中从主页通向目标网页的最优路径,从而指导爬虫快速定位到含有Web实体的目标网页.LPC算法分为两个阶段:首先,使用概率无向图模型CRF,学习从网站主页通往目标网页的链接路径模型,CRF模型能够融合超连接和网页中的各种特征,包括状态特征和转移特征;其次,结合增强学习技术和训练的CRF模型对爬行前端队列的超链接进行优先级评分.一种来自增强学习的折扣回报方法通过利用路径分类阶段学习的CRF模型来计算连接的回报值。在多个领域大量真实数据上的实验结果表明,所提出的适用CRF模型指导的链接路径预测爬行算法LPC的性能明显优于其他聚焦爬行算法.  相似文献   
472.
为对比分析自然循环竖排管、强制循环竖排管和强制循环横排管3种集热方式对全玻璃真空管太阳热水器有效集热量、平均有效集热效率和瞬时有效集热量等的影响,以3组30支Φ58×1800 mm的全玻璃真空管太阳热水器为研究对象,在兰州地区搭建试验平台,在太阳辐射为19.0~25.7 MJ/m2,风速为0.5~1.3m/s,平均环境...  相似文献   
473.
Wireless multimodal interactions over long distance (WMILD) would give rise to numerous thrilling applications, such as remote touching and immersive teleoperations. However, long distances can induce large propagation delays, which makes it difficult to meet the ultra-low latency requirements in haptic-visual interactions. Considering existing works mainly focused on the wireless access part, this paper designs an end-to-end framework for general WMILD applications based on the digital twin (DT) technology and proposes an intelligent resource allocation and parameter compression scheme to guarantee WMILD performance under constraint network resources. In the framework, user device can acquire real-time remote interactions by performing local interactions with nearby base station (BS), where a DT of the remote side is deployed to predict the remote haptic-visual feedbacks. A reliable DT updating process is carefully designed to guarantee the DT accurately model its dynamic physical counterpart. To optimize the updating reliability, we formulate the resource allocation and parameter compression to be a constraint-Markov decision problem, under the constraints on energy consumption, multimodal interactions and updating latencies. Then, a safe deep reinforcement learning algorithm is proposed to adapt resources and compression according to the dynamic DT updating workload, multimodal data-streams and remote transmission capacities. Simulation shows the framework can achieve high updating reliability compared with baselines.  相似文献   
474.
In hot metal forming processes, the temperature of the forming tool progressively increases under serial production conditions. Water-based two-phase lubricants may be applied to cool the forming tool and moderate temperature, in which the liquid agent would evaporate or decompose rapidly with dry matter deposited on the tooling surface during the dwelling time before the forming process commences. Herein, an interactive friction model for a two-phase lubricant is developed to predict the transient lubricant behaviors, i.e., predicting the effects of tool temperature and dwelling time on the friction coefficient evolution and lubricant breakdown. Friction tests between a warm pin and hot aluminum workpiece are conducted using the advanced friction testing system, TriboMate, to validate the modeling results.  相似文献   
475.
DNA interstrand crosslinks (ICLs) are highly toxic DNA lesions, and induce cell death by blocking DNA strand separation. Most ICL agents aiming to kill cancer cells, also generate adverse side effects to normal cells. H2O2-inducible DNA ICL agents are highly selective for targeting cancer cells, as the concentration of H2O2 is higher in cancer cells than normal cells. Previous studies have focused on arylboronate-based precursors, reacting with H2O2 to generate reactive quinone methides (QMs) crosslinking DNA. Here we explore phenyl selenide-based precursors 1 – 3 as H2O2-inducible DNA ICL agents. The precursors 1 – 3 can be activated by H2O2 to generate the good benzylic leaving group and promote production of reactive QMs to crosslink DNA. Moreover, the DNA cross-linking ability is enhanced by the introduction of substituents in the para-position of the phenolic hydroxyl group. From the substituents explored (H, OMe, F), the introduction of electron donating group (OMe) shows a pronounced elevating effect. Further mechanistic studies at the molecular and DNA levels confirm alkylation sites located mainly at dAs, dCs and dGs in DNA. Additionally, cellular experiments reveal that agents 1 – 3 exhibit higher cytotoxicity toward H1299 human lung cancer cells compared to clinically used drugs, by inducing cellular DNA damage, apoptosis and G0/G1 cell cycle arrest. This study provides a strategy to develop H2O2-inducible DNA interstrand cross-linkers.  相似文献   
476.
Proteolysis-targeting chimeras (PROTACs) provide a powerful technique to degrade targeted proteins utilizing the cellular ubiquitin-proteasome system. The major concern is the host toxicity resulting from their poor selectivity. Inducible PROTACs responding to exogenous stimulus, such as light, improve their specificity, but it is difficult for photo-activation in deep tissues. Herein, we develop H2O2-inducible PROTAC precursors 2 / 5 , which can be activated by endogenous H2O2 in cancer cells to release the active PROTACs 1 / 4 to effectively degrade targeted proteins. This results in the intended cytotoxicity towards cancer cells while targeted protein in normal cells remains almost unaffected. The higher Bromodomain-containing protein 4 (BRD4) degradation activity and cytotoxicity of 2 towards cancer cells is mainly due to the higher endogenous concentration of H2O2 in cancer cells (A549 and H1299), characterized by H2O2-responsive fluorescence probe 3 . Western blot assays and cytotoxicity experiments demonstrate that 2 degrades BRD4 more effectively and is more cytotoxic in H2O2-rich cancer cells than in H2O2-deficient normal cells. This method is also extended to estrogen receptor (ER)-PROTAC precursor 5 , showing H2O2-dependent ER degradation ability. Thus, we establish a novel strategy to induce targeted protein degradation in a H2O2-dependent way, which has the potential to improve the selectivity of PROTACs.  相似文献   
477.
研究了Hf的添加对一种镍基高硼铸造高温合金显微组织、凝固行为、疏松情况的影响。显微组织分析、差热分析(DSC)、等温凝固淬火试验结果显示,Hf在合金的凝固过程中强烈偏析于枝晶间区域,并在凝固末期以Ni5Hf相的形式从残余液相中析出;Hf的加入降低了合金的液/固相线温度、γ/γ′共晶相和硼化物析出温度,延缓了合金的凝固过程,加宽了凝固温度范围,显著增大了γ/γ′共晶相的含量。自主设计并浇铸一种薄壁管铸件,渗透检验结果显示Hf的加入使得薄壁管的疏松明显减少,合金的疏松倾向性显著降低。  相似文献   
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号