EEG extended source imaging with structured sparsity and $$L_1$$ -norm residual

Neural Computing and Applications - It is a long-standing challenge to reconstruct the locations and extents of cortical neural activities from electroencephalogram (EEG) recordings, especially...     

禄口机场T2航站楼Kalzip直立锁边铝合金屋面系统设计

禄口机场T2航站楼主楼屋面为9个弧形双曲屋面,由Kalzip直立锁边铝合金屋面系统、柔性卷材防水系统、承重钢底板系统等组成。作者针对该大型屋面系统的特点,就屋面抗风等关键节点的设计问题,提出了相应的解决措施,使屋面综合性能达到最优。     

某深基坑设计方案地铁设施保护技术审查分析

阐述地铁设施保护技术审查的基本流程,以及在不同审查阶段应重点关注的内容,并通过某工程实例介绍了控制保护区内深基坑设计方案的地铁设施保护技术审查和可行性论证的过程,实践表明该做法可有效保护地铁设施的安全,专家的审查意见对今后类似的深基坑支护结构设计和施工均具有一定的参考价值。     

琶洲PZB1401项目Kalzip直立锁边金属屋面系统及屋面绿化

广州琶洲PZB1401项目中庭会展厅上空的双坡屋面,采用Kalzip直立锁边铝镁锰合金屋面系统及屋面绿化,各构造层次如65/400型直立锁边铝镁锰合金屋面板等,均选用了优质、经济、环保、自重轻、使用寿命长的材料;对屋脊、山墙、檐口等节点进行了精心设计及施工,成功地解决了屋面防水防渗等问题。     

面向边缘计算应用的宽度孪生网络

边缘计算是将计算、存储、通信等任务分配到网络边缘的计算模式. 它强调在用户终端附近执行数据处理过程, 以达到降低延迟, 减少能耗, 保护用户隐私等目的. 然而网络边缘的计算、存储、能源资源有限, 这给边缘计算应用的推广带来了新的挑战. 随着边缘智能的兴起, 人们更希望将边缘计算应用与人工智能技术结合起来, 为我们的生活带来更多的便利. 许多人工智能方法, 如传统的深度学习方法, 需要消耗大量的计算、存储资源, 并且伴随着巨大的时间开销. 这不利于强调低延迟的边缘计算应用的推广. 为了解决这个问题, 我们提出将宽度学习系统(Broad learning system, BLS)等浅层网络方法应用到边缘计算应用领域, 并且设计了一种宽度孪生网络算法. 我们将宽度学习系统与孪生网络结合起来用于解决分类问题. 实验结果表明我们的方法能够在取得与传统深度学习方法相似精度的情况下降低时间和资源开销, 从而更好地提高边缘计算应用的性能.     

Hydrogen-induced degradation in SrTiO3-based grain boundary barrier layer ceramic capacitors

Hydrogen-induced degradation in SrTiO₃-based grain boundary barrier layer ceramic capacitors was studied through electrochemical hydrogen charging, in which the capacitors were placed in 0.01 M NaOH solution with hydrogen deposited on their electrodes from the electrolysis of water. The properties of the capacitors were greatly degraded after 0.5 h of treatment: The capacitance was dramatically decreased and the dielectric loss was dramatically increased over the frequency range of 10²–10⁵ Hz, the leakage current was increased by orders of magnitude. It was proposed that atomic hydrogen diffused relatively easily along the grain boundaries and induced a reduction reaction to the grain boundary layer, which resulted in the degradation observed. Hydrogen-induced degradation is more serious in SrTiO₃-based grain boundary barrier layer ceramic capacitors than in other ceramic capacitors and great efforts should be made to prevent hydrogen-induced degradation in them.     

雷电波对通信设备的危害研究

科学技术不断发展使得通信设备的应用已越来越广泛,无线通信技术也已普及,但通信系统的计算和传输部分还是离不开精密设备,因此对通信设备的保护仍需受到重视。文章研究雷电波对通信设备造成危害的原理及其对目前通信设备造成的威胁和损害,探索科学、有效的防雷技术应用,以期对通信设备的正常工作和安全运转提供保障。     

广电机房智能运维管理的研究与应用

当大数据和人工智能等技术赋能传统行业并驱动变革时,广电行业中原本完全依靠个人经验、重复枯燥的机房运维工作也逐渐实现了智能运维。因此,分析广电机房智能运维管理的现状,详细阐述智能运维管理的发展并对其进行展望,包括基础设备与机房的可视化管理、运维自动化以及智能巡检机器人等。     

中国母乳多不饱和脂肪酸组成、特点与其满足婴儿膳食营养素摄入量的比较分析

脂肪是母乳的重要组分,母乳中的脂肪酸是母乳脂肪中变化较大的常量营养素,尤其是多不饱和脂肪酸(PUFA),在婴儿营养、免疫及神经认知功能上发挥着重要作用。旨在为更科学合理的母乳喂养提供营养策略,综述了中国母乳中n-6 PUFA、n-3 PUFA、亚油酸(LA)、α-亚麻酸(ALA)、花生四烯酸(ARA)、二十二碳六烯酸(DHA)的含量以及LA/ALA比值,并根据0～6月龄婴儿PUFA的参考摄入量,分析中国母乳中PUFA的组成及特点。我国母乳成熟乳中LA含量偏高,ARA含量相对稳定,ALA和DHA含量偏低且波动范围大,LA/ALA比值在4.25～33.44之间。可通过膳食干预改善母乳中PUFA组成,提高母乳喂养婴儿生长发育所需的PUFA含量。     

Semantic material name matching system for building energy analysis

As an importance of saving energy in building sector and reducing greenhouse gas emissions is rapidly growing, many tools for building energy analysis (BEA) have been developed. However, such tools still have much room for improvement: manual input of thermal properties of a building material; input of incorrect data; and semantic conflicts between building designers and engineers. This research proposes a semantic material name matching system to find a standardized material name and its associated material property values, and enter them into a BEA tool. The core of our approach is to define ontologies capturing the concepts associated with the standard materials and their property values that are required by BEA tools. The proposed system uses an IFCXML file of a building information model as a source of input data, captures material data in the files, and automates the process of material property input. Using the proposed system, engineers will be able to increase their efficiency in entering required data into BEA tools and to reduce the possibility of erroneous data input.